Intelligent optimization shared energy storage system based on artificial intelligence algorithm

Abstract

The invention discloses an intelligent optimization shared energy storage system based on an artificial intelligence algorithm, and relates to the technical field of power system management, and the system comprises a power central control module which is used for collecting and preprocessing the historical data of a power utilization main body; the electricity consumption prediction module is used for predicting future electricity consumption demands by training a long short term memory neural network model and evaluating electricity consumption capacity; the charging and discharging management and control module is used for optimizing a charging and discharging strategy of the energy storage equipment according to the predicted power consumption data and the electricity price data and adjusting an energy storage output proportion by utilizing a genetic algorithm; the data encryption module is used for encrypting and chaining the power transaction data by using a block chain technology so as to ensure the data security; and the abnormity diagnosis module is used for carrying out real-time monitoring and early warning on system abnormity through block chain data analysis. According to the method, efficient optimal configuration and utilization of power resources are achieved, data security and privacy protection are enhanced, efficient anomaly detection and fault diagnosis capabilities are achieved, and the dynamic response and flexibility of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management, and in particular to an intelligent optimization shared energy storage system based on an artificial intelligence algorithm. Background Art

[0002] With the rapid growth of society's demand for electricity, the development and utilization of new energy sources has expanded rapidly around the world, and the construction of smart grids and distributed energy storage systems has increasingly become an important development direction for the power industry. As an emerging power management method, shared energy storage systems can integrate the energy storage resources of multiple power users to achieve more efficient power utilization. However, there are still some key technical difficulties in the management and optimization of existing shared energy storage systems.

[0003] Traditional shared energy storage systems mainly use preset rules or simple control algorithms to manage charging and discharging. These methods usually rely on fixed control strategies and are difficult to dynamically respond to complex market conditions and changing load demands. In practical applications, this static control method will lead to resource waste and low energy storage efficiency, and cannot fully achieve the optimal allocation of power resources. In addition, traditional systems lack the ability to intelligently analyze the electricity consumption behavior of different electricity users, making it difficult to predict future electricity demand based on historical data and environmental conditions.

[0004] In a shared energy storage system, the electricity consumption data of different electricity users needs to be transmitted, shared and stored. In the existing technology, many systems lack effective encryption protection mechanisms during data transmission, which can easily lead to user data leakage and security risks.

[0005] Shared energy storage systems may face a variety of abnormal situations in actual operation, such as equipment failure, abnormal fluctuations in power load, etc. The abnormal diagnosis and fault detection of existing systems usually rely on manual monitoring or simple threshold alarms, which cannot identify and locate abnormal problems in a timely and accurate manner. Summary of the invention

[0006] Based on the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide an intelligent optimization shared energy storage system based on artificial intelligence algorithm to solve the above-mentioned technical problems.

[0007] To achieve the above objectives, the present invention provides the following technical solutions: an intelligent optimization shared energy storage system based on artificial intelligence algorithm, comprising:

[0008] The power central control module is used to collect and organize the historical data of electricity users, connect to the shared energy storage system platform, and pre-process the historical electricity consumption data of electricity users, including feature conversion, feature screening and dimensionality reduction processing;

[0009] The electricity consumption prediction module is used to train the long short-term memory neural network algorithm model through the pre-processed historical electricity consumption data, and predict the future electricity consumption data of the electricity users to evaluate the electricity consumption capacity on the demand side;

[0010] The charging and discharging control module is used to finely control the charging and discharging of shared energy storage equipment according to the predicted power consumption data and electricity price data of the power users. The output strategy of the shared energy storage equipment is determined by the time-sharing load. The shortfall is balanced by the power of the grid. The actual output ratio of the energy storage unit is optimized by genetic algorithms.

[0011] The data encryption module is used to encrypt and store data using blockchain technology to ensure data security and effectiveness. When the electricity user conducts electricity transactions through the shared energy storage system, the transaction data and the current energy storage operation status are uploaded to the chain to form a block with real-time information. The power transaction within the shared energy storage system is realized through the consensus mechanism to ensure data security and effectiveness.

[0012] The abnormal diagnosis module is used to obtain blockchain data within the shared energy storage range, analyze it, ensure the data security of the power generation unit and the power consumption unit, realize abnormal monitoring of the power output of the power generation unit through the abnormal diagnosis algorithm, and provide early warning of potential risks.

[0013] The present invention is further configured such that the dimensions of the historical data of the electricity user include: electricity consumption data, time data, electricity price data, municipal power load, shared energy storage system load and ambient temperature;

[0014] The feature conversion includes: converting the time data into time features, the time features include season features, week features and hour features; converting the electricity price data, city power load, shared energy storage system load and ambient temperature into electricity price features, city power load features, shared energy storage system load features and ambient temperature features; converting the power consumption data into power consumption features; standardizing the time features, electricity price features, city power load features, shared energy storage system load features and ambient temperature features to set them as feature sets;

[0015] The feature screening includes: setting the power consumption feature as a target feature, and setting the feature set as a feature set to be screened; calculating the resonance index and linear connection degree between the target feature and the features in the feature set to be screened, calculating the comprehensive correlation index according to the resonance index and the linear connection degree, selecting the feature whose comprehensive correlation index is greater than a set comprehensive correlation index threshold, and setting it as the first screening feature set;

[0016] The dimensionality reduction process includes: performing principal component analysis on the first screening feature set to obtain a second screening feature set after dimensionality reduction.

[0017] The present invention is further configured that the calculation logic of the resonance index is: RI(F j ,Y)=∑ y∈Y P(y|F j )logP(y|F j )-∑ y∈Y P(y)logP(y), where RI(F j ,Y) is the jth feature F j and the resonance index of the target feature Y, P(y) is the probability that the target variable Y takes the value y, P(y|F j ) is the value of the given feature F j The probability that the target variable Y takes the value y under the condition of

[0018] The calculation logic of the linear connection degree is: Among them, LLD(F j ,Y) is the jth feature F j The linear connection degree with the target feature Y, n is the number of data, is the jth feature F j The weighted average value is calculated as follows: F j [i] is the value of the jth feature of the i-th data, w i is the weight of the i-th data, and the calculation logic is: median(F j ) is the jth feature F j The median of , ∈ is a constant, is the value of the jth feature of the i-th data, Y[i] is the value of the target feature Y of the i-th data, is the weighted average of the target variable Y, and the calculation logic is:

[0019] The present invention is further configured to normalize the resonance index and the linear connection degree, and calculate the comprehensive association index according to the normalized resonance index and the linear connection degree, and the calculation logic is: CAI(F j )=α·RI(F j ,Y)+β·LLD(F j ,Y), where CAI(F j ) is the jth feature F j The comprehensive correlation index, α and β are weight coefficients, both greater than or equal to 0, and α+β=1.

[0020] The present invention is further configured to train a long short-term memory neural network algorithm model through preprocessed historical electricity consumption data, and predict the future electricity consumption data of the electricity user to evaluate the electricity consumption capacity on the demand side, including:

[0021] The second screening feature set is combined with the target feature to form a data set, wherein each sample includes a feature value and a corresponding target value, and the data is organized into a time series format according to the needs of the electricity consumption forecast;

[0022] The data set is divided into a training set and a test set, the second screening feature set is used as input, the target feature is used as output, and the long short-term memory neural network algorithm model is trained;

[0023] Use mean square error as the loss function and train until the loss function converges or reaches the maximum preset training rounds;

[0024] Predictions are made on the test set and performance indicators are calculated. When the performance indicators meet the requirements, the future electricity consumption data of the electricity users is predicted. The future electricity consumption data of all electricity users are added together to obtain the electricity consumption capacity on the demand side.

[0025] The present invention is further configured such that the charge and discharge control module includes:

[0026] The data integration unit is used to integrate the predicted electricity consumption data U(t) and the electricity price data Pri(t) into a data set D = {U(t), Pri(t)} and format the data set into a time series form. To support time series analysis, N is the number of records at different time points;

[0027] The time-sharing load analysis unit is used to define the time-sharing load L(t) as the power demand at time t, and estimate it by interpolation method. δ(·) is the unit pulse function;

[0028] When the electricity price data Pri(t) is greater than the electricity price threshold, discharge is performed, and the discharge amount is: Cd(t) = S(t) - L(t); when the electricity price data Pri(t) is greater than the electricity price threshold, charging is performed, and the charging amount is: Cc(t) = L(t) - S(t), where S(t) is the current energy storage state of the energy storage device. When discharging, when L(t)>S(t), the grid power is used for balancing, and the balancing amount is: E balance (t) = L(t) - S(t).

[0029] The present invention is further configured to use a genetic algorithm to optimize the output ratio of energy storage monomers when the shared energy storage device is charged and discharged;

[0030] The charge and discharge output ratio of all energy storage units is defined as: PP = [P1, P2,, P m ], m is the number of energy storage units, P m is the output ratio of the mth energy storage unit, the sum is one, and;

[0031] Define the charging fitness function: Among them, CcP j (t) is the charging output ratio of the jth energy storage unit, Pri j (t) Charging price of the jth energy storage unit;

[0032] Define the discharge fitness function: CdP j (t) The discharge output ratio of the jth energy storage unit, Pri j (t) the discharge electricity price of the jth energy storage unit;

[0033] Randomly generate multiple charging and discharging output ratio sets, calculate the charging fitness function and the discharging fitness function, wherein when charging, the charging fitness function is minimized, and when discharging, the charging fitness function is maximized;

[0034] Select a preset number of parents from the minimum charging fitness function or the maximum discharging fitness function for crossover to generate the next generation of individuals, until the maximum number of iterations or the target fitness function, select the charging and discharging output ratio set with the largest fitness and set it as the output ratio of the energy storage monomer.

[0035] The present invention is further configured such that the data encryption module comprises:

[0036] A data encryption unit, used to encrypt the transaction data and the current energy storage operation status, wherein the encryption algorithm includes a symmetric encryption algorithm or an asymmetric encryption algorithm;

[0037] Organizing block units, organizing the encrypted data into blocks, where the block header includes a timestamp, a hash value of the previous block and a Nonce value, and the block body includes encrypted transaction data and energy storage status data;

[0038] The on-chain processing unit is used to broadcast the block to all nodes in the blockchain network, ensure that all nodes receive the new block, realize power trading in the shared energy storage system through the consensus mechanism, and ensure data security and effectiveness.

[0039] The present invention is further configured such that the abnormality diagnosis module comprises:

[0040] Obtain blockchain data within the shared energy storage range, decrypt it, and obtain a data set;

[0041] Quartile the numerical data in the data set to obtain the first quartile, the second quartile, the third quartile and the interquartile range, wherein the first quartile is the value at the 25% position in the data set, the second quartile is the median of the data set, the third quartile is the value at the 75% position in the data, and the interquartile range is the value of the third quartile minus the first quartile;

[0042] When the numerical data in the data set is greater than the sum of the third quartile and 1.5 times the interquartile range, or when the numerical data in the data set is less than the difference between the first quartile and 1.5 times the interquartile range, the corresponding data will be marked, abnormal monitoring will be generated, and early warning will be issued.

[0043] The present invention is further configured such that the shared energy storage system supports virtual power plants and spot market transactions with the electricity market, so as to provide local consumption and surplus grid-connected modes, thereby reducing the impact on the power grid system.

[0044] The present invention provides an intelligent optimization shared energy storage system based on an artificial intelligence algorithm. The power central control module is used to collect and organize the historical data of the power user, access it to the shared energy storage system platform, and pre-process the historical power consumption data of the power user, including feature conversion, feature screening and dimensionality reduction processing; the power consumption prediction module is used to train the long short-term memory neural network algorithm model through the pre-processed historical power consumption data, and predict the future power consumption data of the power user, and evaluate the power consumption capacity on the demand side; the charge and discharge control module is used to perform fine control on the charge and discharge of the shared energy storage device according to the predicted power consumption data and electricity price data of the power user, and determine the output strategy of the shared energy storage device through the time-sharing load, and the shortfall is compensated by the power grid. The data encryption module is used to encrypt and store data using blockchain technology to ensure data security and effectiveness. When the electricity user conducts electricity trading through the shared energy storage system, the transaction data and the current energy storage operation status are uploaded to the chain to form a block with real-time information. The power transaction within the shared energy storage system is realized through the consensus mechanism to ensure data security and effectiveness. The abnormal diagnosis module is used to obtain blockchain data within the shared energy storage range and analyze it to ensure the data security of the power generation unit and the power consumption unit. The abnormal diagnosis algorithm is used to realize abnormal monitoring of the power output of the power generation unit and to provide early warning of potential risks. The beneficial effects include:

[0045] 1. Improve the optimal configuration and utilization efficiency of power resources: Through the current prediction module and the power central control module, the present invention can deeply analyze the historical power consumption data of the power users, predict future power demand, and dynamically adjust the energy storage strategy in combination with real-time electricity prices and market conditions, thereby improving the utilization efficiency of power resources, reducing electricity costs, and avoiding the problem of resource waste caused by static regulation in traditional shared energy storage systems;

[0046] 2. Enhance data security and privacy protection: The data encryption module is used to encrypt sensitive information such as transaction data and electricity consumption data in the energy storage system, thereby effectively preventing the data from being stolen or tampered with during transmission and storage. In addition, the use of blockchain encryption storage mechanism ensures the immutability and traceability of data, further improving the security of the system and user privacy protection. Through the design of the data encryption module, the present invention provides multi-level security protection in the data transmission and storage links of the shared energy storage system, effectively reducing the risk of data leakage and privacy infringement, and ensuring the security and credibility of the system in a multi-user environment;

[0047] 3. Achieve efficient anomaly detection and fault diagnosis: The anomaly diagnosis module can monitor and analyze the data in the shared energy storage system in real time. Once an abnormal fluctuation or fault signal is detected, the system can automatically identify and trigger the early warning mechanism. Compared with the traditional alarm method that relies on manual monitoring or simple thresholds, the intelligent anomaly detection function of the present invention can quickly and accurately identify potential problems, help to timely maintain and optimize the system operation status, thereby improving the reliability and stability of the energy storage system;

[0048] 4. Improve the dynamic response capability and flexibility of the system: The present invention optimizes the energy storage strategy according to real-time market conditions and environmental factors through dynamic feature screening and dimensionality reduction processing technology, thereby enhancing the responsiveness of the system. In the case of large fluctuations in electricity prices or drastic changes in electricity demand, the system can quickly adjust the charging and discharging mode of the energy storage unit to achieve the optimal allocation of power resources. This efficient and flexible dynamic response mechanism enables the present invention to achieve optimal operating results in various complex electricity market environments and has strong adaptability.

[0049] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing 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. In the drawings:

[0051] Figure 1 A schematic structural diagram of an intelligent optimization shared energy storage system based on an artificial intelligence algorithm is shown as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0053] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0054] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0055] An intelligent optimization shared energy storage system based on artificial intelligence algorithm, such as Figure 1 As shown, including:

[0056] The power central control module is used to collect and organize the historical data of electricity users, connect to the shared energy storage system platform, and pre-process the historical electricity consumption data of electricity users, including feature conversion, feature screening and dimensionality reduction processing;

[0057] The electricity consumption prediction module is used to train the long short-term memory neural network algorithm model through the pre-processed historical electricity consumption data, and predict the future electricity consumption data of the electricity users to evaluate the electricity consumption capacity on the demand side;

[0058] The charging and discharging control module is used to finely control the charging and discharging of shared energy storage equipment according to the predicted power consumption data and electricity price data of the power users. The output strategy of the shared energy storage equipment is determined by the time-sharing load. The shortfall is balanced by the power of the grid. The actual output ratio of the energy storage unit is optimized by genetic algorithms.

[0059] The data encryption module is used to encrypt and store data using blockchain technology to ensure data security and effectiveness. When the electricity user conducts electricity transactions through the shared energy storage system, the transaction data and the current energy storage operation status are uploaded to the chain to form a block with real-time information. The power transaction within the shared energy storage system is realized through the consensus mechanism to ensure data security and effectiveness.

[0060] The abnormal diagnosis module is used to obtain blockchain data within the shared energy storage range, analyze it, ensure the data security of the power generation unit and the power consumption unit, realize abnormal monitoring of the power output of the power generation unit through the abnormal diagnosis algorithm, and provide early warning of potential risks.

[0061] The present invention is further configured such that the dimensions of the historical data of the electricity user include: electricity consumption data, time data, electricity price data, municipal power load, shared energy storage system load and ambient temperature;

[0062] The feature conversion includes: converting the time data into time features, the time features include season features, week features and hour features; converting the electricity price data, city power load, shared energy storage system load and ambient temperature into electricity price features, city power load features, shared energy storage system load features and ambient temperature features; converting the power consumption data into power consumption features; standardizing the time features, electricity price features, city power load features, shared energy storage system load features and ambient temperature features to set them as feature sets;

[0063] The feature screening includes: setting the power consumption feature as a target feature, and setting the feature set as a feature set to be screened; calculating the resonance index and linear connection degree between the target feature and the features in the feature set to be screened, calculating the comprehensive correlation index according to the resonance index and the linear connection degree, selecting the feature whose comprehensive correlation index is greater than a set comprehensive correlation index threshold, and setting it as the first screening feature set;

[0064] The dimensionality reduction process includes: performing principal component analysis on the first screening feature set to obtain a second screening feature set after dimensionality reduction.

[0065] The present invention is further configured that the calculation logic of the resonance index is: RI(F j ,Y)=∑ y∈Y P(y|F j )logP(y|F j )-∑ y∈Y P(y)logP(y), where RI(F j ,Y) is the jth feature F j and the resonance index of the target feature Y, P(y) is the probability that the target variable Y takes the value y, P(y|F j ) is the value of the given feature Fj The probability that the target variable Y takes the value y under the condition of F; Specifically, the above resonance index is used to measure the ability to distinguish between features, by calculating the probability that the target variable Y in each category takes the value y under the condition of F j The joint probability P(y|F j ), measure feature F j The degree of correlation with the target variable Y can effectively identify features with high correlation with the target variable, thereby helping to select the features that contribute most to the model effect and improve the performance of the model.

[0066] The calculation logic of the linear connection degree is: Among them, LLD(F j ,Y) is the jth feature F j The linear connection degree with the target feature Y, n is the number of data, is the jth feature F j The weighted average value is calculated as follows: F j [i] is the value of the jth feature of the i-th data, w i is the weight of the i-th data, and the calculation logic is: median(F j ) is the jth feature F j The median of , ∈ is a constant, is the value of the jth feature of the i-th data, Y[i] is the value of the target feature Y of the i-th data, is the weighted average of the target variable Y, and the calculation logic is: Specifically, the linear connection degree is calculated by weighted average to calculate the feature F j The linear discreteness between the feature and the target variable Y is used to evaluate the explanatory power of the feature on the target feature. First, the weighted mean and weighted standard deviation are calculated, and then the linear connection between the feature and the target is obtained through the linear coupling formula for feature selection and evaluation. The constant ∈ ranges from [0.1, 1]. By calculating the linear coupling degree between the feature and the target feature, the features with high explanatory power for the target feature can be effectively identified, thereby improving the accuracy and discrimination ability of the model. The weighted average combined with the median to calculate the weight can effectively reduce the impact of outliers on the calculation results, making the coupling degree calculation more robust.

[0067] The present invention is further configured to normalize the resonance index and the linear connection degree, and calculate the comprehensive association index according to the normalized resonance index and the linear connection degree, and the calculation logic is: CAI(F j )=α·RI(F j ,Y)+β·LLD(F j ,Y), where CAI(Fj ) is the jth feature F j The comprehensive correlation index, α and β are weight coefficients, both greater than or equal to 0, and α + β = 1. Specifically, by considering both nonlinear resonance and linear coupling, the comprehensive correlation index provides a more comprehensive feature evaluation method, making feature selection more accurate and comprehensive. The weights of the resonance index and linear coupling can be controlled by adjusting α and β, thereby adapting to the characteristic requirements of different data sets and providing greater flexibility for feature selection.

[0068] The present invention is further configured to train a long short-term memory neural network algorithm model through preprocessed historical electricity consumption data, and predict the future electricity consumption data of the electricity user to evaluate the electricity consumption capacity on the demand side, including:

[0069] The second screening feature set is combined with the target feature to form a data set, wherein each sample includes a feature value and a corresponding target value, and the data is organized into a time series format according to the needs of the electricity consumption forecast;

[0070] The data set is divided into a training set and a test set, the second screening feature set is used as input, the target feature is used as output, and the long short-term memory neural network algorithm model is trained;

[0071] Use mean square error as the loss function and train until the loss function converges or reaches the maximum preset training rounds;

[0072] Predictions are made on the test set, and performance indicators are calculated. When the performance indicators meet the requirements, the future electricity consumption data of the electricity users are predicted, and the future electricity consumption data of all electricity users are summed to obtain the electricity consumption capacity on the demand side. Specifically, the Long Short-Term Memory (LSTM) neural network is a special recursive neural network (RNN) that is good at processing time series data. LSTM can capture long-term dependencies through the "memory gate" and "forget gate" mechanisms, and is suitable for dynamic prediction of electricity demand. It is an existing technology and will not be described here. The LSTM model can effectively capture the time dependency of electricity data, thereby accurately predicting future electricity demand and helping decision makers to allocate resources reasonably.

[0073] The present invention is further configured such that the charge and discharge control module includes:

[0074] The data integration unit is used to integrate the predicted electricity consumption data U(t) and the electricity price data Pri(t) into a data set D = {U(t), Pri(t)} and format the data set into a time series form. To support time series analysis, N is the number of records at different time points;

[0075] The time-sharing load analysis unit is used to define the time-sharing load L(t) as the power demand at time t, and estimate it by interpolation method. δ(·) is a unit pulse function; Specifically, the above calculation logic defines the calculation method of the time-sharing load L(t), which is used to represent the power demand at a specific time t. The power consumption data U(t) at multiple time points are interpolated by the interpolation method. i ) and the unit pulse function δ(tt i ) are multiplied and accumulated to generate a load curve L(t) at any time t, effectively converting discrete power consumption data points into continuous load demand, which is suitable for predicting and regulating the load distribution of the system; the unit pulse function is used to generate a function of an instantaneous value at a specific time point. δ(tt i ) at t = t i , and 0 at other time points, which is used to selectively activate the power demand U(t i ), an over-time load analysis unit, the present invention can estimate the power demand at any time point, thereby improving the monitoring accuracy of load changes; the interpolation method converts discrete load data into a continuous load function, avoiding the sudden change phenomenon that may exist in traditional discrete predictions, thereby making the load prediction curve smoother and closer to the actual situation.

[0076] When the electricity price data Pri(t) is greater than the electricity price threshold, discharge is performed, and the discharge amount is: Cd(t) = S(t) - L(t); when the electricity price data Pri(t) is greater than the electricity price threshold, charging is performed, and the charging amount is: Cc(t) = L(t) - S(t), where S(t) is the current energy storage state of the energy storage device. When discharging, when L(t)>S(t), the grid power is used for balancing, and the balancing amount is: E balance (t) = L(t) - S(t).

[0077] The present invention is further configured to use a genetic algorithm to optimize the output ratio of energy storage monomers when the shared energy storage device is charged and discharged;

[0078] The charge and discharge output ratio of all energy storage units is defined as: PP = [P1, P2,, P m ], m is the number of energy storage units, P m is the output ratio of the mth energy storage unit, the sum is one, and;

[0079] Define the charging fitness function: Among them, CcP j (t) is the charging output ratio of the jth energy storage unit, Pri j (t) The charging price of the jth energy storage unit; define the discharge fitness function: CdP j(t) The discharge output ratio of the jth energy storage unit, Pri j (t) The discharge price of the jth energy storage unit; Specifically, the charging fitness function F is defined cha And define the discharge fitness function F discha They are used to evaluate the adaptability of the charging and discharging strategies of the energy storage system in a specific time period. These two functions calculate the fitness value by accumulating the product of the charging and discharging output ratio of the energy storage unit and the electricity price, thereby reflecting the charging and discharging efficiency of the energy storage system under different electricity price conditions. This method can help optimize the charging and discharging scheduling of the energy storage unit so that the system can manage energy at the optimal cost in different time periods.

[0080] Randomly generate multiple charging and discharging output ratio sets, calculate the charging fitness function and the discharging fitness function, wherein when charging, the charging fitness function is minimized, and when discharging, the charging fitness function is maximized;

[0081] Select a preset number of parents from the minimum charging fitness function or the maximum discharging fitness function for crossover to generate the next generation of individuals, until the maximum number of iterations or the target fitness function, select the charging and discharging output ratio set with the largest fitness and set it as the output ratio of the energy storage monomer.

[0082] The present invention is further configured such that the data encryption module comprises:

[0083] A data encryption unit, used to encrypt the transaction data and the current energy storage operation status, wherein the encryption algorithm includes a symmetric encryption algorithm or an asymmetric encryption algorithm;

[0084] Organizing block units, organizing the encrypted data into blocks, where the block header includes a timestamp, a hash value of the previous block and a Nonce value, and the block body includes encrypted transaction data and energy storage status data;

[0085] The chain processing unit is used to broadcast the block to all nodes in the blockchain network, ensure that all nodes receive the new block, realize the power transaction in the shared energy storage system through the consensus mechanism, and ensure the data security and effectiveness. Specifically, the data encryption module adopts blockchain technology to ensure the security of transaction data in the shared energy storage system by encryption and chaining. The data encryption module consists of the following units, each of which works together to achieve the secure storage and trusted propagation of data; data encryption unit: responsible for encrypting the transaction data and the current energy storage operation status to ensure the security of sensitive information in transmission and storage. The encryption algorithm can be symmetric encryption or asymmetric encryption, and the most appropriate encryption method is selected according to security requirements; organization block unit: packing the encrypted data into blocks. Each block includes a block header and a block body. The block header contains a timestamp, a hash value of the previous block, and a random number (Nonce value) to ensure the order and immutability of the block; the block body contains the actual encrypted transaction data and energy storage status data; chain processing unit: broadcast the generated block to all nodes in the blockchain network. Through the consensus mechanism, all nodes can verify and receive new blocks, thereby achieving data synchronization and verification across the entire network. This mechanism ensures the authenticity and security of power transaction data in the shared energy storage system.

[0086] The present invention is further configured such that the abnormality diagnosis module comprises:

[0087] Obtain blockchain data within the shared energy storage range, decrypt it, and obtain a data set;

[0088] Quartile the numerical data in the data set to obtain the first quartile, the second quartile, the third quartile and the interquartile range, wherein the first quartile is the value at the 25% position in the data set, the second quartile is the median of the data set, the third quartile is the value at the 75% position in the data, and the interquartile range is the value of the third quartile minus the first quartile;

[0089] When the numerical data in the data set is greater than the sum of the third quartile and 1.5 times the interquartile range, or when the numerical data in the data set is less than the difference between the first quartile and 1.5 times the interquartile range, the corresponding data will be marked, abnormal monitoring will be generated, and early warning will be issued. Specifically, by calculating the quartiles and intervals of the data, a deeper understanding of the central tendency and discreteness of the data can be obtained, making anomaly detection more robust and having a higher tolerance for extreme values; once the data is detected to be out of the normal range, the system can immediately mark and generate warning information, thereby discovering potential equipment failures or operational abnormalities in advance and ensuring the stability and safety of the shared energy storage system.

[0090] The present invention is further configured such that the shared energy storage system supports virtual power plants and spot market transactions with the power market, so as to provide local consumption and surplus power access mode, and reduce the impact on the power grid system. Specifically, the shared energy storage system further enhances the adaptability to the power market, supports the application of virtual power plants, and can participate in spot market transactions in the power market, thereby realizing local power consumption and surplus power access. This mechanism can optimize the economic benefits of the shared energy storage system and alleviate the load pressure on the power grid system to a certain extent. Virtual power plant support: Virtual Power Plant (VPP) is a distributed energy management method that can integrate decentralized energy storage equipment and power generation units into a "virtual" power supply entity, and uniformly dispatch and manage multiple energy storage units in the shared energy storage system. Through virtual power plant management, the shared energy storage system can respond to changes in demand in the power market in a clustered manner, balance supply and demand fluctuations, and improve the operating efficiency of the energy storage system; Spot market transactions in the power market: Spot market transactions are a real-time power trading mode. Through the trading platform, the energy storage system can perform charging and discharging operations according to the real-time electricity price. The shared energy storage system charges when electricity prices are low and discharges when electricity prices are high to maximize profits. At the same time, spot market transactions provide a dynamic pricing mechanism for energy storage systems, enabling them to participate in market competition more flexibly. Local consumption and surplus grid access: The local consumption mode enables the shared energy storage system to prioritize the consumption of locally generated renewable energy electricity (such as solar and wind power), reduce dependence on purchased electricity, and increase energy self-sufficiency. The surplus grid access mode allows excess electricity to be sold online when local electricity consumption is saturated, further realizing the profitability of the energy storage system.

[0091] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0092] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0093] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0094] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0095] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0097] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0098] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0099] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0100] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0101] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent optimization shared energy storage system based on artificial intelligence algorithm, characterized in that: include: The power central control module is used to collect and organize the historical data of the power users, connect to the shared energy storage system platform, and pre-process the historical power consumption data of the power users. Including feature conversion, feature screening and dimensionality reduction processing; The electricity consumption prediction module is used to train the long short-term memory neural network algorithm model through the pre-processed historical electricity consumption data, and predict the future electricity consumption data of the electricity users to evaluate the electricity consumption capacity on the demand side; The charging and discharging control module is used to finely control the charging and discharging of shared energy storage equipment according to the predicted power consumption data and electricity price data of the power users. The output strategy of the shared energy storage equipment is determined by the time-sharing load. The shortfall is balanced by the power of the grid. The actual output ratio of the energy storage unit is optimized by genetic algorithms. The data encryption module is used to encrypt and store data using blockchain technology to ensure data security and effectiveness. When the electricity user conducts electricity transactions through the shared energy storage system, the transaction data and the current energy storage operation status are uploaded to the chain to form a block with real-time information. The power transaction within the shared energy storage system is realized through the consensus mechanism to ensure data security and effectiveness. The abnormal diagnosis module is used to obtain blockchain data within the shared energy storage range, analyze it, ensure the data security of the power generation unit and the power consumption unit, realize abnormal monitoring of the power output of the power generation unit through the abnormal diagnosis algorithm, and provide early warning of potential risks.

2. According to claim 1, an intelligent optimization shared energy storage system based on artificial intelligence algorithm is characterized in that: The dimensions of the historical data of electricity users include: electricity consumption data, time data, electricity price data, city power load, shared energy storage system load and ambient temperature; The feature conversion includes: converting the time data into time features, wherein the time features include season features, week features and hour features; converting the electricity price data, city power load, shared energy storage system load and ambient temperature into electricity price features, city power load features, shared energy storage system load features and ambient temperature characteristics; converting the power consumption data into power consumption characteristics; standardizing the time characteristics, electricity price characteristics, city power load characteristics, shared energy storage system load characteristics and ambient temperature characteristics to set them as a feature set; The feature screening includes: setting the power consumption feature as a target feature, and setting the feature set as a feature set to be screened; calculating the resonance index and linear connection degree between the target feature and the features in the feature set to be screened, calculating the comprehensive correlation index according to the resonance index and the linear connection degree, selecting the feature whose comprehensive correlation index is greater than a set comprehensive correlation index threshold, and setting it as the first screening feature set; The dimensionality reduction process includes: performing principal component analysis on the first screening feature set to obtain a second screening feature set after dimensionality reduction.

3. According to claim 2, an intelligent optimization shared energy storage system based on artificial intelligence algorithm is characterized in that: The calculation logic of the resonance index is: RI(F j ,Y)=∑ y∈Y P(y|F j )logP(y|F j )-∑ y∈Y P(y)logP(y), Among them, RI(F j ,Y) is the jth feature F j and the resonance index of the target feature Y, P(y) is the probability that the target variable Y takes the value y, P(y|F j ) is the given feature F j The probability that the target variable Y takes the value y under the condition of The calculation logic of the linear connection degree is: Among them, LLD(F j ,Y) is the jth feature F j The linear connection degree with the target feature Y, n is the number of data, is the jth feature F j The weighted average value is calculated as follows: F j [i] is the value of the jth feature of the i-th data, w i is the weight of the i-th data, and the calculation logic is: is the jth feature F j The median of , ∈ is a constant, is the value of the jth feature of the i-th data, Y[i] is the value of the target feature Y of the i-th data, is the weighted average of the target variable Y, and the calculation logic is:

4. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 3 is characterized in that: The resonance index and linear connection degree are normalized, and the comprehensive association index is calculated based on the normalized resonance index and linear connection degree. The calculation logic is: CAI(F j )=α·RI(F j ,Y)+β·LLD(F j ,Y), where CAI(F j ) is the jth feature F j The comprehensive correlation index, α and β are weight coefficients, both greater than or equal to 0, and α+β=1.

5. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 1 is characterized in that: The long short-term memory neural network algorithm model is trained through the pre-processed historical electricity consumption data, and the future electricity consumption data of the electricity users is predicted to evaluate the electricity consumption capacity on the demand side, including: The second screening feature set is combined with the target feature to form a data set, wherein each sample includes a feature value and a corresponding target value, and the data is organized into a time series format according to the needs of the electricity consumption forecast; The data set is divided into a training set and a test set, the second screening feature set is used as input, the target feature is used as output, and the long short-term memory neural network algorithm model is trained; Use mean square error as the loss function and train until the loss function converges or reaches the maximum preset training rounds; Predictions are made on the test set and performance indicators are calculated. When the performance indicators meet the requirements, the future electricity consumption data of the electricity users is predicted. The future electricity consumption data of all electricity users are added together to obtain the electricity consumption capacity on the demand side.

6. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 1 is characterized in that: The charge and discharge control module includes: The data integration unit is used to integrate the predicted electricity consumption data U(t) and the electricity price data Pri(t) into a data set D = {U(t), Pri(t)} and format the data set into a time series form. To support time series analysis, N is the number of records at different time points; The time-sharing load analysis unit is used to define the time-sharing load L(t) as the power demand at time t, and estimate it by interpolation method. δ(·) is the unit pulse function; When the electricity price data Pri(t) is greater than the electricity price threshold, discharge is performed, and the discharge amount is: Cd(t) = S(t) - L(t); when the electricity price data Pri(t) is greater than the electricity price threshold, charging is performed, and the charging amount is: Cc(t) = L(t) - S(t), where S(t) is the current energy storage state of the energy storage device. When discharging, when L(t)>S(t), the grid power is used for balancing, and the balancing amount is: E balance (t) = L(t) - S(t).

7. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 6 is characterized in that: When the shared energy storage device is charged and discharged, a genetic algorithm is used to optimize the output ratio of the energy storage unit; The charge and discharge output ratio of all energy storage units is defined as: PP = [P1, P2,, P m ], m is the number of energy storage units, P m is the output ratio of the mth energy storage unit, the sum is one, and; Define the charging fitness function: Among them, CcP j (t) is the charging output ratio of the jth energy storage unit, Pri j (t) Charging price of the jth energy storage unit; Define the discharge fitness function: CdP j (t) The discharge output ratio of the jth energy storage unit, Pri j (t) the discharge price of the jth energy storage unit; Randomly generate multiple charging and discharging output ratio sets, calculate the charging fitness function and the discharging fitness function, wherein when charging, the charging fitness function is minimized, and when discharging, the charging fitness function is maximized; Select a preset number of parents from the minimum charging fitness function or the maximum discharging fitness function for crossover to generate the next generation of individuals, until the maximum number of iterations or the target fitness function, select the charging and discharging output ratio set with the largest fitness and set it as the output ratio of the energy storage monomer.

8. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 1 is characterized in that: The data encryption module comprises: A data encryption unit, used to encrypt the transaction data and the current energy storage operation status, wherein the encryption algorithm includes a symmetric encryption algorithm or an asymmetric encryption algorithm; Organizing block units, organizing the encrypted data into blocks, where the block header includes a timestamp, a hash value of the previous block and a Nonce value, and the block body includes encrypted transaction data and energy storage status data; The on-chain processing unit is used to broadcast the block to all nodes in the blockchain network, ensure that all nodes receive the new block, realize power trading in the shared energy storage system through the consensus mechanism, and ensure data security and effectiveness.

9. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 1 is characterized in that: The abnormality diagnosis module comprises: Obtain blockchain data within the shared energy storage range, decrypt it, and obtain a data set; Quartile the numerical data in the data set to obtain the first quartile, the second quartile, the third quartile and the interquartile range, wherein the first quartile is the value at the 25% position in the data set, the second quartile is the median of the data set, the third quartile is the value at the 75% position in the data, and the interquartile range is the value of the third quartile minus the first quartile; When the numerical data in the data set is greater than the sum of the third quartile and 1.5 times the interquartile range, or when the numerical data in the data set is less than the difference between the first quartile and 1.5 times the interquartile range, the corresponding data will be marked, abnormal monitoring will be generated, and early warning will be issued.

10. The intelligent optimization shared energy storage system based on artificial intelligence algorithm according to claim 1, characterized in that: The shared energy storage system supports virtual power plants and spot market transactions with the electricity market to provide local consumption and surplus grid-connected modes, thereby reducing the impact on the power grid system.

Images