Data encryption method and system for balance area of power distribution network

Abstract

The invention discloses a data encryption method and system for a balance area of a power distribution network, which is applied to the field of data encryption, and comprises the following steps: carrying out analysis processing on obtained to-be-encrypted data of a target balance area of the power distribution network to obtain load information and behavior information of the to-be-encrypted data; determining a first encryption level of the load information; determining a second encryption level of the behavior information; determining an encryption algorithm of the to-be-encrypted data based on the first encryption level and the second encryption level; carrying out quantitative coding processing on the data source information, the power grid equipment attribute parameters and the operation fluctuation condition, and generating an encryption key matched with an encryption algorithm; and encrypting the to-be-encrypted data of the target power distribution network balance area at least by using the encryption algorithm and the encryption key. According to the data encryption method and system for the balance area of the power distribution network provided by the invention, the problem of insufficient security or resource waste caused by lack of risk distinguishing capability of a traditional encryption mechanism is effectively solved, and the security of data encryption of the balance area of the power distribution network is improved.

Description

Technical Field

[0001] This invention relates to the field of power distribution network balance zone technology, and in particular to a data encryption method and system for power distribution network balance zone. Background Technology

[0002] The distribution network balance zone aggregates a large number of dispersed distributed power sources and energy storage systems to achieve collaborative management and optimized control. During this process, massive amounts of data are exchanged at high frequency; precise encryption protection of communication data ensures the secure operation of data exchange within the distribution network balance zone.

[0003] Current technologies do not differentiate between different grid data; instead, all grid data is uniformly encrypted. Existing encryption methods extract equipment information from the grid data and encrypt the data according to pre-defined encryption rules for that equipment information. However, data in balanced distribution grid areas is highly heterogeneous and scenario-sensitive. The security risks of data from the same device under different operating conditions or from different sources can vary drastically. Low-strength encryption may be used in high-risk scenarios, leading to command tampering or sensitive data leakage; conversely, over-encryption may occur in low-risk scenarios, wasting terminal computing and communication resources. This encryption mechanism struggles to meet the dual requirements of security and efficiency in balanced distribution grid areas. Summary of the Invention

[0004] This invention provides a data encryption method and system for the balanced distribution network area, which solves the technical problem of existing technologies that use predefined rules to analyze and encrypt data from a single dimension, thereby improving the security and efficiency of data encryption in the balanced distribution network area.

[0005] To address the aforementioned technical problems, this invention provides a data encryption method and system for the balanced distribution network area, the method comprising: The data to be encrypted in the target distribution network balance zone is parsed and processed to obtain the load information and behavior information of the data to be encrypted. Based on the data source information corresponding to the behavioral information, the security risk result of the data to be encrypted is determined, and based on the security risk result and the operational fluctuation determined by the load information, the first encryption level of the load information is determined; Based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information, a second encryption level of the behavioral information is determined; Based on the first encryption level and the second encryption level, an encryption algorithm for the data to be encrypted is determined; the data source information, the power grid equipment attribute parameters, and the operating fluctuation are quantized and encoded to generate an encryption key that matches the encryption algorithm; The data to be encrypted in the target distribution network balance zone is encrypted using at least the encryption algorithm and the encryption key.

[0006] Preferably, determining the first encryption level of the load information based on the security risk results and the operational fluctuations determined by the load information includes: Extract multi-dimensional time-series data from the load information; The multi-dimensional time series data is used to construct a three-dimensional tensor and input into a pre-constructed analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics. The short-term sudden fluctuation characteristics and the long-term regularity characteristics are analyzed to determine the operational fluctuation situation; The security risk results and the operational fluctuations are used to construct a two-dimensional judgment matrix to obtain the first encryption level.

[0007] Preferably, determining the second encryption level of the behavior information based on the topology of the target distribution network and the power grid equipment attribute parameters indicated by the behavior information includes: Extract the node connection structure from the topology of the target distribution network to obtain the node hierarchy type; The behavioral information is parsed to obtain the attribute parameters of the power grid equipment; One-hot encoding is performed on the node level type, and digital encoding is performed on the attribute parameters of the power grid equipment to obtain a two-dimensional feature vector; The sensitivity fusion features are obtained by analyzing the two-dimensional feature vectors. Based on the aforementioned sensitivity fusion features, a second encryption level for behavioral information is obtained.

[0008] Preferably, the step of quantizing and encoding the data source information, the power grid equipment attribute parameters, and the operational fluctuations to generate an encryption key matching the encryption algorithm includes: The data source identifier is extracted from the data source information; The attribute parameters of the power grid equipment are decomposed to obtain the equipment identifier; The operational fluctuations are quantified and encoded to obtain operational fluctuation data; Based on the data source identifier, the device identifier, and the operational fluctuation data, a quantization encoding vector is obtained; The encryption key is generated based on the quantization encoding vector and the key length requirement of the encryption algorithm.

[0009] Preferably, the step of encrypting the data to be encrypted in the target distribution network balance zone using at least the encryption algorithm and the encryption key further includes: Collect fault-related data of the target distribution network balance zone to obtain fault alarm signals and emergency dispatch instructions; The encryption level decision is determined based on the fault level and emergency priority. Based on the encryption level decision result, the encryption algorithm, and the encryption key, a second encryption algorithm and a second encryption key are obtained; The data to be encrypted is compressed to obtain compressed data to be encrypted. The compressed data to be encrypted is encrypted based on the second encryption algorithm and the second encryption key.

[0010] Another aspect of the present invention provides a data encryption system for the balanced zone of a power distribution network, comprising: The parsing module is used to parse and process the data to be encrypted in the target distribution network balance area to obtain the load information and behavior information of the data to be encrypted. The first module is used to determine the security risk result of the data to be encrypted based on the data source information corresponding to the behavioral information, and to determine the first encryption level of the load information based on the security risk result and the operating fluctuation determined by the load information. The second module is used to determine the second encryption level of the behavior information based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavior information. The key module is used to determine the encryption algorithm of the data to be encrypted based on the first encryption level and the second encryption level; and to quantize and encode the data source information, the power grid equipment attribute parameters and the operating fluctuations to generate an encryption key that matches the encryption algorithm. An encryption module is used to encrypt the data to be encrypted in the target distribution network balance zone using at least the encryption algorithm and the encryption key.

[0011] Preferably, determining the first encryption level of the load information based on the security risk results and the operational fluctuations determined by the load information includes: The extraction unit is used to extract multi-dimensional time-series data from the load information; The network unit is used to construct a three-dimensional tensor from the multi-dimensional time series data and input it into a pre-constructed analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics. The analysis unit is used to analyze the short-term sudden fluctuation characteristics and the long-term regularity characteristics to determine the operational fluctuation situation; The first unit is used to construct a two-dimensional judgment matrix by combining the security risk results with the operational fluctuations to obtain the first encryption level.

[0012] Preferably, the second module includes: Node unit, used to extract node connection structure from the topology of the target distribution network to obtain node hierarchy type; The device unit is used to parse the behavioral information to obtain the attribute parameters of the power grid equipment; The node level type is encoded using one-hot encoding, and the attribute parameters of the power grid equipment are digitally encoded to obtain a two-dimensional feature vector. A sensitivity unit is used to analyze the two-dimensional feature vector to obtain sensitivity fusion features; The second encryption unit is used to obtain a second encryption level for behavioral information based on the sensitivity fusion features.

[0013] Preferably, the key module includes: An identification unit is used to extract a data source identifier from the data source information; The disassembly unit is used to disassemble the attribute parameters of the power grid equipment to obtain the equipment identifier; A quantization unit is used to quantize and encode the operational fluctuations to obtain operational fluctuation data. The encoding vector unit is used to obtain a quantized encoding vector based on the data source identifier, the device identifier, and the operational fluctuation data; An encryption key unit is used to generate the encryption key based on the quantization encoding vector and the key length requirement of the encryption algorithm.

[0014] Preferably, the encryption module further includes: The data acquisition unit is used to collect fault-related data of the target distribution network balance zone and obtain fault alarm signals and emergency dispatch instructions. The decision-making unit is used to determine the encryption level decision based on the fault level and emergency priority. The second encryption unit is used to obtain a second encryption algorithm and a second encryption key based on the encryption level decision result, the encryption algorithm, and the encryption key; A compression unit is used to compress the data to be encrypted to obtain compressed data to be encrypted; A compression encryption unit is used to encrypt the compressed data to be encrypted based on the second encryption algorithm and the second encryption key.

[0015] Compared with the prior art, the beneficial effects of the present invention are at least one of the following: This invention parses the data to be encrypted to extract load and behavioral information. Based on the data source information corresponding to the behavioral information, it assesses the security risk level of the data to be encrypted and, combined with the operational fluctuations determined by the load information, comprehensively determines the first encryption level corresponding to the load information. Simultaneously, based on the topology of the target distribution network and the power grid equipment attribute parameters indicated by the behavioral information, it determines the second encryption level corresponding to the behavioral information. Furthermore, this invention integrates the first and second encryption levels, dynamically selecting an encryption algorithm that matches the overall security requirements. It further normalizes and quantizes the data source information, the power grid equipment attribute parameters, and the operational fluctuations to generate an encryption key compatible with the selected encryption algorithm. Finally, this invention encrypts the data to be encrypted in the target distribution network's balanced area using at least the encryption algorithm and the encryption key. This eliminates the need for encryption based on preset rules in existing technologies. This invention effectively solves the security deficiencies or resource waste caused by the lack of risk differentiation capabilities in traditional encryption mechanisms, significantly improving the security and efficiency of data interaction in the balanced area of ​​the distribution network. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a data encryption method for a power distribution network balance zone in one embodiment of the present invention. Figure 2 This is a schematic diagram of the data encryption system in the power distribution network balance zone according to one embodiment of the present invention; Figure label: Among them, 11 is the parsing module; 12 is the first module; 13 is the second module; 14 is the key module; and 15 is the encryption module. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joint" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. The term "and / or" as used herein includes any and all combinations of one or more of the related listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] In the distribution network balance zone, collaborative optimization is achieved by aggregating resources such as distributed power sources and energy storage. This involves massive amounts of data interacting frequently, requiring precise encryption to ensure security. However, existing technologies use a uniform encryption strategy for all data, encrypting only equipment information according to preset rules. This ignores the differences in security risks of data under different operating states and sources, resulting in insufficient protection in high-risk scenarios and waste of resources in low-risk scenarios, making it difficult to balance security and efficiency.

[0022] One embodiment of the present invention provides a data encryption method for the balanced zone of a distribution network. For details, please refer to [link to relevant documentation]. Figure 1 , Figure 1 The diagram shown is a flowchart illustrating a data encryption method for a distribution network balance zone according to one embodiment of the present invention, including: S1. Parse and process the data to be encrypted in the target distribution network balance zone to obtain the load information and behavior information of the data to be encrypted. S2. Based on the data source information corresponding to the behavioral information, determine the security risk result of the data to be encrypted, and based on the security risk result and the operational fluctuation determined by the load information, determine the first encryption level of the load information; S3. Determine the second encryption level of the behavioral information based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information; S4. Based on the first encryption level and the second encryption level, determine the encryption algorithm for the data to be encrypted; quantize and encode the data source information, power grid equipment attribute parameters and operating fluctuations to generate an encryption key that matches the encryption algorithm; S5. Encrypt the data to be encrypted in the target distribution network balance area using at least an encryption algorithm and an encryption key.

[0023] Data in the distribution network balance zone is highly heterogeneous and highly sensitive to different scenarios. The security risk level of the same equipment varies significantly under different operating states, such as normal operation, fault response, or dispatch control, or under different data sources, such as telemetry, remote control, or market transactions. If only the equipment information is based on preset fixed encryption rules, it is impossible to dynamically reflect the actual operating context and potential threat level of the data. This can easily lead to high-risk critical instructions being tampered with or leaked due to insufficient encryption strength. At the same time, excessive encryption of low-sensitivity routine data results in a waste of terminal computing and communication resources. Therefore, it is difficult to meet the comprehensive requirements of the distribution network balance zone in terms of security, real-time performance, and resource efficiency.

[0024] Preferably, the data to be encrypted in the target distribution network balance zone is parsed and processed to obtain the load information and behavior information of the data to be encrypted. The acquired data to be encrypted from the target distribution network balance zone is parsed to obtain load information and behavioral information. The target distribution network balance zone refers to a specific area in the distribution network where the load distribution is relatively balanced and independent power dispatch and management are possible. The data to be encrypted refers to data collected during distribution network operation that contains raw user electricity consumption information and requires encryption protection. This type of data is usually stored in binary or a specific format and includes voltage and current acquisition values ​​and user electricity consumption time records. Load information refers to data reflecting the power consumption within the distribution network balance zone, including total regional electricity consumption, average load for each time period, and peak load. Behavioral information refers to data reflecting the electricity consumption patterns of users or equipment, including electricity consumption time distribution, changes in electricity consumption patterns, and abnormal electricity consumption records. In practice, a data format parsing tool is first used to read the data to be encrypted. A protocol parsing algorithm is used to extract the address field, function code, and data field from the data frame, separating the raw fields such as voltage, current, and electricity consumption time. Then, the parsed raw fields are cleaned, and anomaly detection algorithms such as the 3σ criterion are used to calculate voltage and current values. Based on the mean and standard deviation, outliers exceeding ±3 times the standard deviation of the mean are removed. Missing electricity usage records are then filled using algorithms such as linear interpolation to obtain a regularized raw dataset. Load information is then extracted from this regularized raw dataset. Time-series statistical algorithms are used to calculate the average current and voltage values ​​per hour for each day, and the active power for each time period is calculated. This yields the average load and peak load for the total regional electricity consumption for each time period. Next, behavioral information is extracted. Clustering algorithms are used to cluster user electricity consumption data by time series, dividing it into peak, valley, and flat electricity consumption periods. Association rule mining algorithms are used to analyze changes in electricity consumption on different dates, identifying stable electricity consumption patterns and abnormal electricity consumption records, such as a user consistently using low load at night but suddenly exhibiting high load during the day. This process aims to extract key information supporting distribution network scheduling and management from the data to be encrypted, avoiding the raw data being disorganized and unusable. The benefits include providing accurate data support for subsequent distribution network load forecasting and safety management, and providing clear data classification criteria for targeted encryption of the data to be encrypted.

[0025] Preferably, based on the data source information corresponding to the behavioral information, the security risk result of the data to be encrypted is determined. Based on the security risk result and the operational fluctuation determined by the load information, the first encryption level of the load information is determined. Multi-dimensional time-series data is extracted from the load information; a three-dimensional tensor is constructed from the multi-dimensional time-series data and input into a pre-constructed analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics; the short-term sudden fluctuation characteristics and long-term regularity characteristics are analyzed to determine the operational fluctuation; a two-dimensional judgment matrix is ​​constructed from the security risk result and the operational fluctuation to obtain the first encryption level. The security risk result of the data to be encrypted is determined based on the data source information corresponding to the behavioral information. Then, the first encryption level of the load information is determined by combining the operational fluctuations determined by the load information. The data source information refers to the relevant information of the device or user that generates the data to be encrypted, including the data acquisition terminal number, user electricity category, and data transmission link type, such as the dedicated acquisition terminal number of industrial users and the ordinary acquisition terminal number of residential users. The security risk result refers to the assessment of the risk of leakage or tampering of the data to be encrypted, which is usually divided into three levels: high risk, medium risk, and low risk. Multi-dimensional time series data refers to data extracted from the load information that contains multiple electricity parameters and is arranged in chronological order, such as a time series composed of active power, reactive power, and current parameters collected every 15 minutes. The three-dimensional tensor is a matrix structure constructed by multi-dimensional time series data according to the time dimension, parameter dimension, and sample dimension. For example, a tensor with a time dimension of 24 hours, a parameter dimension of 3 samples, and a parameter dimension of 30 days; short-term sudden fluctuation characteristics refer to the drastic changes in load data within a short period of time, such as a sudden increase of 50% in power within one hour; long-term regularity characteristics refer to the stable change patterns of load data over a longer period, such as higher power at 8 am and 10 pm on weekdays; volatility factor is a numerical value that quantifies the degree of fluctuation in load data, usually ranging from 0 to 1, with larger values ​​indicating more severe fluctuations; operational volatility refers to the degree of load operation stability based on the volatility factor, divided into three categories: severe fluctuation, general fluctuation, and stable; and a two-dimensional judgment matrix is ​​a matrix constructed with safety risk results as the row dimension and operational volatility as the column dimension. Each cell in the matrix corresponds to a first encryption level, which is the encryption strength level of the load information determined by the two-dimensional judgment matrix.In practical implementation, the security risk results are first assessed based on the data source information, and a risk assessment index system is established. This system includes the security authentication level of the data acquisition terminal, the sensitivity of the user's electricity category, and the encryption status of the transmission link. Each index is assigned a weight, such as 0.4 for authentication level, 0.3 for user category, and 0.3 for transmission link. The fuzzy comprehensive evaluation method is used to calculate the risk score, with a score greater than 80 indicating high risk, 60 to 80 indicating medium risk, and below 60 indicating low risk. Next, multi-dimensional time-series data is extracted from the load information. The time interval is set to 15 minutes, and active power, reactive power, and current data from the past 30 days are extracted to form a dataset of 24×4 (96 time points per day)×3 (parameters)×30 (days). A three-dimensional tensor is then constructed. Finally, the three-dimensional tensor is input into a neural network model based on a combination of convolutional neural networks and long short-term memory networks. The convolutional layers extract short-term sudden fluctuation features. For example, a 3×3 convolutional kernel is used to capture power spikes within one hour, and a long short-term memory network layer extracts long-term regularity features, such as learning the power change pattern over 30 days through 10 hidden layer units. These two features are then input into a fully connected layer, outputting a volatility factor of 0 to 1. The volatility factor is then used to determine the operational volatility: a factor greater than 0.7 indicates severe volatility, 0.3 to 0.7 indicates moderate volatility, and below 0.3 indicates stability. Finally, a two-dimensional judgment matrix is ​​constructed: high risk and severe volatility correspond to Level 1 encryption, high risk and moderate volatility correspond to Level 2 encryption, and other situations correspond to Level 3 encryption. Based on the combination of security risk results and operational volatility, the first encryption level is matched from the matrix. This is done to ensure that the encryption level accurately matches the data risk and operational needs, avoiding over-encryption or under-encryption. The advantage is that it can ensure data security while also considering the efficiency of power distribution network operation data utilization.

[0026] Operational fluctuations reflect the stability of distribution network load data. The more drastic the fluctuations, the more critical the information about changes in the power grid's operating status contained in the data. For example, short-term sudden fluctuations may be associated with sensitive scenarios such as equipment failures or peak electricity consumption. Once such data is leaked or tampered with, it may be maliciously used to interfere with power grid dispatch or cause security risks. Therefore, it requires stronger encryption protection. On the other hand, data with gentle fluctuations, such as long-term stable residential electricity load, has highly predictable changes in its status and lower data sensitivity and leakage risk. Basic encryption strength is sufficient to meet security requirements.

[0027] Preferably, the second encryption level of the behavioral information is determined based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information. The node connection structure is extracted from the topology of the target distribution network to obtain the node hierarchy type; the behavioral information is parsed to obtain the attribute parameters of the power grid equipment; one-hot encoding is performed on the node hierarchy type, and digital encoding is performed on the attribute parameters of the power grid equipment to obtain a two-dimensional feature vector; the two-dimensional feature vector is analyzed to obtain a sensitivity fusion feature; and the second encryption level of the behavioral information is obtained based on the sensitivity fusion feature. The second encryption level of behavioral information is determined based on the topology and behavioral information of the target distribution network, indicating the attribute parameters of the power grid equipment. The topology of the target distribution network refers to the connection structure and hierarchical relationship between various devices such as transformers, switchgear, and lines. For example, a substation transformer connects to a 10kV line, which in turn connects to a user distribution box. The node connection structure refers to the connection method between various device nodes in the topology, such as transformer nodes, line nodes, and user nodes. The node hierarchy type is a category based on the importance of the node in the topology, such as core nodes (substation transformers), important nodes (10kV line switches), and ordinary nodes (user distribution boxes). The power grid equipment attribute parameters are data describing the characteristics of the power grid equipment itself, including equipment model, rated power, operating years, and whether it is a critical protection device. One-hot encoding is an encoding method that converts discrete node hierarchical types into binary vectors. For example, core nodes are encoded as [1,0,0], important nodes as [0,1,0], and ordinary nodes as [0,0,1]. Digital encoding is a processing method that converts power grid equipment attribute parameters into numerical values. For example, equipment models are assigned values ​​from 1 to 5 according to their importance, and rated power is assigned values ​​from 1 to 3 according to the range. The two-dimensional feature vector is a comprehensive feature vector formed by concatenating the node hierarchical vector after one-hot encoding and the equipment attribute vector after digital encoding. Sensitivity fusion feature is a numerical value that reflects the sensitivity of behavioral information, obtained by weighting the two-dimensional feature vector. The second encryption level is the encryption strength level of behavioral information determined according to the sensitivity fusion feature. Similar to the first encryption level, it is divided into level one, level two, and level three, with level one having the highest encryption strength.In practical implementation, the node connection structure is first extracted from the topology data of the target distribution network. A graph theory analysis tool, such as the NetworkX library, is used to traverse the topology graph, identifying the connection objects and hierarchical levels of each node. Nodes are then classified into three categories: core nodes, important nodes, and ordinary nodes, thus obtaining the node hierarchy type. Next, the behavioral information is parsed to extract the corresponding power grid equipment attribute parameters. For example, parameters for equipment with model SCB13-1000kVA, rated power of 1000kVA, 5 years of operation, and being a critical protection device are obtained from the behavioral information. Then, the node hierarchy type is one-hot encoded using the OneHotEncoder tool from Python's sklearn library. Core nodes are converted to [1,0,0], and the power grid equipment attribute parameters are numerically encoded: equipment model SCB13-1000kVA is assigned a value of 5, rated power of 1000kVA is assigned a value of 3, 5 years of operation is assigned a value of 2, and critical protection device is assigned a value of 4. Finally, the two encoded vectors are concatenated, such as the core node encoding... The code [1,0,0] and the device attribute code [5,3,2,4] are concatenated to form [1,0,0,5,3,2,4], resulting in a two-dimensional feature vector. This two-dimensional feature vector is then analyzed, with the node-level encoding weight set to 0.4 and the device attribute encoding weight set to 0.6. The sensitivity fusion feature is calculated, such as (1×0.4+0×0.4+0×0.4)+(5×0.6+3×0.6+2×0.6+4×0.6)=0.4+(3+1.8+1.2+2.4)=0. 0.4 + 8.4 = 8.8; Finally, the second encryption level is determined based on the sensitivity fusion feature. The feature value is set to be greater than 8 for level 1 encryption, 5 to 8 for level 2 encryption, and less than 5 for level 3 encryption. The second encryption level corresponding to 8.8 is level 1. This is done to match the encryption strength based on the location and characteristics of the equipment in the power grid, avoiding uniform encryption that is out of touch with the actual importance of the equipment. The advantage is that it can make the encryption protection of behavioral information more targeted, and reduce the encryption resource consumption of non-critical information while ensuring the information security of critical equipment.

[0028] Preferably, based on the first and second encryption levels, an encryption algorithm for the data to be encrypted is determined; the data source information, power grid equipment attribute parameters, and operational fluctuations are quantized and encoded to generate an encryption key matching the encryption algorithm. The data source identifier is extracted from the data source information; the power grid equipment attribute parameters are decomposed to obtain the equipment identifier; the operational fluctuations are quantized and encoded to obtain operational fluctuation data; a quantized encoding vector is obtained based on the data source identifier, equipment identifier, and operational fluctuation data; and an encryption key is generated based on the quantized encoding vector and the key length requirement of the encryption algorithm. The encryption algorithm for the data to be encrypted is determined based on the first and second encryption levels. Then, an encryption key matching the algorithm is generated by quantizing and encoding data source information, power grid equipment attribute parameters, and operational fluctuations. The encryption algorithm refers to the mathematical operation rules used to encrypt the data. Common algorithms include DES, AES-128 (Data Encryption Standard), and AES-256 (Advanced Encryption Standard with 256-bit key). Different algorithms have different encryption strengths and key lengths. DES has a 56-bit key length, while AES-128 (Advanced Encryption Standard with 256-bit key) has a 128-bit key length. The key consists of two parts: Advanced Encryption Standard (AES-256) key (128-bit key, 256-bit key), Data Source Identifier (data source identifier) ​​(unique identifier extracted from data source information, such as device number of data acquisition terminal or user electricity account number), Device Identifier (data source identifier obtained by disassembling power grid equipment attribute parameters, such as asset number of transformer or model code of switchgear), Operational Fluctuation Data (data source identifier quantized and encoded, such as 3 for severe fluctuations, 2 for normal fluctuations, and 1 for stable fluctuations), Quantization Encoding Vector (data source identifier, device identifier, and operational fluctuation data converted into numerical values ​​and then concatenated into a vector), and Encryption Key (a string or numerical sequence that meets the key length requirements of the encryption algorithm and is used to encrypt and decrypt data).In practical implementation, the encryption algorithm is first determined based on the combination of the first and second encryption levels, establishing a correspondence between levels and algorithms. If both the first and second encryption levels are level one, the AES-256 algorithm is selected; if one is level one and the other is level two, the AES-128 algorithm is selected; if both are level two or lower, the DES algorithm is selected. Next, the data source identifier is extracted from the data source information, such as the acquisition terminal number "DTU-2025-001". It is converted into a numerical sequence using ASCII encoding, with each character corresponding to an ASCII value, such as "D" corresponding to 68, "T" to 84, "U" to 85, and "-" to 45, forming the data source identifier numerical sequence. Then, the power grid equipment attribute parameters are decomposed to obtain the equipment identifier, such as the transformer asset number "BZY-1001". This is also converted into a numerical sequence using ASCII encoding, resulting in the equipment identifier numerical sequence. The operational fluctuations are then quantized and encoded: severe fluctuations are assigned a value of 3, moderate fluctuations are assigned a value of 2, and stability is assigned a value of 1, resulting in operational fluctuation data. Next, the data source identifier value sequence, the device identifier value sequence, and the operational fluctuation data are concatenated to obtain a quantized encoding vector. Finally, the quantized encoding vector is processed according to the key length requirements of the encryption algorithm. If the AES-256 algorithm is selected, a 256-bit key is required. Each value in the quantized encoding vector is converted into an 8-bit binary number, and then concatenated to form a binary string. If the length is less than 256 bits, the values ​​in the quantized encoding vector are padded cyclically until the length is met. If the length exceeds the requirement, the first 256 bits are truncated to generate a compliant encryption key. This process ensures that the encryption algorithm and key precisely match the security requirements of the data, avoiding security risks caused by a mismatch between algorithm strength and key. The benefit is improved encryption security of the data to be encrypted, while ensuring a close connection between the key generation process and the data, reducing the risk of key leakage.

[0029] Finally, the data to be encrypted in the target distribution network balance area is encrypted using at least an encryption algorithm and an encryption key. Fault-related data from the target distribution network balance area is collected to obtain fault alarm signals and emergency dispatch instructions; based on the fault level and emergency priority, the encryption level decision result is determined; based on the encryption level decision result, encryption algorithm, and encryption key, a second encryption algorithm and a second encryption key are obtained; the data to be encrypted is compressed to obtain compressed data to be encrypted; the compressed data to be encrypted is then encrypted using the second encryption algorithm and the second encryption key. The data to be encrypted in the target distribution network balance area is encrypted using at least an encryption algorithm and an encryption key. Simultaneously, in fault scenarios, the data to be transmitted is processed through a specific process. Fault-related data refers to information collected when a fault occurs in the distribution network balance area, including equipment fault type, fault location, and fault impact range. Fault alarm signals are system alert signals issued when a fault occurs, such as overcurrent alarm signals, overvoltage alarm signals, and equipment abnormality alarm signals. Emergency dispatch instructions are operational instructions issued to handle the fault, such as load transfer instructions, equipment shutdown instructions, and emergency repair dispatch instructions. The fault level is a classification based on the fault's impact range and severity, such as a Level 1 fault (large area). The fault levels are classified into three levels: Level 1 (power outage), Level 2 (partial power outage), and Level 3 (single device failure). Emergency priority is a classification based on the urgency of the fault handling, such as emergency priority (response within 10 minutes), general priority (response within 30 minutes), and low priority (response within 1 hour). The encryption level decision result is the encryption level of the data to be transmitted determined by combining the fault level and emergency priority. The second encryption algorithm is the encryption algorithm selected based on this decision result. The second encryption key is the key that matches the second encryption algorithm. The data to be transmitted is the fault alarm signal and emergency dispatch instruction that need to be transmitted in the fault scenario. The compressed data to be encrypted is the data after compression processing of the data to be transmitted.In practical implementation, firstly, in a normal scenario, a predetermined encryption algorithm and key are used to encrypt the data to be encrypted using symmetric encryption. The original data is divided into blocks according to the algorithm requirements, and each block is XORed and transformed with the key to generate encrypted data. Next, in a fault scenario, fault-related data is collected through sensors and monitoring systems in the distribution network. For example, overcurrent alarm signals are generated by current sensors detecting overcurrent, and emergency dispatch instructions for load transfer are issued through the dispatch center system. Then, based on the fault level and emergency priority, an encryption level decision is determined, establishing corresponding rules: Level 1 faults and emergency priorities correspond to Level 1 encryption, Level 2 faults and general priorities correspond to Level 2 encryption, and Level 3 faults and low priorities correspond to Level 3 encryption. Finally, based on this decision, a second encryption algorithm is selected, and Level 1 encryption is applied. AES-256 is selected for secondary encryption, AES-128 for secondary encryption, and DES for tertiary encryption. Following the previous key generation logic, a second encryption key is generated by combining the fault location code and emergency command number. The data to be transmitted is then compressed using a compression algorithm. A sliding window is used to find duplicate data sequences, which are replaced with pointers and length identifiers to reduce data volume, resulting in compressed data to be encrypted. Finally, the compressed data is encrypted using the second encryption algorithm and the second encryption key, processing data blocks according to the algorithm's key length requirements to complete the encryption operation. This approach ensures data security in both normal and fault scenarios. Furthermore, data compression improves transmission efficiency in fault scenarios, balancing data security and fault handling timeliness, thus ensuring efficient data transmission and emergency response in the distribution network.

[0030] The core reason for designing a complex encryption process is that the balance zone, as a key unit for load scheduling, security management, and fault response in the distribution network, directly determines the stable operation of the regional power grid and the guarantee of electricity supply for residents. The distribution network balance zone contains a large amount of sensitive data. The load information in the data to be encrypted is related to the electricity consumption patterns of regional industrial production and residential life, while behavioral information involves the core parameters and operating status of power grid equipment. Without multi-dimensional encryption and hierarchical protection, leaks could be maliciously exploited to analyze power grid load vulnerabilities, triggering targeted electricity attacks or causing equipment parameters to be tampered with, thereby disrupting the load balance of the balance zone and causing local power outages or equipment damage. Furthermore, the distribution network balance zone has a complex topology and diverse equipment types. Equipment at different node levels and load data with varying degrees of fluctuation have significantly different security requirements. A single encryption method cannot cover all scenarios. The complex encryption process generates customized keys and algorithms by combining multi-dimensional information such as data source identifiers, equipment attributes, and operational fluctuations. This accurately matches the security requirements of different data in the balance zone, ultimately achieving dual protection of data security and operational efficiency in the balance zone, building a solid data protection barrier for the stable and reliable operation of the entire distribution network.

[0031] Another embodiment of the present invention provides a method for selecting an encryption method after classifying data streams by sensitivity, as detailed below: Firstly, in the time dimension, the powerful time-series feature extraction capabilities of temporal convolutional networks are leveraged to analyze the long-term and short-term characteristics of user electricity load, energy storage device charging and discharging load, and new energy power generation load (wind power, photovoltaic) in the distribution network balance zone. The specific operation is as follows: First, the continuous time-step load values ​​of each device in the distribution network balance zone are processed and converted into a three-dimensional tensor. Then, short-term and long-term time-series patterns are extracted by different convolution kernels. The short-term window (minute level) is used to detect sudden fluctuations, such as a sudden drop in photovoltaic power or instantaneous wind speed disturbances; the long-term window (hour / day level) is used to identify regular patterns, such as the periodic charging and discharging of energy storage devices or the photovoltaic solar cycle.

[0032] Finally, the pooled feature sequence is further mapped to a higher-dimensional feature space through a nonlinear connection layer, increasing the model's expressive power and ultimately outputting a feature vector. After obtaining the extracted basic time features, this invention further calculates volatility factors to enhance the interpretability of the time dimension in sensitivity analysis. Volatility factors mainly include peak-to-valley difference, variance, volatility, and abrupt change detection index. Volatility measures the dynamics of the load by the rate of change between adjacent time steps. In addition, an abrupt change detection index is introduced to identify abnormal changes in the load curve in real time.

[0033] To avoid oversensitivity, a threshold h is further set for mutation detection. The threshold can be determined by statistical analysis of historical data and can be a multiple of the sequence standard deviation. The burst detection result can be converted into a sensitivity enhancement factor. Finally, a comprehensive volatility feature value is formed by weighted linear fusion. The comprehensive volatility feature value is concatenated with the output feature of TCN (Temporal Convolutional Network) to form an enhanced time dimension input.

[0034] Semantic feature extraction primarily targets the identification and analysis of command content during power distribution network dispatching, classifying sensitivity levels based on the importance of the identified command categories. Taking single-sentence recognition as an example, the command content is typically a sentence. The command content is input into the model, and firstly, the sentence is segmented into individual words. Then, based on information from the model's vocabulary, each word is converted into its corresponding encoded representation, resulting in an encoded sequence. An attention mask is created based on the sentence information to determine word positions. The rule is that if a position truly exists, the mask is set to 1; if a position needs to be filled, the mask is set to 0. After the above conversion, encoding, and attention mask creation operations, the sentence has been converted into a numerical format that the machine learning model can process. The converted code is then input into a multi-layer transformer structure, which works through a self-attention mechanism, iteratively calculating to analyze the correlation between words. Feature vectors are recalculated for individual words in the sentence, and then a significance score is calculated for each word. Combining attention weights and semantic distribution, a threshold is set to determine the keyword components of the sentence, retaining the keyword set. The threshold serves as a basic screening threshold, typically set to 0.5. Furthermore, a keyword judgment mechanism based on dynamic word frequency tracking is adopted to ensure that keyword judgment fully adapts to changes in actual conditions. Each round of processing instruction information... Assuming 5000 instruction messages are processed daily, 2000 messages are selected per round; each processing... An update is automatically triggered after each instruction message. When an update is triggered, the frequency of all keywords is extracted to obtain the current round of keyword frequency table. Combined with historical keyword frequencies, the current round of keyword frequency is calculated, with a higher weight of 0.7 to better reflect the latest changes.

[0035] After sorting the original word frequencies of the above words from high to low, a sorted word frequency sequence is obtained. The lowest 10% of the remaining word frequencies are taken as the emergency dispatch instruction category, which is directly assigned the highest sensitivity and is not included in subsequent classification calculations.

[0036] Next, semantic similarity is calculated between keyword vectors and dictionary entry vectors. A threshold is set for the accuracy of semantic judgment. As shown in Table 1, a basic domain dictionary is built based on a large number of actual instruction samples, defining the correspondence between categories and basic sensitivity levels, and making a preliminary judgment on sensitivity levels.

[0037] A real-time dynamic update mechanism is introduced based on word frequency-sensitivity mapping analysis to dynamically adjust the sensitivity level. Specifically, keywords with higher frequencies have lower weights due to their commonality, resulting in a lower sensitivity level; conversely, keywords with lower frequencies have higher weights due to their rarity, resulting in a higher sensitivity level. This ensures that the dictionary can reflect the latest power dispatching business environment in real time. Based on the above, the keyword frequency intervals for classification judgment are divided, and the interval boundaries are dynamically calculated based on the word frequency distribution in this round. To highlight the flexibility and applicability of word frequency updates while retaining the inherent scenario characteristics of the power industry, if a statement contains keywords of multiple categories, the highest level is taken as the initial semantic judgment result. Finally, the keyword vector and the initial sensitivity judgment result are combined to generate a semantic dimension vector, which is then aggregated with other dimensional features and input into the model for further sensitivity analysis.

[0038] At the behavioral level, considering the varying degrees of data security impact on different types of equipment participating in power distribution network dispatching under different business scenarios, user attributes are directly identified and preliminarily coded based on the different security situations. This serves as a further supplement to the analysis of user-level load data sensitivity. The specific operations are as follows: User attributes are digitally encoded based on three characteristics: business scenario, equipment category, and participation in power grid dispatch and demand response. After encoding, each user attribute is transformed into a feature vector. The above time, semantic and behavioral features are standardized and combined with actual sensitivity samples input into the MLP (Multilayer Perceptron). Through training, a weight matrix and bias vector are generated. Furthermore, based on the influence relationship between the combination of behavioral features and time and semantic features on sensitivity, dynamic weight factors are generated and their weighted sum is used to integrate the influence of behavioral dimensions into the time and semantic feature vectors. In the spatial dimension, only the node type is one-hot encoded. The feature vectors from the above three dimensions are concatenated and aggregated, and then input into the transformer encoder layer.

[0039] To provide a more intuitive probabilistic interpretation of the matching results, the Softmax function is introduced, which uses an exponential function to amplify the differences in scores between different categories, making the probability distribution more "sharp" and the classification decision results more accurate.

[0040] To further compare and analyze the cost-saving effect and security level of the above methods, a simple model of the processing of four encryption algorithms is constructed. Computational complexity factors are set for each of the four encryption algorithms. Security factor, transmission time factor This is used to differentiate the simulation experiments. A simple model is also created for the data encryption, transmission, and decryption processes. A simplified model of the encryption process is as follows: in, The encryption processing time (s) is the CPU time required to simulate the algorithm execution. The length of the original data; As the base time constant, and as a scaling factor, it can be adjusted reasonably according to the simulation time scale; f is the computational complexity factor corresponding to the encryption level. The length of the encrypted data; This represents the encryption overhead factor, which is positively correlated with the complexity factor, and indicates the inflation rate of the encrypted data. The simplified model of the transmission process is as follows: in, Network transmission time (s) The basic transmission time constant represents the transmission time per unit data length and per unit transmission factor, and can be reasonably adjusted according to the simulation time scale. The decryption process can be simplified as follows: in, Indicates the decryption processing time (s); This represents the decryption / encryption time ratio. Taking into full account that the decryption process may be slightly faster than the encryption process (e.g., SM2 decryption is faster than encryption), an average parameter can be set to simplify the model. After sensitivity analysis and dynamic selection of encryption algorithms, the total latency for encryption, transmission, and decryption of the data is: By comparing latency, the cost savings can be measured, providing a basis for optimizing reporting frequency. Simultaneously, to quantify the overall security performance of the solution, an average reliability rate metric is introduced. Safety Adequacy Ratio (AR) and Safety Score ; Where N is the total number of samples. This represents the single-pass security factor corresponding to the encryption algorithm used in the nth sample. The number of samples that meet the safety conditions is counted; the AR safety adequacy rate is a key safety indicator for evaluating the correctness of AI decision-making. These are the weighting factors for average reliability and safety adequacy, respectively. Ultimately, the simple model and index calculation formula constructed above can be used to compare and analyze the balance between security and efficiency of this AI dynamic encryption scheme, and measure its optimization effect.

[0041] One embodiment of the present invention provides a data encryption system for the balanced area of ​​a distribution network. For details, please refer to [link / reference]. Figure 2 , Figure 2 The diagram shown is a schematic representation of a data encryption system for a power distribution network balance zone according to one embodiment of the present invention, comprising: The parsing module 11 is used to parse and process the data to be encrypted in the target distribution network balance area to obtain the load information and behavior information of the data to be encrypted. The first module 12 is used to determine the security risk result of the data to be encrypted based on the data source information corresponding to the behavioral information, and to determine the first encryption level of the load information based on the security risk result and the operational fluctuation determined by the load information. The second module 13 is used to determine the second encryption level of the behavioral information based on the topological relationship of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information. The key module 14 is used to determine the encryption algorithm of the data to be encrypted based on the first encryption level and the second encryption level; and to quantize and encode the data source information, power grid equipment attribute parameters and operating fluctuations to generate an encryption key that matches the encryption algorithm. The encryption module 15 is used to encrypt the data to be encrypted in the target distribution network balance area using at least an encryption algorithm and an encryption key.

[0042] Preferably, based on the security risk results and the operational fluctuations determined by the load information, a first encryption level for the load information is determined, including: The extraction unit is used to extract multi-dimensional time-series data from load information; Network units are used to construct three-dimensional tensors from multi-dimensional time series data and input them into a pre-built analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics. The analysis unit is used to analyze the characteristics of short-term sudden fluctuations and long-term regularities to determine the operational fluctuation situation; The first unit is used to construct a two-dimensional judgment matrix by combining the security risk results with the operational fluctuations to obtain the first encryption level.

[0043] Preferably, the second module 13 includes: Node units are used to extract the node connection structure from the topology of the target distribution network to obtain the node hierarchy type; The equipment unit is used to parse behavioral information to obtain the attribute parameters of the power grid equipment; The encoding uses one-hot encoding for node level types and digital encoding for power grid equipment attribute parameters to obtain a two-dimensional feature vector; Sensitivity unit, used to analyze two-dimensional feature vectors to obtain sensitivity fusion features; The second encryption unit is used to obtain the second encryption level of behavioral information based on the sensitivity fusion features.

[0044] Preferably, the key module 14 includes: The identification unit is used to extract the data source identifier from the data source information. The disassembly unit is used to disassemble the attribute parameters of power grid equipment to obtain the equipment identifier; The quantization unit is used to quantize and encode operational fluctuations to obtain operational fluctuation data. The encoding vector unit is used to obtain a quantized encoding vector based on the data source identifier, device identifier, and operational fluctuation data; An encryption key unit is used to generate an encryption key based on the quantization encoding vector and the key length requirements of the encryption algorithm.

[0045] Preferably, the encryption module 15 further includes: The data acquisition unit is used to collect fault-related data of the target distribution network balance zone and obtain fault alarm signals and emergency dispatch instructions. The decision-making unit is used to determine the encryption level decision based on the fault level and emergency priority. The second encryption unit is used to obtain the second encryption algorithm and the second encryption key based on the encryption level decision result, the encryption algorithm and the encryption key; The compression unit is used to compress the data to be encrypted, resulting in compressed data to be encrypted. The compression encryption unit is used to encrypt compressed data to be encrypted based on a second encryption algorithm and a second encryption key.

[0046] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0047] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform steps in the data encryption method for the distribution network balance zone as described in the above embodiments, for example... Figure 1 Steps S1 to S5 as described above.

[0048] This invention parses the data to be encrypted to obtain load and behavioral information, determines security risk results based on the data source information corresponding to the behavioral information, and dynamically generates a first encryption level by combining the operational fluctuations reflected by the load information. Simultaneously, a second encryption level is determined based on the topology of the target distribution network and the power grid equipment attribute parameters indicated by the behavioral information, thereby achieving refined identification of security requirements for different data under different operating scenarios. Based on this, the solution adaptively selects an encryption algorithm according to the first and second encryption levels, and quantifies and encodes the data source information, the power grid equipment attribute parameters, and the operational fluctuations to generate an encryption key matching the selected encryption algorithm. This enables differentiated and scenario-aware precise encryption of the data to be encrypted. Therefore, it effectively avoids the problems of insufficient protection of high-risk data or over-encryption of low-risk data caused by the use of uniform preset encryption rules in existing technologies. While ensuring the security of key instructions and sensitive data in the distribution network balance zone, it significantly reduces the terminal computing burden and communication resource consumption, improving the overall system operating efficiency and security resilience.

[0049] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A data encryption method for a balanced distribution network zone, characterized in that, include: The data to be encrypted in the target distribution network balance zone is parsed and processed to obtain the load information and behavior information of the data to be encrypted. Based on the data source information corresponding to the behavioral information, the security risk result of the data to be encrypted is determined, and based on the security risk result and the operational fluctuation determined by the load information, the first encryption level of the load information is determined; Based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information, a second encryption level of the behavioral information is determined; Based on the first encryption level and the second encryption level, an encryption algorithm for the data to be encrypted is determined; the data source information, the power grid equipment attribute parameters, and the operating fluctuation are quantized and encoded to generate an encryption key that matches the encryption algorithm; The data to be encrypted in the target distribution network balance zone is encrypted using at least the encryption algorithm and the encryption key.

2. The data encryption method for the balance zone of a distribution network as described in claim 1, characterized in that, The determination of the first encryption level of the load information based on the security risk results and the operational fluctuations determined by the load information includes: Extract multi-dimensional time-series data from the load information; The multi-dimensional time series data is used to construct a three-dimensional tensor and input into a pre-constructed analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics. The short-term sudden fluctuation characteristics and the long-term regularity characteristics are analyzed to determine the operational fluctuation situation; The security risk results and the operational fluctuations are used to construct a two-dimensional judgment matrix to obtain the first encryption level.

3. The data encryption method for the distribution network balance zone as described in claim 1, characterized in that, The step of determining the second encryption level of the behavioral information based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavioral information includes: Extract the node connection structure from the topology of the target distribution network to obtain the node hierarchy type; The behavioral information is parsed to obtain the attribute parameters of the power grid equipment; One-hot encoding is performed on the node level type, and digital encoding is performed on the attribute parameters of the power grid equipment to obtain a two-dimensional feature vector; The sensitivity fusion features are obtained by analyzing the two-dimensional feature vectors. Based on the aforementioned sensitivity fusion features, a second encryption level for behavioral information is obtained.

4. The data encryption method for the balanced distribution network area as described in claim 1, characterized in that, The step of quantizing and encoding the data source information, the power grid equipment attribute parameters, and the operational fluctuations to generate an encryption key matching the encryption algorithm includes: The data source identifier is extracted from the data source information; The attribute parameters of the power grid equipment are decomposed to obtain the equipment identifier; The operational fluctuations are quantified and encoded to obtain operational fluctuation data; Based on the data source identifier, the device identifier, and the operational fluctuation data, a quantization encoding vector is obtained; The encryption key is generated based on the quantization encoding vector and the key length requirement of the encryption algorithm.

5. The data encryption method for the balanced distribution network area as described in claim 1, characterized in that, The step of encrypting the data to be encrypted in the target distribution network balance area using at least the encryption algorithm and the encryption key further includes: Collect fault-related data of the target distribution network balance zone to obtain fault alarm signals and emergency dispatch instructions; The encryption level decision is determined based on the fault level and emergency priority. Based on the encryption level decision result, the encryption algorithm, and the encryption key, a second encryption algorithm and a second encryption key are obtained; The data to be encrypted is compressed to obtain compressed data to be encrypted. The compressed data to be encrypted is encrypted based on the second encryption algorithm and the second encryption key.

6. A data encryption system for a balanced distribution network zone, characterized in that, include: The parsing module is used to parse and process the data to be encrypted in the target distribution network balance area to obtain the load information and behavior information of the data to be encrypted. The first module is used to determine the security risk result of the data to be encrypted based on the data source information corresponding to the behavioral information, and to determine the first encryption level of the load information based on the security risk result and the operating fluctuation determined by the load information. The second module is used to determine the second encryption level of the behavior information based on the topology of the target distribution network and the attribute parameters of the power grid equipment indicated by the behavior information. The key module is used to determine the encryption algorithm of the data to be encrypted based on the first encryption level and the second encryption level; and to quantize and encode the data source information, the power grid equipment attribute parameters and the operating fluctuations to generate an encryption key that matches the encryption algorithm. An encryption module is used to encrypt the data to be encrypted in the target distribution network balance zone using at least the encryption algorithm and the encryption key.

7. The data encryption system for the distribution network balance zone as described in claim 6, characterized in that, The determination of the first encryption level of the load information based on the security risk results and the operational fluctuations determined by the load information includes: The extraction unit is used to extract multi-dimensional time-series data from the load information; The network unit is used to construct a three-dimensional tensor from the multi-dimensional time series data and input it into a pre-constructed analysis network to obtain short-term sudden fluctuation characteristics and long-term regularity characteristics. The analysis unit is used to analyze the short-term sudden fluctuation characteristics and the long-term regularity characteristics to determine the operational fluctuation situation; The first unit is used to construct a two-dimensional judgment matrix by combining the security risk results with the operational fluctuations to obtain the first encryption level.

8. The data encryption system for the distribution network balance zone as described in claim 6, characterized in that, The second module includes: Node unit, used to extract node connection structure from the topology of the target distribution network to obtain node hierarchy type; The device unit is used to parse the behavioral information to obtain the attribute parameters of the power grid equipment; The node level type is encoded using one-hot encoding, and the attribute parameters of the power grid equipment are digitally encoded to obtain a two-dimensional feature vector. A sensitivity unit is used to analyze the two-dimensional feature vector to obtain sensitivity fusion features; The second encryption unit is used to obtain a second encryption level for behavioral information based on the sensitivity fusion features.

9. The data encryption system for the distribution network balance zone as described in claim 6, characterized in that, The key module includes: An identification unit is used to extract a data source identifier from the data source information; The disassembly unit is used to disassemble the attribute parameters of the power grid equipment to obtain the equipment identifier; A quantization unit is used to quantize and encode the operational fluctuations to obtain operational fluctuation data. The encoding vector unit is used to obtain a quantized encoding vector based on the data source identifier, the device identifier, and the operational fluctuation data; An encryption key unit is used to generate the encryption key based on the quantization encoding vector and the key length requirement of the encryption algorithm.

10. The data encryption system for the distribution network balance zone as described in claim 6, characterized in that, The encryption module further includes: The data acquisition unit is used to collect fault-related data of the target distribution network balance zone and obtain fault alarm signals and emergency dispatch instructions. The decision-making unit is used to determine the encryption level decision based on the fault level and emergency priority. The second encryption unit is used to obtain a second encryption algorithm and a second encryption key based on the encryption level decision result, the encryption algorithm, and the encryption key; A compression unit is used to compress the data to be encrypted to obtain compressed data to be encrypted; A compression encryption unit is used to encrypt the compressed data to be encrypted based on the second encryption algorithm and the second encryption key.

Images