Integrated power data safety monitoring and analyzing system

Abstract

The invention discloses an integrated power data safety monitoring and analysis system, which relates to the technical field of power data safety and comprises a data acquisition module, a safety monitoring module, a risk assessment module, an early warning decision module and an emergency response module. Advanced technologies and innovative algorithms are integrated, efficient cooperation of all the modules is achieved, comprehensive, accurate and pre-processing of data acquisition, precise anomaly detection of safety monitoring, scientific and accurate grading of risk assessment, timely and effective early warning decision making and rapid and appropriate emergency response are included, the innovative algorithms improve the accuracy and reliability of monitoring analysis, and the system is suitable for large-scale popularization and application. The method adapts to safety requirements of different scales of power systems and changes, can effectively cope with novel safety threats, reduces safety incident loss to the greatest extent, and guarantees stable operation of the power systems.

Description

Technical Field

[0001] The present invention relates to the technical field of power data security, and in particular to an integrated power data security monitoring and analysis system. Background Art

[0002] In today's era of high dependence on electricity, the stable operation of the power system is of vital importance. With the continuous development of the power system and the acceleration of the intelligent process, a large amount of power data is generated, transmitted and stored. These power data cover everything from the operating parameters of power equipment to the overall status of the power grid, from the user's power consumption information to the impact of the environment on the power system. However, the importance of power data also makes it face many security risks.

[0003] There are some serious problems in the existing power data security monitoring and analysis systems. On the one hand, many system functions are scattered, and there is a lack of effective collaborative working mechanism between modules. For example, each link often acts independently and cannot form an organic whole. This not only leads to a waste of resources, but also reduces the efficiency and response speed of the system. On the other hand, the existing algorithms lack innovation and are difficult to cope with the increasingly complex power data security challenges. Traditional anomaly detection methods may not be able to accurately identify new security threats, and risk assessment algorithms may not be able to fully consider the impact of various factors, resulting in inaccurate assessment results. In addition, the existing systems are also insufficient in dealing with large-scale data processing and real-time monitoring, and it is difficult to meet the high requirements of the power system for data security. Summary of the invention

[0004] The purpose of the present invention is to provide an integrated power data security monitoring and analysis system to solve the deficiencies in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solutions: an integrated power data safety monitoring and analysis system, including a data acquisition module, a safety monitoring module, a risk assessment module, a warning decision module and an emergency response module.

[0006] Preferably, the data acquisition module uses high-precision intelligent sensors to collect operating parameters and environmental data of power equipment in real time, and transmits the data to the safety monitoring module through a combination of wireless communication technology and optical fiber communication technology. During the data transmission process, the edge computing device is used to perform preliminary processing on the collected data, including data cleaning, denoising, compression, encryption and other operations.

[0007] Preferably, the safety monitoring module adopts an anomaly detection algorithm that combines a generative adversarial network based on deep learning and an autoencoder. The generator generates data similar to normal power data, the discriminator distinguishes between real data and data generated by the generator, and the autoencoder encodes and decodes normal power data to learn the potential feature representation of the data. When new data is input, the reconstruction error is first judged by the autoencoder. If it exceeds a certain threshold, it is considered that an anomaly may exist, and then the discriminator is used to further confirm whether the data is abnormal data.

[0008] Preferably, the risk assessment module establishes a scientific and reasonable power data security risk assessment index system, adopts a risk assessment algorithm based on the combination of fuzzy logic and hierarchical analysis method, establishes a judgment matrix through expert scoring to determine the indicator weights, performs a consistency check, and then determines the fuzzy evaluation set and membership function of each indicator based on the collected data and expert experience to obtain a fuzzy evaluation matrix, and determines the risk level of power data security through fuzzy synthesis operations.

[0009] Preferably, the early warning decision module adopts an early warning decision algorithm based on reinforcement learning, represents the safety status of power data as a state vector, selects an appropriate early warning level and decision action according to the current state, designs a reasonable reward function, and uses historical data to train the reinforcement learning algorithm to continuously optimize the early warning decision strategy.

[0010] Preferably, the emergency response module formulates corresponding emergency plans according to different risk levels and types of security incidents, takes emergency measures when a security incident occurs and notifies relevant personnel to handle it, and tracks and records the security incidents.

[0011] The beneficial effects of the present invention are as follows:

[0012] In the present invention, firstly, by integrating advanced technologies and innovative algorithms, efficient collaboration between various modules of the system is achieved. The data acquisition module can accurately and comprehensively collect power data, providing a reliable data basis for subsequent safety monitoring and analysis. The safety monitoring module adopts a unique anomaly detection algorithm combining a generative adversarial network and an autoencoder, which can accurately identify potential abnormal data and improve the safety of the system. The risk assessment module uses an algorithm based on a combination of fuzzy logic and hierarchical analysis to scientifically and reasonably determine the safety risk level of power data, providing an accurate basis for early warning decisions. The early warning decision module is based on a reinforcement learning algorithm and can make timely and accurate early warnings and decisions according to the real-time status of the system, effectively reducing safety risks. The emergency response module quickly activates the emergency plan when a safety incident occurs, takes effective emergency measures, and minimizes losses.

[0013] Secondly, the system of the present invention has good scalability and adaptability. Whether it is power systems of different sizes or the ever-changing power data security requirements, the system can be flexibly adjusted and optimized to meet various complex application scenarios;

[0014] Finally, the innovative algorithm of this system improves the accuracy and reliability of power data security monitoring and analysis. The combination of generative adversarial networks and autoencoders, as well as the application of reinforcement learning in early warning decision-making, enable the system to better respond to new security threats and provide strong guarantees for the stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a functional schematic diagram of each module of the present invention;

[0016] Figure 2 Schematic diagram of the operating steps of the system of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The present invention provides Figure 1 An integrated power data safety monitoring and analysis system shown includes a data acquisition module, a safety monitoring module, a risk assessment module, a warning decision module and an emergency response module.

[0019] The data acquisition module uses high-precision intelligent sensors to collect the operating parameters and environmental data of the power equipment in real time, and transmits the data to the safety monitoring module through a combination of wireless communication technology and optical fiber communication technology. During the data transmission process, the edge computing device is used to perform preliminary processing on the collected data, including data cleaning, denoising, compression and encryption.

[0020] The safety monitoring module adopts an anomaly detection algorithm that combines a generative adversarial network based on deep learning and an autoencoder. The generator generates data similar to normal power data, the discriminator distinguishes between real data and data generated by the generator, and the autoencoder encodes and decodes normal power data to learn the potential feature representation of the data. When new data is input, the reconstruction error is first judged by the autoencoder. If it exceeds a certain threshold, it is considered that there may be an anomaly, and then the discriminator is used to further confirm whether the data is abnormal data.

[0021] The risk assessment module establishes a scientific and reasonable power data security risk assessment indicator system, adopts a risk assessment algorithm based on the combination of fuzzy logic and hierarchical analysis method, establishes a judgment matrix through expert scoring to determine the indicator weights, performs consistency test, and then determines the fuzzy evaluation set and membership function of each indicator based on the collected data and expert experience to obtain the fuzzy evaluation matrix, and determines the risk level of power data security through fuzzy synthesis operation.

[0022] The early warning decision module adopts an early warning decision algorithm based on reinforcement learning, represents the safety status of power data as a state vector, selects the appropriate early warning level and decision action according to the current state, designs a reasonable reward function, and uses historical data to train the reinforcement learning algorithm to continuously optimize the early warning decision strategy.

[0023] The emergency response module formulates corresponding emergency plans according to different risk levels and types of security incidents, takes emergency measures when security incidents occur and notifies relevant personnel to handle them, while tracking and recording security incidents.

[0024] The data acquisition module, safety monitoring module, risk assessment module, early warning decision module and emergency response module of the present invention also include the following contents:

[0025] 1. Data acquisition module

[0026] The data acquisition module uses high-precision intelligent sensors, such as current sensors, voltage sensors, temperature sensors, humidity sensors, etc., which are installed in key parts of power equipment to collect operating parameters and environmental data of power equipment in real time;

[0027] These sensors have self-calibration, self-diagnosis and remote communication functions, which can ensure the accuracy and reliability of data. At the same time, they use a combination of high-speed and stable wireless communication technology and optical fiber communication technology to quickly and accurately transmit the collected data to the safety monitoring module. During the data transmission process, the edge computing device performs preliminary processing on the collected data, including data cleaning, denoising, compression and encryption, to improve the quality and security of the data.

[0028] 2. Safety Monitoring Module

[0029] This module uses an anomaly detection algorithm that combines a deep learning-based generative adversarial network (GAN) and an autoencoder (AE). This algorithm can effectively identify anomalies in power data and provide a strong guarantee for the safe operation of the system.

[0030] Model construction: The generative adversarial network consists of a generator and a discriminator. The generator is used to generate data similar to normal power data, and the discriminator is used to distinguish between real data and data generated by the generator. At the same time, an autoencoder is constructed, consisting of an encoder and a decoder, to encode and decode normal power data and learn the potential feature representation of the data;

[0031] Model training: define the generator's loss function L G = -log(D(G(z))), where z is a random noise vector; the loss function of the discriminator is L D = -log(D(x))-log(1-D(G(z))), where x is the real data; the loss function of the autoencoder is L AE =||xD e (E(x))|| 2 where ||·|| 2 Express the binorm; the total loss function of joint training is L = αL D +βL G +γL AE Among them, α, β, and γ are weight coefficients, and these coefficients are adjusted to balance the contributions of the three parts;

[0032] A large amount of normal power data is used to jointly train the generative adversarial network and the autoencoder. During the training process, the generator continuously generates data close to the real data, the discriminator continuously improves its ability to distinguish between real data and generated data, and the autoencoder continuously optimizes its encoding and decoding capabilities.

[0033] Abnormal judgment: When new data x new When input, it is first encoded and decoded through the autoencoder to calculate the reconstruction error e = ||x new -D e (E(x new ))|| 2 If the reconstruction error exceeds a certain threshold T AE , then the data may be abnormal, and then the data is input into the discriminator of the generative adversarial network. If the discriminator determines that the data is the probability D(x new ) is high, it is further confirmed that the data is abnormal data. That is, when D(x new )>T D When it is judged as abnormal, T D is the threshold of the discriminator.

[0034] 3. Risk Assessment Module

[0035] The risk assessment module establishes a scientific and reasonable power data security risk assessment indicator system, including data integrity risk, data confidentiality risk, data availability risk, system stability risk and other aspects;

[0036] Adopt risk assessment algorithm based on fuzzy logic and analytic hierarchy process to ensure the accuracy and reliability of assessment results;

[0037] Determine the indicator weights using the analytic hierarchy process: divide the risk assessment indicator system into the target layer, the criterion layer, and the indicator layer;

[0038] By means of expert scoring, a judgment matrix A = (a ij ) n×n , where a ij Indicates the relative importance of the i-th indicator and the j-th indicator, and calculates the maximum eigenvalue λ of the judgment matrix max and the corresponding eigenvector W = (w1, w2, ..., w n ) T ;

[0039] Conduct consistency checks and consistency indicators The random consistency index RI can be obtained by looking up the table, and the consistency ratio If CR<0.1, the judgment matrix is ​​considered to have satisfactory consistency, and the eigenvector W is the weight vector of each indicator;

[0040] Fuzzy logic assessment risk level: According to the collected data and expert experience, the fuzzy evaluation set V = {V1, V2, ..., Vm} and the membership function of each indicator are determined. For each indicator i, its membership rij to each evaluation level is calculated according to its actual value, and the fuzzy evaluation matrix R = (rij) nxm is obtained. Through fuzzy logic reasoning, each indicator is evaluated to obtain the fuzzy evaluation result B = WoR = (b1, b2, ..., bm), where o represents the fuzzy synthesis operation, which can be calculated by weighted average method and other methods. The risk level of power data security is determined according to the fuzzy evaluation results.

[0041] 4. Early Warning Decision Module

[0042] The early warning decision module adopts an early warning decision algorithm based on reinforcement learning, which can make timely and accurate warnings and decisions according to the real-time status of the system, effectively reducing safety risks;

[0043] State representation: The safety state of power data is represented as a state vector S = (s1, s2, ..., sk), including multiple factors such as the number of abnormal data, risk level, system load, etc.

[0044] Action selection: According to the current state S, select the appropriate warning level and decision action A, such as issuing a low-level warning, a medium-level warning, a high-level warning, or taking corresponding security measures, such as strengthening data encryption, increasing access control, etc.;

[0045] Reward function design: According to the effect of early warning decision, a reasonable reward function R(S,A) is designed. If the early warning decision can prevent the occurrence of security incidents in a timely and effective manner, a positive reward will be given; if the early warning decision is inappropriate, resulting in the occurrence of security incidents or aggravating the impact of security incidents, a negative reward will be given;

[0046] Reinforcement learning algorithm training: Use historical data to train the reinforcement learning algorithm, continuously optimize the early warning decision strategy, and improve the accuracy and effectiveness of early warning decisions. The Q-learning algorithm can be used, and its update formula is: Where Q(S,A) represents the Q value of taking action A in state S, α is the learning rate, γ is the discount factor, and S′ is the next state.

[0047] 5. Emergency Response Module

[0048] The emergency response module formulates corresponding emergency plans according to different risk levels and types of security incidents, including emergency response procedures, emergency resource allocation, and division of personnel responsibilities;

[0049] When a security incident occurs, the emergency plan is immediately activated and appropriate emergency measures are taken, such as cutting off the power supply of the faulty equipment, isolating the affected network area, restoring data backup, etc. At the same time, relevant personnel are notified in a timely manner to handle the emergency, and the security incident is tracked and recorded for subsequent analysis and summary.

[0050] like Figure 2 As shown, the system of the present invention includes the following specific embodiments:

[0051] System deployment: Install the sensors of the data acquisition module at key locations of the power equipment, deploy the safety monitoring module, risk assessment module, early warning decision module and emergency response module in the control center of the power system, connect each module through the network to form an integrated power data security monitoring and analysis system.

[0052] Data collection and preprocessing: Sensors collect operating parameters and environmental data of power equipment in real time, transmit the data to edge computing devices for preprocessing through data transmission technology, and then transmit the preprocessed data to the safety monitoring module.

[0053] Safety monitoring and anomaly detection: The safety monitoring module uses an anomaly detection algorithm to monitor power data in real time. When abnormal data is detected, a warning signal is immediately issued and the abnormal data is transmitted to the risk assessment module.

[0054] Risk assessment and early warning decision-making: The risk assessment module uses a risk assessment algorithm to assess the security risks of power data. The early warning decision module uses an early warning decision algorithm to issue early warnings and execute decisions based on the risk assessment results.

[0055] Emergency response and handling: When a security incident occurs, the emergency response module immediately activates the emergency plan and takes appropriate emergency measures to handle it.

[0056] In summary, the integrated power data security monitoring and analysis system of the present invention has significant technical advantages and practical value, can effectively improve the security of power data, and through the detailed design and collaborative work of each module, provides a strong guarantee for the safe and stable operation of the power system.

[0057] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An integrated power data security monitoring and analysis system, characterized by: It includes data acquisition module, safety monitoring module, risk assessment module, early warning decision module and emergency response module; The safety monitoring module adopts an anomaly detection algorithm that combines a generative adversarial network based on deep learning and an autoencoder. The generator generates data similar to normal power data, the discriminator distinguishes between real data and data generated by the generator, and the autoencoder encodes and decodes normal power data to learn the potential feature representation of the data. When new data is input, the reconstruction error is first judged by the autoencoder. If it exceeds the threshold, it is considered that there may be an anomaly, and then the discriminator is used to further confirm whether the data is abnormal data.

2. The integrated power data security monitoring and analysis system according to claim 1 is characterized in that: The data acquisition module uses high-precision intelligent sensors to collect the operating parameters and environmental data of the power equipment in real time, and transmits the data to the safety monitoring module through a combination of wireless communication technology and optical fiber communication technology. During the data transmission process, the edge computing device is used to perform preliminary processing on the collected data, including data cleaning, denoising, compression and encryption operations.

3. The integrated power data security monitoring and analysis system according to claim 1 is characterized in that: The risk assessment module establishes a power data security risk assessment index system, adopts a risk assessment algorithm based on the combination of fuzzy logic and hierarchical analysis method, establishes a judgment matrix through expert scoring to determine the indicator weights, performs consistency verification, and then determines the fuzzy evaluation set and membership function of each indicator based on the collected data and expert experience to obtain a fuzzy evaluation matrix, and determines the risk level of power data security through fuzzy synthesis operations.

4. The integrated power data security monitoring and analysis system according to claim 3 is characterized in that: The early warning decision module adopts an early warning decision algorithm based on reinforcement learning, represents the safety status of power data as a state vector, selects the early warning level and decision action according to the current status, designs a reward function, and uses historical data to train the reinforcement learning algorithm to continuously optimize the early warning decision strategy.

5. The integrated power data security monitoring and analysis system according to claim 4 is characterized in that: The emergency response module formulates corresponding emergency plans according to different risk levels and types of security incidents, takes emergency measures when security incidents occur and notifies relevant personnel to handle them, while tracking and recording security incidents.

Images