A power grid communication method based on quantum key

Abstract

The present invention proposes a power grid communication method based on quantum key, which constructs an optimal QKD protocol selection neural network model, and trains the optimal QKD protocol selection neural network model based on a training set to obtain the trained optimal QKD protocol selection neural network model; by obtaining the secret number rate, detection efficiency, background error, number of light pulses sent by a communication sender and communication distance between two communicating parties of the detector corresponding to the QKD protocol applicable to the target quantum hardware as input, the data transmission rate of the QKD protocol applicable to the target quantum hardware is predicted, and an optimal QKD protocol is selected as the target QKD protocol; power grid communication is performed based on the target QKD protocol, thereby improving the security and efficiency of power grid communication.

Description

Technical Field

[0001] The present invention relates to the field of power grid communication, and in particular to a power grid communication method based on quantum keys. Background Art

[0002] With the continuous development of power systems and the improvement of their intelligence level, the requirements for the security and efficiency of power system communications are becoming increasingly higher. Traditional power system communication methods have problems such as low security and vulnerability to attacks. Therefore, it is of great significance to study a safe, efficient and reliable power system communication method.

[0003] Quantum key distribution (QKD) is an encryption technology based on the principles of quantum mechanics, which can share keys between communicating parties to ensure the security of communication. However, the existing QKD technology still has certain limitations in practical applications, such as limited transmission distance and high cost. Therefore, in order to further improve the security and efficiency of power system communication, it is necessary to select a QKD protocol suitable for the current scenario. There are already methods for selecting QKD protocols, but their efficiency and performance are low.

[0004] Neural network is a computational model that simulates the structure of neurons in the human brain and has powerful learning and processing capabilities. In recent years, neural networks have achieved remarkable results in image recognition, speech recognition and other fields. However, there are relatively few studies on the application of neural networks in the field of power system communications and the selection of QKD. Summary of the invention

[0005] The purpose of the present invention is to provide a power grid communication method based on quantum key to improve the security and efficiency of power grid communication.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A power grid communication method based on neural network and quantum key comprises the following steps:

[0008] S1: Obtain N QKD protocols, where N is an integer greater than 2;

[0009] S2: Based on simulation, obtain the secret rate Y0, detection efficiency η, background error ed of the detectors of the N QKD protocols, the number of optical pulses N sent by the communication sender, and the communication distance L between the two communicating parties, and combine the above parameters into a 5-dimensional vector X = [Y0, ed, η, N, L], the data transmission rate Y of the N QKD protocols for different scenarios, and divide the obtained data into a training set and a test set;

[0010] S3: constructing an optimal QKD protocol selection neural network model, and training the optimal QKD protocol selection neural network model based on a training set to obtain a trained optimal QKD protocol selection neural network model;

[0011] S4: Obtain target quantum hardware, obtain M QKD protocols suitable for the target quantum hardware from N QKD protocols, where M≤N, select a neural network model based on the optimal QKD protocol after training, obtain the dark mark rate, detection efficiency, background error, number of optical pulses sent by the communication sender, and communication distance between the two communicating parties of the detector corresponding to the M QKD protocols suitable for the target quantum hardware as input, predict the data transmission rate of the M QKD protocols suitable for the target quantum hardware, and select the optimal QKD protocol as the target QKD protocol;

[0012] S5: Perform power grid communication based on the target QKD protocol.

[0013] Furthermore, the specific structure of the optimal QKD protocol selection neural network model is as follows:

[0014] Input layer, first stage feature extraction module, second stage feature extraction module, third stage feature extraction module, Relu layer, output layer;

[0015] The first-stage feature extraction module to the third-stage feature extraction module are all used for feature extraction, and the predicted transmission rate is obtained through the output layer.

[0016] The first-stage feature extraction module includes a first convolution layer, a second convolution layer, a first feature fusion layer, a third convolution layer, a second feature fusion layer, a fourth convolution layer, and a third feature fusion layer connected in sequence;

[0017] The first feature fusion layer fuses the outputs of the first convolutional layer and the second convolutional layer;

[0018] The second feature fusion layer fuses the outputs of the first feature fusion layer and the third convolutional layer;

[0019] The third feature fusion layer fuses the outputs of the second feature fusion layer and the fourth convolutional layer;

[0020] The output of the third feature fusion layer is used as the output of the first-stage feature extraction module.

[0021] The second stage feature extraction module includes:

[0022] The first branch convolution layer B1, the second branch convolution layer B2, the first branch fusion layer, the third branch convolution layer B3, the fourth branch convolution layer B4, the second branch fusion layer;

[0023] The filter of the first branch convolutional layer B1 is set to 16;

[0024] The filter of the second branch convolutional layer B2 is set to 32;

[0025] The filter of the third branch convolutional layer B3 is set to 32;

[0026] The filter of the fourth branch convolutional layer B4 is set to 64;

[0027] The structures of the first branch fusion layer and the second branch fusion layer are both: Conv+Maxpooling;

[0028] The output of the second branch fusion layer is used as the output of the second stage feature extraction module.

[0029] The third-stage feature extraction module includes three sequentially connected separable convolution modules, and the structure of the separable convolution module is SepConv+Relu+BN.

[0030] The present invention proposes a power grid communication method based on quantum key, which constructs an optimal QKD protocol selection neural network model, and trains the optimal QKD protocol selection neural network model based on a training set to obtain the trained optimal QKD protocol selection neural network model; by obtaining the secret number rate, detection efficiency, background error, number of light pulses sent by a communication sender and communication distance between two communicating parties of the detector corresponding to the QKD protocol applicable to the target quantum hardware as input, the data transmission rate of the QKD protocol applicable to the target quantum hardware is predicted, and an optimal QKD protocol is selected as the target QKD protocol; power grid communication is performed based on the target QKD protocol, thereby improving the security and efficiency of power grid communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific implementation of the present invention or the technical solution in the prior art, the drawings required for describing the specific implementation or the prior art will be briefly introduced below. Obviously, the drawings described below are only one implementation of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 This is the structural diagram of the first stage feature extraction module;

[0033] Figure 2 This is the structural diagram of the second stage feature extraction module;

[0034] Figure 3This is the structural diagram of the third stage feature extraction module. DETAILED DESCRIPTION

[0035] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the technical scheme in the specific implementation mode of the present invention is clearly and completely described below to further illustrate the present invention. Obviously, the specific implementation mode described is only a part of the implementation mode of the present invention, rather than all styles.

[0036] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0037] A power grid communication method based on neural network and quantum key comprises the following steps:

[0038] S1: Obtain N QKD protocols, where N is an integer greater than 2;

[0039] S2: Based on simulation, obtain the secret rate Y0, detection efficiency η, background error ed of the detectors of the N QKD protocols, the number of optical pulses N sent by the communication sender, and the communication distance L between the two communicating parties, and combine the above parameters into a 5-dimensional vector X = [Y0, ed, η, N, L], the data transmission rate Y of the N QKD protocols for different scenarios, and divide the obtained data into a training set and a test set;

[0040] S3: constructing an optimal QKD protocol selection neural network model, and training the optimal QKD protocol selection neural network model based on a training set to obtain a trained optimal QKD protocol selection neural network model;

[0041] S4: Obtain target quantum hardware, obtain M QKD protocols applicable to the target quantum hardware, where M≤N, select a neural network model based on the optimal QKD protocol after training, obtain the dark mark rate, detection efficiency, background error, number of optical pulses sent by the communication sender, and communication distance between the two communicating parties of the detector corresponding to the M QKD protocols applicable to the target quantum hardware as input, predict the data transmission rate of the M QKD protocols applicable to the target quantum hardware, and select the optimal one QKD protocol as the target QKD protocol;

[0042] S5: Perform power grid communication based on the target QKD protocol.

[0043] Furthermore, the specific structure of the optimal QKD protocol selection neural network model is as follows:

[0044] Input layer, first stage feature extraction module, second stage feature extraction module, third stage feature extraction module, Relu layer, output layer;

[0045] The first-stage feature extraction module to the third-stage feature extraction module are all used for feature extraction, and the predicted transmission rate is obtained through the output layer.

[0046] As attached Figure 1 As shown, the first-stage feature extraction module includes a first convolution layer, a second convolution layer, a first feature fusion layer, a third convolution layer, a second feature fusion layer, a fourth convolution layer, and a third feature fusion layer connected in sequence;

[0047] The first feature fusion layer fuses the outputs of the first convolutional layer and the second convolutional layer;

[0048] The second feature fusion layer fuses the outputs of the first feature fusion layer and the third convolutional layer;

[0049] The third feature fusion layer fuses the outputs of the second feature fusion layer and the fourth convolutional layer;

[0050] The output of the third feature fusion layer is used as the output of the first-stage feature extraction module.

[0051] As attached Figure 2 As shown, the second stage feature extraction module includes:

[0052] The first branch convolution layer B1, the second branch convolution layer B2, the first branch fusion layer, the third branch convolution layer B3, the fourth branch convolution layer B4, the second branch fusion layer;

[0053] The filter of the first branch convolutional layer B1 is set to 16;

[0054] The filter of the second branch convolutional layer B2 is set to 32;

[0055] The filter of the third branch convolutional layer B3 is set to 32;

[0056] The filter of the fourth branch convolutional layer B4 is set to 64;

[0057] The structures of the first branch fusion layer and the second branch fusion layer are both: Conv+Maxpooling;

[0058] The output of the second branch fusion layer is used as the output of the second stage feature extraction module.

[0059] As attached Figure 3 As shown, the third-stage feature extraction module includes three separable convolution modules connected in sequence, and the structure of the separable convolution module is SepConv+Relu+BN.

[0060] The present invention proposes a power grid communication method based on quantum key, which constructs an optimal QKD protocol selection neural network model, and trains the optimal QKD protocol selection neural network model based on a training set to obtain the trained optimal QKD protocol selection neural network model; by obtaining the secret number rate, detection efficiency, background error, number of light pulses sent by a communication sender and communication distance between two communicating parties of the detector corresponding to the QKD protocol applicable to the target quantum hardware as input, the data transmission rate of the QKD protocol applicable to the target quantum hardware is predicted, and an optimal QKD protocol is selected as the target QKD protocol; power grid communication is performed based on the target QKD protocol, thereby improving the security and efficiency of power grid communication.

[0061] The above describes the main technical features and basic principles of the present invention and the related advantages. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the concept or basic features of the present invention. Therefore, no matter from which point of view, the above specific embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the present invention.

[0062] In addition, it should be understood that although the present specification is described according to various implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each implementation method may also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A power grid communication method based on quantum key, characterized in that: The method comprises the following steps: S1: Obtain N QKD protocols, where N is an integer greater than 2; S2: Based on simulation, obtain the secret rate Y0, detection efficiency η, background error ed of the detectors of the N QKD protocols, the number of optical pulses N sent by the communication sender, and the communication distance L between the two communicating parties, and combine the above parameters into a 5-dimensional vector X = [Y0, ed, η, N, L], the data transmission rate Y of the N QKD protocols for different scenarios, and divide the obtained data into a training set and a test set; S3: constructing an optimal QKD protocol selection neural network model, and training the optimal QKD protocol selection neural network model based on a training set to obtain a trained optimal QKD protocol selection neural network model; S4: Obtain target quantum hardware, obtain M QKD protocols suitable for the target quantum hardware from N QKD protocols, where M≤N, select a neural network model based on the optimal QKD protocol after training, obtain the dark mark rate, detection efficiency, background error, number of optical pulses sent by the communication sender, and communication distance between the two communicating parties of the detector corresponding to the M QKD protocols suitable for the target quantum hardware as input, predict the data transmission rate of the M QKD protocols suitable for the target quantum hardware, and select the optimal QKD protocol as the target QKD protocol; S5: Performing grid communication based on the target QKD protocol; The specific structure of the optimal QKD protocol selection neural network model is as follows: Input layer, first stage feature extraction module, second stage feature extraction module, third stage feature extraction module, Relu layer, output layer; The first stage feature extraction module to the third stage feature extraction module are all used for feature extraction, and the predicted transmission rate is obtained through the output layer; The first-stage feature extraction module includes a first convolution layer, a second convolution layer, a first feature fusion layer, a third convolution layer, a second feature fusion layer, a fourth convolution layer, and a third feature fusion layer connected in sequence; The first feature fusion layer fuses the outputs of the first convolutional layer and the second convolutional layer; The second feature fusion layer fuses the outputs of the first feature fusion layer and the third convolutional layer; The third feature fusion layer fuses the outputs of the second feature fusion layer and the fourth convolutional layer; The output of the third feature fusion layer is used as the output of the first stage feature extraction module; The second stage feature extraction module includes: The first branch convolution layer B1, the second branch convolution layer B2, the first branch fusion layer, the third branch convolution layer B3, the fourth branch convolution layer B4, the second branch fusion layer; The filter of the first branch convolutional layer B1 is set to 16; The filter of the second branch convolutional layer B2 is set to 32; The filter of the third branch convolutional layer B3 is set to 32; The filter of the fourth branch convolutional layer B4 is set to 64; The structures of the first branch fusion layer and the second branch fusion layer are both: Conv+Maxpooling; The output of the second branch fusion layer is used as the output of the second stage feature extraction module; The third-stage feature extraction module includes three sequentially connected separable convolution modules, and the structure of the separable convolution module is SepConv+Relu+BN.

Images