A safety control method for an industrial control cabinet

Abstract

The application relates to the technical field of intelligent control, and discloses an industrial-grade control cabinet safety control method, which comprises the following steps: step 1, collecting quantum arbitrary number generator data to generate a key and evaluating the key entropy by using an edit distance measure; step 2, adjusting a key update strategy by using Markov decision according to the key generated in step 1 and updating the key when an attack is detected; step 3, collecting network traffic according to the update strategy in step 2 to build a topology analysis model, determining a network attack by using persistent homology, and issuing an alarm; and step 4, building an adjacency matrix according to the traffic data and equipment connection information in step 3 and calculating a graph Laplacian matrix. The application adopts quantum arbitrary number generation and Markov decision to optimize the key update strategy, dynamically updates the key when an attack is detected, guarantees the randomness of the key and the anti-attack ability, and can effectively reduce the key leakage risk and improve the system anti-quantum computing attack ability compared with the existing static key management scheme.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, specifically to a safety control method for industrial-grade control cabinets. Background Technology

[0002] Industrial-grade control cabinets are a crucial component of modern industrial automation and intelligent manufacturing, widely used in the power, communications, transportation, and intelligent manufacturing industries. Their core functions include equipment control, data processing, and network communication. However, the continuous upgrading of network attack methods, the increasing complexity of system topologies, and the growing demand for human-machine interaction have revealed numerous shortcomings in existing control cabinets regarding security, maintainability, and user experience, hindering the pace of intelligent development.

[0003] Traditional control cabinets primarily rely on static key mechanisms for communication encryption and authentication. However, if these static keys are leaked or cracked, the system's security is severely threatened. Furthermore, with the development of quantum computing technology, the resistance to attacks from traditional encryption algorithms has significantly decreased, making the system vulnerable to quantum computing attacks. Existing solutions typically rely on updating keys at fixed intervals, but lack intelligent dynamic adjustment mechanisms, failing to respond promptly when attacks are detected, leading to potential security vulnerabilities.

[0004] The topology of industrial control networks is not fixed; the addition or removal of equipment, changes in connection status, and dynamic adjustments to network load all affect the overall system architecture. Currently, most control cabinet systems use manual topology configuration, which is cumbersome and prone to errors. Furthermore, most existing topology learning methods are based on static graph models, which cannot adapt to real-time changing network environments, resulting in limited system scalability and high maintenance costs.

[0005] Industrial control cabinet display interfaces are typically used for status monitoring and operation command input. Different ambient lighting conditions can affect display quality, leading to illegible information and even misoperation. Existing interface layouts are usually preset and lack the ability to adapt to lighting conditions, failing to maintain optimal readability in various environments. Some systems employ automatic brightness adjustment technology, but this fails to fully integrate with changes in system topology, resulting in poor interface adaptability and a limited user experience.

[0006] Therefore, those skilled in the art provide an industrial-grade control cabinet safety control method to solve the problems mentioned above. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides an industrial-grade control cabinet safety control method to solve the problems mentioned in the background section.

[0008] To achieve the above objectives, the present invention provides the following technical solution: 1. An industrial-grade control cabinet safety control method, comprising:

[0009] Step S1: Collect data from the quantum arbitrary number generator to generate a key and evaluate the key entropy using the edit distance metric;

[0010] Step S2: Based on the key generated in step S1, the key update strategy is adjusted using Markov decision and the key is updated when an attack is detected.

[0011] Step S3: Based on the update strategy in step S2, collect network traffic to build a topology analysis model to persistently identify network attacks and issue alarms.

[0012] Step S4: Construct an adjacency matrix and calculate the graph Laplace matrix based on the traffic data and device connection information from step S3.

[0013] Step S5: Based on the graph Laplacian matrix from step S4, train a graph convolutional network to determine module connections and update the system topology;

[0014] Step S6: Based on the system topology in step S5, new data is collected when device connections change to construct an adjacency matrix and the system structure is adjusted according to the updated topology;

[0015] Step S7: Based on the system structure in step S6, collect the lighting data of the display interface to construct a lighting model and use differentiable rendering to calculate the interface layout to complete the display update.

[0016] In step S7, the interface lighting model M calculated in step S6 I Furthermore, the adjusted interface layout includes:

[0017] Step S7.1: Based on the illumination influence matrix G calculated in step S6 and the optimized interface layout scheme C, collect the final interface display data:

[0018] D I ={d1,d2,...,d n}, where d n To define the interface display state in the k-th time slice, define the interface display feature matrix: X D =ψ(D I ,C), where ψ(·) is a mapping function based on display data and layout scheme to calculate interface stability and user interaction experience;

[0019] Step S7.2, based on the interface display feature matrix X calculated in step S7.1 D Construct an interface stability evaluation function:

[0020]

[0021] Where αij is the interface element e i and e jInteraction weights between ||X Di -X Dj || 2 To calculate the relative degree of change of interface elements in order to evaluate the consistency and stability of the interface display;

[0022] Step S7.3: Based on the interface stability S calculated in step S7.2, define the final interface optimization target:

[0023] Ω D =argmaxg(S,C,G),

[0024] Where g(·) is the optimization objective function that comprehensively considers interface stability, layout scheme, and illumination effects, based on Ω. D Calculate the final interface adjustment scheme and feed back the interface optimization information to step S6 to further optimize the lighting model and interface layout.

[0025] Preferably, step S1, when the output data of the quantum arbitrary number generator satisfies the stable sampling condition, further includes:

[0026] Step S1.1, define the output data set of the quantum arbitrary number generator:

[0027] R = {r1, r2, ..., r} n},

[0028] Where r1 is any number obtained from the first sampling, r2 is any number obtained from the second sampling, and r n Let n be any number obtained in the nth sampling, where n is the number of samplings;

[0029] Calculate key candidates:

[0030] Among them, a i r is the weight factor corresponding to the i-th sampling. i Let K1 be any number obtained from the i-th sampling, and K1 be a key candidate;

[0031] Step S1.2, define the reference sequence: T = {t1, t2, ..., t} m}, where t1 is the first element of the reference sequence, t2 is the second reference value, and t m is the last element of the reference sequence, m is the length of the reference sequence, and T is the reference sequence;

[0032] Calculate edit distance:

[0033] Where D is the sum of the mismatch indices between key candidate K1 and reference sequence T at each position, and β j Let L be the mismatch index at position j, and L be the comparison length.

[0034] Step S1.3, define the normalized entropy: H = D / L,

[0035] Where H is the entropy value of the key candidate, D is the mismatch index of the key candidate K1 and the reference sequence T at each position, and L is the comparison length;

[0036] Set a high entropy threshold y. If H≥y, determine that the key candidate meets the high entropy requirement.

[0037] Preferably, in step S2, after the key candidates determined in step S1 meet the high entropy requirement, the following further steps are included:

[0038] Step S2.1: Based on the key candidate K1 calculated in step S1 and the reference sequence T, define the key update probability distribution: p K ={p1,p2,...,p n},

[0039] Where p1 is the probability that key candidate K1 takes the first state, p2 is the probability that key candidate K1 takes the second state, and p n To determine the probability that key candidate K1 takes the nth state, construct a Markov decision process: M = (S, A, P, R, γ).

[0040] Where S is the set of key states, A is the set of update policies, P is the state transition probability matrix, R is the state reward function, and γ is the discount factor;

[0041] Step S2.2: Based on the Markov decision process constructed in step S2.1, calculate the state transition probabilities:

[0042]

[0043] Where P(s′|s,a) is the probability that the current state s transitions to the next state s′ from policy a, s is the current state, s′ is the next state, a is the key update policy adopted, and δ i p is the adjustment factor for the i-th state. n Let K1 be the probability that the key candidate K1 takes the nth state. To update the probability distribution p of the key K Perform a summation calculation;

[0044] Optimize strategy A to maximize the cumulative reward function:

[0045]

[0046] Where J(A) is the cumulative reward function, γ is the discount factor, and R(s) is the cumulative reward function. t ,a tTo take strategy a at time step t t And it is in state s t The reward value obtained at that time To perform cumulative calculations over all time steps t;

[0047] Step S2.3: After the update strategy A0 obtained from the optimization in step S2.2 is determined, based on the key update trigger condition Θ, it is determined whether to perform a key update, and the trigger condition is defined as follows:

[0048] Θ = f(D,H,η),

[0049] Where D is the sum of the mismatch indexes between key candidate K1 and reference sequence T at each position, H is the entropy value of key candidate, and η is the attack detection feedback signal;

[0050] If Θ exceeds the set threshold, perform a key update and replace the old key K0 with the new key K1.

[0051] Preferably, step S3, after determining the key update strategy A0 obtained in step S2, further includes:

[0052] Step S3.1: Based on the new key K1 and update policy A0 determined in step S2, collect network traffic data: X = {x1, x2, ..., x...} m}, where x m For the data packets collected in the m-th time slice, where m is the total number of time slices collected, define a network traffic feature vector:

[0053] F = g(X, K1),

[0054] Where F is the network traffic feature vector, and g(·) is the key-weighted feature extraction function to generate traffic topology data for subsequent attack detection;

[0055] Step S3.2: Construct a topology data analysis model based on the traffic feature vector F calculated in step S3.1.

[0056] G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, W is the edge weight matrix, and G is the network traffic topology data analysis model.

[0057] Calculate the topological persistent homology feature: P = h(G), where P is the topological persistent homology feature and h(·) is the topological feature extraction function to identify potential anomalous patterns;

[0058] Step S3.3: Based on the topological features P extracted in step S3.2, define the attack detection judgment function:

[0059] θ=ψ(P,K1,A0),

[0060] Where θ is the attack detection judgment function, ψ(·) is the detection function that comprehensively considers topological features, key and update strategy, and an attack judgment threshold θ0 is set. If θ≥θ0, an alarm signal is triggered and attack information η is fed back to step S2 to adjust the key update strategy.

[0061] Preferably, step S4, after determining the network attack characteristics and triggering the alarm signal in step S3, further includes:

[0062] Step S4.1, based on the topology data analysis model constructed in step S3:

[0063] G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, G is the network traffic topology data analysis model, and W is the edge weight matrix, W = [w ij ], where w ij For node v i With v j The strength of the connection between them;

[0064] Define adjacency matrix A: A = [a ij ], where a ij For node v i With v j The connection between them, if there is a direct connection: a ij =w ij Otherwise a ij =0, v i v j Let i and j be the network nodes in the topology model;

[0065] Calculate the graph Laplacian matrix: L = DA, where D is the degree matrix and L is the graph Laplacian matrix, which is used to represent the global information of the network topology and serves as the input for subsequent graph convolution.

[0066] Step S4.2: Based on the graph Laplacian matrix L calculated in step S4.1, train the system topology model M using a graph convolutional network. t =f(L,X),

[0067] Where X is the node feature matrix, f(·) is the graph convolution operation, and the node embedding is calculated as follows:

[0068] Z = σ(LXW), where W is the training parameter matrix and σ(·) is the activation function to learn the module connection method in the topology;

[0069] Step S4.3, based on the system topology model M obtained in step S4.2... t Define the topology update strategy:

[0070] Ω = argmaxg(Z,P), where g(·) is the objective function based on node embedding and topology feature optimization. The system topology is adjusted according to Ω, and the topology change information is fed back to step S3 to optimize the subsequent network attack detection and key update strategy.

[0071] Preferably, in step S5, the graph Laplacian matrix L calculated in step S4 and the training system topology model M are used... t The next step includes:

[0072] Step S5.1, based on the topology model M obtained in step S4... t Collect data on changes in device connections:

[0073] Y = {y1, y2, ..., y} k}, where y k Define a new adjacency matrix for the device connection information recorded in the k-th time slice: A′=[a′ ij ], where a′ ij For node v i With v j A new connection relationship exists between them, if a direct connection exists: a′ ij =w′ ij Otherwise: a′ ij =0, calculate the updated graph Laplacian matrix: L′=D′-A′, where D′ is the updated degree matrix;

[0074] Step S5.2: Based on the updated graph Laplacian matrix L′ calculated in step S5.1, a new topology model, M, is calculated using a graph convolutional network. t ′=f(L′,X′), where X′ is the updated node feature matrix, f(·) is the graph convolution operation, and the new node embedding is calculated as:

[0075] Z′=σ(L′X′W′), where W′ is the new training parameter matrix and σ(·) is the activation function to optimize the system topology;

[0076] Step S5.3: Based on the new topology model M calculated in step S5.2 t Define the topology optimization objective:

[0077] Ω′=argmax g(Z′,ξ),

[0078] Wherein, g(·) is an optimization function calculated based on node embedding and topology change ξ, which readjusts the system topology according to Ω′ and feeds back the new topology adjustment information to step S4 to further optimize subsequent network analysis and attack detection.

[0079] Preferably, step S6, after calculating the new topology in step S5 and adjusting the system topology, further includes:

[0080] Step S6.1: Based on the updated graph Laplacian matrix L′ and optimized topology obtained in step S5, collect the illumination data of the display interface:

[0081] I = {i1, i2, ..., i} m}, where i m Define the illumination model for the illumination data of the m-th time slice:

[0082] M I =φ(I,L′), where φ(·) is a mapping function based on lighting data and topology, used to calculate the impact of lighting on the interface display;

[0083] Step S6.2, based on the lighting model M calculated in step S6.1 I Construct the illumination effect matrix:

[0084] G = [g ij ], where g ij For interface element e i and e j The degree of influence of lighting between elements is defined by the differentiable rendering calculation formula: C = σ(GX) I W I ),

[0085] Among them, X I W is the lighting feature matrix of the interface elements. I The interface layout adjustment scheme is calculated using the training parameter matrix and σ(·) as the activation function.

[0086] Step S6.3: Based on the interface layout adjustment scheme C calculated in step S6.2, define the final interface update strategy: Ω I =argmaxf(C,Z′),

[0087] Where f(·) is the optimization objective function, which adjusts the interface based on the calculated interface lighting information and topology embedding Z′, and feeds back the interface adjustment information to step S5 to optimize the system topology and interactive experience.

[0088] A terminal device includes a processor, a memory, and a communication interface for performing industrial-grade control cabinet security control methods.

[0089] A storage medium storing a computer program, which, when executed, enables a terminal device to implement an industrial-grade control cabinet safety control method.

[0090] This invention provides a safety control method for industrial-grade control cabinets. It has the following beneficial effects:

[0091] 1. This invention adopts a quantum arbitrary number generation and Markov decision optimization key update strategy, which dynamically updates the key when an attack is detected, ensuring key arbitrariness and attack resistance. Compared with existing static key management schemes, it can effectively reduce the risk of key leakage and improve the system's resistance to quantum computing attacks.

[0092] 2. This invention adopts a topology learning method based on graph convolutional networks, which can automatically learn and optimize the device connection structure, enabling the system to adapt to dynamically changing device topologies. Traditional methods rely on manual topology adjustments, which are inefficient and prone to errors. This solution can achieve intelligent adjustments, improve operation and maintenance efficiency, and support flexible system expansion.

[0093] 3. This invention combines light perception and differentiable rendering technology to achieve adaptive adjustment of the interface, ensuring that the interface is clearly visible under different ambient lighting conditions. Compared with the traditional fixed interface layout method, this invention can dynamically optimize the display effect according to changes in lighting, improve user visual comfort, and reduce the risk of misoperation. Attached Figure Description

[0094] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0095] To enable those skilled in the art to understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0096] The present invention will now be described in detail with reference to the accompanying drawings:

[0097] Example:

[0098] Please see the appendix Figure 1 This invention provides a safety control method for an industrial-grade control cabinet, characterized by comprising:

[0099] Step S1: Collect data from the quantum arbitrary number generator to generate a key and evaluate the key entropy using the edit distance metric;

[0100] Step S2: Based on the key generated in step S1, the key update strategy is adjusted using Markov decision and the key is updated when an attack is detected.

[0101] Step S3: Based on the update strategy in step S2, collect network traffic to build a topology analysis model to persistently identify network attacks and issue alarms.

[0102] Step S4: Construct an adjacency matrix and calculate the graph Laplace matrix based on the traffic data and device connection information from step S3.

[0103] Step S5: Based on the graph Laplacian matrix from step S4, train a graph convolutional network to determine module connections and update the system topology;

[0104] Step S6: Based on the system topology in step S5, new data is collected when device connections change to construct an adjacency matrix and the system structure is adjusted according to the updated topology;

[0105] Step S7: Based on the system structure in step S6, collect the lighting data of the display interface to construct a lighting model and use differentiable rendering to calculate the interface layout to complete the display update.

[0106] In step S1, when the output data of the quantum arbitrary number generator satisfies the stable sampling condition, the following is further included:

[0107] Step S1.1, define the output data set of the quantum arbitrary number generator:

[0108] R = {r1, r2, ..., r} n},

[0109] Where r1 is any number obtained from the first sampling, r2 is any number obtained from the second sampling, and r n Let n be any number obtained in the nth sampling, where n is the number of samplings;

[0110] Calculate key candidates:

[0111] Among them, a i r is the weight factor corresponding to the i-th sampling. i Let K1 be any number obtained from the i-th sampling, and K1 be a key candidate;

[0112] Step S1.2, define the reference sequence: T = {t1, t2, ..., t} m}, where t1 is the first element of the reference sequence, t2 is the second reference value, and t m is the last element of the reference sequence, m is the length of the reference sequence, and T is the reference sequence;

[0113] Calculate edit distance:

[0114] Where D is the sum of the mismatch indices between key candidate K1 and reference sequence T at each position, and β j Let L be the mismatch index at position j, and L be the comparison length.

[0115] Step S1.3, define the normalized entropy: H = D / L,

[0116] Where H is the entropy value of the key candidate, D is the mismatch index of the key candidate K1 and the reference sequence T at each position, and L is the comparison length;

[0117] Set a high entropy threshold y. If H≥y, determine that the key candidate meets the high entropy requirement.

[0118] In step S2, after the key candidates determined in step S1 meet the high entropy requirement, the following further steps are included:

[0119] Step S2.1: Based on the key candidate K1 calculated in step S1 and the reference sequence T, define the key update probability distribution: p K ={p1,p2,...,p n},

[0120] Where p1 is the probability that key candidate K1 takes the first state, p2 is the probability that key candidate K1 takes the second state, and p n To determine the probability that key candidate K1 takes the nth state, construct a Markov decision process: M = (S, A, P, R, γ).

[0121] Where S is the set of key states, A is the set of update policies, P is the state transition probability matrix, R is the state reward function, and γ is the discount factor;

[0122] Step S2.2: Based on the Markov decision process constructed in step S2.1, calculate the state transition probabilities:

[0123]

[0124] Where P(s′|s,a) is the probability that the current state s transitions to the next state s′ from policy a, s is the current state, s′ is the next state, a is the key update policy adopted, and δ i p is the adjustment factor for the i-th state. n Let K1 be the probability that the key candidate K1 takes the nth state. To update the probability distribution p of the key K Perform a summation calculation;

[0125] Optimize strategy A to maximize the cumulative reward function:

[0126]

[0127] Where J(A) is the cumulative reward function, γ is the discount factor, and R(s) is the cumulative reward function. t ,a tTo take strategy a at time step t t And it is in state s t The reward value obtained at that time To perform cumulative calculations over all time steps t;

[0128] Step S2.3: After the update strategy A0 obtained from the optimization in step S2.2 is determined, based on the key update trigger condition Θ, it is determined whether to perform a key update, and the trigger condition is defined as follows:

[0129] Θ = f(D,H,η),

[0130] Where D is the sum of the mismatch indexes between key candidate K1 and reference sequence T at each position, H is the entropy value of key candidate, and η is the attack detection feedback signal;

[0131] If Θ exceeds the set threshold, perform a key update and replace the old key K0 with the new key K1.

[0132] In step S3, after the key update strategy A0 obtained in step S2 is determined, the following further steps are included:

[0133] Step S3.1: Based on the new key K1 and update policy A0 determined in step S2, collect network traffic data: X = {x1, x2, ..., x...} m}, where x m For the data packets collected in the m-th time slice, where m is the total number of time slices collected, define a network traffic feature vector:

[0134] F = g(X, K1),

[0135] Where F is the network traffic feature vector, and g(·) is the key-weighted feature extraction function to generate traffic topology data for subsequent attack detection;

[0136] Step S3.2: Construct a topology data analysis model based on the traffic feature vector F calculated in step S3.1.

[0137] G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, W is the edge weight matrix, and G is the network traffic topology data analysis model.

[0138] Calculate the topological persistent homology feature: P = h(G), where P is the topological persistent homology feature and h(·) is the topological feature extraction function to identify potential anomalous patterns;

[0139] Step S3.3: Based on the topological features P extracted in step S3.2, define the attack detection judgment function:

[0140] θ=ψ(P,K1,A0),

[0141] Where θ is the attack detection judgment function, ψ(·) is the detection function that comprehensively considers topological features, key and update strategy, and an attack judgment threshold θ0 is set. If θ≥θ0, an alarm signal is triggered and attack information η is fed back to step S2 to adjust the key update strategy.

[0142] In step S4, after determining the network attack characteristics and triggering the alarm signal in step S3, the following further steps are included:

[0143] Step S4.1, based on the topology data analysis model constructed in step S3:

[0144] G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, G is the network traffic topology data analysis model, and W is the edge weight matrix, W = [w ij ], where w ij For node v i With v j The strength of the connection between them;

[0145] Define adjacency matrix A: A = [a ij ], where a ij For node v i With v j The connection between them, if there is a direct connection: a ij =w ij Otherwise a ij =0, v i v j Let i and j be the network nodes in the topology model;

[0146] Calculate the graph Laplacian matrix: L = DA, where D is the degree matrix and L is the graph Laplacian matrix, which is used to represent the global information of the network topology and serves as the input for subsequent graph convolution.

[0147] Step S4.2: Based on the graph Laplacian matrix L calculated in step S4.1, train the system topology model M using a graph convolutional network. t =f(L,X),

[0148] Where X is the node feature matrix, f(·) is the graph convolution operation, and the node embedding is calculated as follows:

[0149] Z = σ(LXW), where W is the training parameter matrix and σ(·) is the activation function to learn the module connection method in the topology;

[0150] Step S4.3, based on the system topology model M obtained in step S4.2... t Define the topology update strategy:

[0151] Ω = argmaxg(Z,P), where g(·) is the objective function based on node embedding and topology feature optimization. The system topology is adjusted according to Ω, and the topology change information is fed back to step S3 to optimize the subsequent network attack detection and key update strategy.

[0152] In step S5, the graph Laplacian matrix L calculated in step S4 and the training system topology model M are... t The next step includes:

[0153] Step S5.1, based on the topology model M obtained in step S4... t Collect data on changes in device connections:

[0154] Y = {y1, y2, ..., y} k}, where y k Define a new adjacency matrix for the device connection information recorded in the k-th time slice: A′=[a′ ij ], where a′ ij For node v i With v j A new connection relationship exists between them, if a direct connection exists: a′ ij =w′ ij Otherwise: a′ ij =0, calculate the updated graph Laplacian matrix: L′=D′-A′, where D′ is the updated degree matrix;

[0155] Step S5.2: Based on the updated graph Laplacian matrix L′ calculated in step S5.1, a new topology model, M, is calculated using a graph convolutional network. t ′=f(L′,X′), where X′ is the updated node feature matrix, f(·) is the graph convolution operation, and the new node embedding is calculated as:

[0156] Z′=σ(L′X′W′), where W′ is the new training parameter matrix and σ(·) is the activation function to optimize the system topology;

[0157] Step S5.3: Based on the new topology model M calculated in step S5.2 t Define the topology optimization objective:

[0158] Ω′=argmax g(Z′,ξ),

[0159] Wherein, g(·) is an optimization function calculated based on node embedding and topology change ξ, which readjusts the system topology according to Ω′ and feeds back the new topology adjustment information to step S4 to further optimize subsequent network analysis and attack detection.

[0160] In step S6, after calculating the new topology in step S5 and adjusting the system topology, the following further steps are included:

[0161] Step S6.1: Based on the updated graph Laplacian matrix L′ and optimized topology obtained in step S5, collect the illumination data of the display interface:

[0162] I = {i1, i2, ..., i} m}, where i m Define the illumination model for the illumination data of the m-th time slice:

[0163] M I =φ(I,L′), where φ(·) is a mapping function based on lighting data and topology, used to calculate the impact of lighting on the interface display;

[0164] Step S6.2, based on the lighting model M calculated in step S6.1 I Construct the illumination effect matrix:

[0165] G = [g ij ], where g ij For interface element e i and e j The degree of influence of lighting between elements is defined by the differentiable rendering calculation formula: C = σ(GX) I W I ),

[0166] Among them, X I W is the lighting feature matrix of the interface elements. I The interface layout adjustment scheme is calculated using the training parameter matrix and σ(·) as the activation function.

[0167] Step S6.3: Based on the interface layout adjustment scheme C calculated in step S6.2, define the final interface update strategy: Ω I =argmaxf(C,Z′),

[0168] Where f(·) is the optimization objective function, which adjusts the interface based on the calculated interface lighting information and topology embedding Z′, and feeds back the interface adjustment information to step S5 to optimize the system topology and interactive experience.

[0169] In step S7, the interface lighting model M calculated in step S6 I Furthermore, the adjusted interface layout includes:

[0170] Step S7.1: Based on the illumination influence matrix G calculated in step S6 and the optimized interface layout scheme C, collect the final interface display data:

[0171] DI ={d1,d2,...,d n}, where d n To define the interface display state in the k-th time slice, define the interface display feature matrix: X D =ψ(D I ,C), where ψ(·) is a mapping function based on display data and layout scheme to calculate interface stability and user interaction experience;

[0172] Step S7.2, based on the interface display feature matrix X calculated in step S7.1 D Construct an interface stability evaluation function:

[0173]

[0174] Where αij is the interface element e i and e j Interaction weights between them To calculate the relative degree of change of interface elements in order to evaluate the consistency and stability of the interface display;

[0175] Step S7.3: Based on the interface stability S calculated in step S7.2, define the final interface optimization target:

[0176] Ω D =argmaxg(S,C,G),

[0177] Where g(·) is the optimization objective function that comprehensively considers interface stability, layout scheme, and illumination effects, based on Ω. D Calculate the final interface adjustment scheme and feed back the interface optimization information to step S6 to further optimize the lighting model and interface layout.

[0178] The advantage of step S1 is that it uses a quantum arbitrary number generator to ensure the arbitrariness of the key, which improves encryption security. In addition, it combines the edit distance entropy evaluation method to ensure that the generated key meets the high entropy requirement and effectively resists quantum computing attacks.

[0179] The benefit of step S2 is that by using a Markov decision process to dynamically adjust the key update strategy, the level of intelligence in key management is improved, and the key can be quickly adjusted when an attack is detected, thus avoiding the security risks caused by the key remaining unchanged for a long time.

[0180] The benefits of step S3 include collecting real-time traffic data, extracting topological features using persistent coherence methods, accurately detecting abnormal traffic patterns, and feeding back to the key management system when an alarm signal is triggered, thereby achieving linkage optimization between key and security detection.

[0181] The benefit of step S4 is that it builds a network topology model based on traffic data and device connection relationships, improves the understanding of the system structure, and calculates the graph Laplace matrix to obtain global topology features, providing a foundation for subsequent topology optimization.

[0182] The benefits of step S5 are that it uses graph convolutional networks to train the topology model, realizes automated topology learning and optimization, improves network adaptability, and can automatically adjust the topology structure when device connections change, reducing manual intervention and improving operation and maintenance efficiency.

[0183] The advantage of step S6 is that by combining ambient lighting data and using differentiable rendering technology, the interface can be dynamically adjusted, improving readability. Compared with traditional static interface solutions, this method can automatically adapt to changes in lighting and reduce the risk of misoperation.

[0184] The benefits of step S7 are that by using the interface stability evaluation function, it ensures that the user experience can be maintained after the interface layout is adjusted, and by combining the lighting model to optimize the interface display scheme, the comfort of human-computer interaction and the accuracy of operation are improved.

[0185] In summary, this invention enhances the security, improves system maintainability, and optimizes the user experience of industrial-grade control cabinets through quantum key management, network topology optimization, and intelligent interface adjustment technologies. Unlike traditional solutions, this method dynamically adapts to environmental changes and optimizes the system structure by incorporating intelligent learning technology, providing a reliable and efficient security control solution for intelligent manufacturing, the Industrial Internet of Things, and critical infrastructure.

[0186] A terminal device includes a processor, a memory, and a communication interface for performing industrial-grade control cabinet security control methods.

[0187] A storage medium storing a computer program, which, when executed, enables a terminal device to implement an industrial-grade control cabinet safety control method.

[0188] The terminal device and storage medium provided by this invention enhance system security, improve maintainability, and optimize user experience by implementing an industrial-grade control cabinet security control method. Compared with traditional solutions, this solution can adaptively adjust keys, intelligently optimize topology, and dynamically adapt interfaces, making it widely applicable to intelligent manufacturing, industrial IoT, and critical infrastructure, providing a secure, efficient, and intelligent solution for industrial control systems.

[0189] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A safety control method for an industrial-grade control cabinet, characterized in that, include: Step S1: Collect data from the quantum arbitrary number generator to generate a key and evaluate the key entropy using the edit distance metric; Step S2: Based on the key generated in step S1, the key update strategy is adjusted using Markov decision and the key is updated when an attack is detected. Step S3: Based on the update strategy in step S2, collect network traffic to build a topology analysis model to persistently identify network attacks and issue alarms. Step S4: Construct an adjacency matrix and calculate the graph Laplace matrix based on the traffic data and device connection information from step S3. Step S5: Based on the graph Laplacian matrix from step S4, train a graph convolutional network to determine module connections and update the system topology; Step S6: Based on the system topology in step S5, new data is collected when device connections change to construct an adjacency matrix and the system structure is adjusted according to the updated topology; Step S7: Based on the system structure in step S6, collect the lighting data of the display interface to construct a lighting model and use differentiable rendering to calculate the interface layout to complete the display update. In step S7, the interface lighting model M calculated in step S6 I Furthermore, the adjusted interface layout includes: Step S7.1: Based on the illumination influence matrix G calculated in step S6 and the optimized interface layout scheme C, collect the final interface display data: D I ={d1,d2,...,d n }, where d n To define the interface display state in the k-th time slice, define the interface display feature matrix: X D =ψ(D I ,C), where ψ(·) is a mapping function based on display data and layout scheme to calculate interface stability and user interaction experience; Step S7.2, based on the interface display feature matrix X calculated in step S7.1 D Construct an interface stability evaluation function: Where, α ij For interface element e i and e j Interaction weights between them To calculate the relative degree of change of interface elements in order to evaluate the consistency and stability of the interface display; Step S7.3: Based on the interface stability S calculated in step S7.2, define the final interface optimization target: Ω D =argmaxg(S,C,G), Where g(·) is the optimization objective function that comprehensively considers interface stability, layout scheme, and illumination effects, based on Ω. D Calculate the final interface adjustment scheme and feed back the interface optimization information to step S6 to further optimize the lighting model and interface layout.

2. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S1, when the output data of the quantum arbitrary number generator satisfies the stable sampling condition, the following is further included: Step S1.1, define the output data set of the quantum arbitrary number generator: R＝{r1,r2,...,r n}， Where r1 is any number obtained from the first sampling, r2 is any number obtained from the second sampling, and r n Let n be any number obtained in the nth sampling, where n is the number of samplings; Calculate key candidates: Among them, a i r is the weight factor corresponding to the i-th sampling. i Let K1 be any number obtained from the i-th sampling, and K1 be a key candidate. Step S1.2, define the reference sequence: T = {t1, t2, ..., t} m }, where t1 is the first element of the reference sequence, t2 is the second reference value, and t m is the last element of the reference sequence, m is the length of the reference sequence, and T is the reference sequence; Calculate edit distance: Where D is the sum of the mismatch indices between key candidate K1 and reference sequence T at each position, and β j Let L be the mismatch index at position j, and L be the comparison length. Step S1.3, define the normalized entropy: H = D / L, Where H is the entropy value of the key candidate, D is the mismatch index of the key candidate K1 and the reference sequence T at each position, and L is the comparison length; Set a high entropy threshold y. If H≥y, determine that the key candidate meets the high entropy requirement.

3. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S2, after the key candidates determined in step S1 meet the high entropy requirement, the following further steps are included: Step S2.1: Based on the key candidate K1 calculated in step S1 and the reference sequence T, define the key update probability distribution: p K ={p1,p2,...,p n }, Where p1 is the probability that key candidate K1 takes the first state, p2 is the probability that key candidate K1 takes the second state, and p n To determine the probability that key candidate K1 takes the nth state, construct a Markov decision process: M = (S, A, P, R, γ). Where S is the set of key states, A is the set of update policies, P is the state transition probability matrix, R is the state reward function, and γ is the discount factor; Step S2.2: Based on the Markov decision process constructed in step S2.1, calculate the state transition probabilities: Where P(s′|s,a) is the probability that the current state s transitions to the next state s′ from policy a, s is the current state, s′ is the next state, a is the key update policy adopted, and δ i p is the adjustment factor for the i-th state. n Let K1 be the probability that the key candidate K1 takes the nth state. To update the probability distribution p of the key K Perform a summation calculation; Optimize strategy A to maximize the cumulative reward function: Where J(A) is the cumulative reward function, γ is the discount factor, and R(s) is the cumulative reward function. t ,a t To take strategy a at time step t t And it is in state s t The reward value obtained at that time To perform cumulative calculations over all time steps t; Step S2.3: After the update strategy A0 obtained from the optimization in step S2.2 is determined, based on the key update trigger condition Θ, it is determined whether to perform a key update, and the trigger condition is defined as follows: Θ = f(D,H,η), Where D is the sum of the mismatch indexes between key candidate K1 and reference sequence T at each position, H is the entropy value of key candidate, and η is the attack detection feedback signal; If Θ exceeds the set threshold, perform a key update and replace the old key K0 with the new key K1.

4. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S3, after the key update strategy A0 obtained in step S2 is determined, the following further steps are included: Step S3.1: Based on the new key K1 and update policy A0 determined in step S2, collect network traffic data: X = {x1, x2, ..., x...} m }, where x m For the data packets collected in the m-th time slice, where m is the total number of time slices collected, define a network traffic feature vector: F = g(X, K1), Where F is the network traffic feature vector, and g(·) is the key-weighted feature extraction function to generate traffic topology data for subsequent attack detection; Step S3.2: Construct a topology data analysis model based on the traffic feature vector F calculated in step S3.

1. G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, W is the edge weight matrix, and G is the network traffic topology data analysis model. Calculate the topological persistent homology feature: P = h(G), where P is the topological persistent homology feature and h(·) is the topological feature extraction function to identify potential anomalous patterns; Step S3.3: Based on the topological features P extracted in step S3.2, define the attack detection judgment function: θ = ψ(P,K1,A0), Where θ is the attack detection judgment function, ψ(·) is the detection function that comprehensively considers topological features, key and update strategy, and an attack judgment threshold θ0 is set. If θ≥θ0, an alarm signal is triggered and attack information η is fed back to step S2 to adjust the key update strategy.

5. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S4, after determining the network attack characteristics and triggering the alarm signal in step S3, the following further steps are included: Step S4.1, based on the topology data analysis model constructed in step S3: G = (V, E, W), where V is the set of network nodes, E is the set of connections between nodes, G is the network traffic topology data analysis model, and W is the edge weight matrix, W = [w ij ], where w ij For node v i With v j The strength of the connection between them; Define adjacency matrix A: A = [a ij ], where a ij For node v i With v j The connection between them, if a direct connection exists: a ij =w ij Otherwise a ij =0, v i v j Let i and j be the network nodes in the topology model; Calculate the graph Laplacian matrix: L = DA, where D is the degree matrix and L is the graph Laplacian matrix, which is used to represent the global information of the network topology and serves as the input for subsequent graph convolution. Step S4.2: Based on the graph Laplacian matrix L calculated in step S4.1, use the graph convolutional network to train the system topology model: M. t =f(L,X), Where X is the node feature matrix, f(·) is the graph convolution operation, and the node embedding is calculated as follows: Z = σ(LXW), where W is the training parameter matrix and σ(·) is the activation function to learn the module connection method in the topology; Step S4.3, based on the system topology model M obtained in step S4.2... t Define the topology update strategy: Ω = argmaxg(Z,P), where g(·) is the objective function based on node embedding and topology feature optimization. The system topology is adjusted according to Ω, and the topology change information is fed back to step S3 to optimize the subsequent network attack detection and key update strategy.

6. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S5, the graph Laplacian matrix L calculated in step S4 and the training system topology model M are... t The next step includes: Step S5.1, based on the topology model M obtained in step S4... t Collect data on changes in device connections: Y = {y1, y2, ..., y} k }, where y k Define a new adjacency matrix for the device connection information recorded in the k-th time slice: A′=[a′ ij ], where a′ ij For node v i With v j A new connection relationship exists between them, if a direct connection exists: a′ ij =w′ ij Otherwise: a′ ij =0, calculate the updated graph Laplacian matrix: L′=D′-A′, where D′ is the updated degree matrix; Step S5.2: Based on the updated graph Laplacian matrix L′ calculated in step S5.1, a new topology model, M, is calculated using a graph convolutional network. t ′=f(L′,X′), where X′ is the updated node feature matrix, f(·) is the graph convolution operation, and the new node embedding is calculated as: Z′=σ(L′X′W′), where W′ is the new training parameter matrix and σ(·) is the activation function to optimize the system topology; Step S5.3: Based on the new topology model M calculated in step S5.2 t Define the topology optimization objective: Ω′=argmax g(Z′,ξ), Wherein, g(·) is an optimization function calculated based on node embedding and topology change ξ, which readjusts the system topology according to Ω′ and feeds back the new topology adjustment information to step S4 to further optimize subsequent network analysis and attack detection.

7. The industrial-grade control cabinet safety control method according to claim 1, characterized in that, In step S6, after calculating the new topology in step S5 and adjusting the system topology, the following further steps are included: Step S6.1: Based on the updated graph Laplacian matrix L′ and optimized topology obtained in step S5, collect the illumination data of the display interface: I = {i1, i2, ..., i} m }, where i m Define the illumination model for the illumination data of the m-th time slice: M I =φ(I,L′), where φ(·) is a mapping function based on lighting data and topology, used to calculate the impact of lighting on the interface display; Step S6.2, based on the lighting model M calculated in step S6.1 I Construct the illumination effect matrix: G = [g ij ], where g ij For interface element e i and e j The degree of influence of lighting between elements is defined by the differentiable rendering calculation formula: C = σ(GX) I W I ), Among them, X I W is the lighting feature matrix of the interface elements. I The interface layout adjustment scheme is calculated using the training parameter matrix and σ(·) as the activation function. Step S6.3: Based on the interface layout adjustment scheme C calculated in step S6.2, define the final interface update strategy: Ω I =argmaxf(C,Z′), Where f(·) is the optimization objective function, which adjusts the interface based on the calculated interface lighting information and topology embedding Z′, and feeds back the interface adjustment information to step S5 to optimize the system topology and interactive experience.

8. A terminal device, characterized in that, It includes a processor, a memory, and a communication interface for performing the industrial-grade control cabinet safety control method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The device stores a computer program that, when executed, causes the terminal device to implement the industrial-grade control cabinet safety control method as described in any one of claims 1 to 7.

Images