A method and system for automated construction of business topology in industrial environments

Abstract

This invention discloses an automated business topology construction method and system based on industrial environments, belonging to the field of industrial business topology automation. It utilizes neural networks to analyze the logical relationships between devices in real time and dynamically adjust node priorities. Combined with a topology optimization mechanism using genetic algorithms / particle swarm optimization, it reduces the false alarm rate in scenarios such as device firmware upgrades and configuration changes. An event-triggered automatic topology update mechanism eliminates the need for manual maintenance of fingerprint and rule bases. By monitoring network status in real time and triggering adaptive reconstruction, it achieves automated discovery and construction of industrial business topologies. It can quickly generate complete and accurate topology maps without manual intervention. During topology reconstruction, configuration backup and gradual switching mechanisms are implemented to ensure business continuity.

Description

Technical Field

[0001] This invention belongs to the field of industrial business topology automation, specifically relating to a business topology automation construction method and system based on an industrial environment. Background Technology

[0002] Currently, industrial control systems have become an important component of critical infrastructure. In industrial control systems, equipment identification and behavior monitoring typically rely on the following technical means: Static fingerprint database comparison: Asset identification is performed by using a static feature database such as pre-set device MAC address and firmware version, which cannot adapt to dynamic changes in device behavior; Threshold alarm mechanism: Setting fixed thresholds can easily lead to false alarms or missed alarms; Single-dimensional monitoring: Independent analysis based on network traffic or device logs is difficult to fully reflect the device status; Rule engine driven: Relying on manually written rule bases to match known attack patterns, it is difficult to identify unknown threats and has high maintenance costs.

[0003] The above methods suffer from poor dynamic adaptability when facing complex and ever-changing industrial environments: they cannot automatically identify scenarios such as firmware upgrades and configuration changes (false alarm rate as high as 35%); they have a single feature dimension: they ignore the temporal characteristics of device behavior and their relationship with the process context (false alarm rate exceeds 40%); they have high maintenance costs: they require manual updates to the fingerprint database and detection rules regularly (averaging 8 person-hours per week); and they have serious detection lag: there is a 12-48 hour window between feature changes and rule effectiveness. Summary of the Invention

[0004] To address the shortcomings of the existing technologies, this application provides a method and system for automated construction of business topology in an industrial environment.

[0005] Firstly, this application proposes an automated method for constructing business topologies in an industrial environment, including the following steps:

[0006] Collect sensor data, control system data, and network communication data from industrial equipment, fuse and process the collected data, and build a real-time database of equipment nodes;

[0007] Based on the neural network, the logical relationships between devices in the real-time database of device nodes are identified, and the node priorities are dynamically sorted. Each device is a node, and each node has a corresponding weight value. The priorities are dynamically sorted according to the size of the weight values ​​to obtain the sorting result.

[0008] A first network topology is generated by combining the sorting results with a genetic algorithm or particle swarm optimization algorithm. Based on changes in service type and network status, a topology reconstruction strategy is dynamically invoked to transform the first network topology into a second network topology, including a ring topology and a star topology, thereby achieving adaptive optimization of the topology.

[0009] Choose the first or second network topology based on actual network requirements, and construct it into an industrial business topology graph using graph theory algorithms.

[0010] By combining periodic polling and event triggering, the system monitors the operating status of devices and network connectivity in real time. When a new device is detected to be accessing the network, the system automatically triggers a topology update process, collects the data of the new device, and adds it to the industrial business topology map to achieve a topology update.

[0011] In some embodiments, a first network topology is generated using a genetic algorithm combined with the sorting results, including:

[0012] Step A1: Randomly generate N binary matrices according to the encoding method to form the initial population. Each binary matrix represents an industrial network topology.

[0013] Step A2: Calculate the fitness value of each individual in the population according to the designed fitness function, where each individual represents a possible topology.

[0014] Step A3: Select individuals with high fitness from the current population to enter the next generation using roulette wheel selection or tournament selection.

[0015] Step A4: Perform crossover operations on the selected individuals according to the set crossover probabilities;

[0016] Step A5: Based on the mutation probability, mutate the individual's genes;

[0017] Step A6: Repeat steps A2-A5 until the set number of iterations is reached or the stopping condition is met. At this point, the topology corresponding to the individual with the highest fitness in the population is the first network topology.

[0018] In some embodiments, the encoding method of the genetic algorithm includes: representing the industrial network topology in binary encoding form. Assuming there are n nodes in the industrial network, an n×n binary matrix T can be constructed, where the elements T in the matrix... ij When T ij When T = 1, it indicates that there is a connection between node i and node j; when T ij When =0, it means there is no connection between node i and node j.

[0019] In some embodiments, a first network topology is generated by combining the particle swarm optimization algorithm with the ranking results, including:

[0020] Step B1: Randomly generate M n×n real number matrices as the initial positions of the particles, and simultaneously initialize the particle velocity matrix V, with velocity matrix elements Vij The range of values ​​for is determined based on the actual situation, and is generally [−1, 1].

[0021] Step B2: Calculate the fitness value of each particle according to the designed fitness function;

[0022] Step B3: Compare the fitness value of each particle's current position with the fitness value of its own historical best position. If the current position is better, update the individual's best position. Compare the fitness values ​​of all particles, find the particle with the highest fitness, and set its position as the global best position.

[0023] Step B4: Update the particle's velocity and position according to the following formula:

[0024] V ij (t+1)=ωV ij (t)+c1r1 ij (t)(P ij (t)−X ij (t))+c2r2 ij (t)(G ij (t)−X ij (t))X ij (t+1)=X ij (t)+V ij (t+1);

[0025] Where t represents the current iteration number, r1 ij and r2 ij P is a random number in the interval [0,1]. ij G represents the individual historical best position of particle i. ij This represents the global optimal position. The updated position is constrained to ensure that the matrix elements are within the range [0,1].

[0026] Step B5: Repeat steps B2-B4 until the maximum number of iterations is reached or the stopping condition is met. At this point, the topology corresponding to the globally optimal position is the first network topology.

[0027] In some embodiments, the particle representation of the particle swarm optimization algorithm includes: representing each particle as an n×n real matrix X, where n is the number of industrial network nodes, and the matrix elements X... ij This represents the probability or weight of a connection between node i and node j, with a value range of [0,1]. ij When X is close to 1, it indicates a higher probability of a connection between node i and node j; when X... ij A value close to 0 indicates a lower probability of connection.

[0028] In some embodiments, the fitness function design of the genetic algorithm or particle swarm optimization algorithm includes: Let R represent the network reliability index, T represent the real-time index, S represent the scalability index, and C represent the cost-effectiveness index. The value range of each index is normalized to be within the interval [0,1]. Then the fitness function formula is: Fitness=w1*R−w2*T−w3*S+w4*C where w1, w2, w3, and w4 are weight coefficients, and w1+w2+w3+w4=1. The weight coefficients need to be adjusted according to the needs of the actual industrial network.

[0029] In some embodiments, the dynamic invocation of topology reconstruction strategy based on service type and network status changes includes: the network status-based conversion is as follows: real-time monitoring of the status parameters of the industrial network; when the bandwidth utilization of a certain link in the network continuously exceeds the threshold and the latency and packet loss rate increase significantly, it is determined that the current topology cannot meet the network requirements, and the first network topology is converted to the corresponding topology according to the preset conversion strategy.

[0030] During the conversion process, the current network configuration and data are backed up first, and then the connection relationships of network devices are gradually adjusted. At the same time, network tests are conducted to ensure the continuity and stability of network services during the conversion process.

[0031] In some embodiments, the topology transformation of the first network topology to obtain a second network topology including a ring topology and a star topology, thereby achieving adaptive optimization of the topology, includes:

[0032] The generation of the circular topology is as follows: In the binary encoding of the genetic algorithm, a continuous sequence of nodes is required, so that these nodes are connected in sequence to form a ring, and each node is only connected to the adjacent nodes on the ring. In the particle swarm optimization algorithm, the initial position and update rules of the particles are set so that the connection between the nodes tends to form a circular structure. When updating the particle position, the probability of increasing the connection weight of the adjacent nodes is increased.

[0033] The star topology is generated as follows: In the binary matrix encoding of the genetic algorithm, a central node is specified, and all other nodes are connected only to this central node. In the matrix, the connection element between the central node and other nodes is 1, and the other elements are 0. In the particle swarm optimization algorithm, when initializing particles, the connection weights of most nodes are concentrated on a specific node to form the prototype of the star structure, and the stability of this connection mode is maintained in subsequent iterations.

[0034] The tree topology is defined by hierarchical encoding to define the connection relationship between parent nodes and child nodes;

[0035] The mesh topology is characterized by relaxing connection constraints, allowing more nodes to establish connections, and using an index in the fitness function to guide the formation of the mesh structure that needs to be transformed.

[0036] Secondly, this application proposes an automated business topology construction system based on an industrial environment, including a data acquisition module, a node configuration module, an industrial network topology analysis module, an industrial business topology graph construction module, and an automatic response module.

[0037] The data acquisition module is used to collect sensor data, control system data and network communication data of industrial equipment, perform fusion processing on the collected data, and build a real-time database of equipment nodes.

[0038] The node configuration module is used to identify the logical relationships between devices in the real-time database of device nodes based on a neural network, and dynamically sort the node priorities. Each device is a node, and each node has a corresponding weight value. The priorities are dynamically sorted according to the size of the weight values ​​to obtain the sorting result.

[0039] The industrial network topology analysis module is used to generate a first network topology by combining the sorting results with a genetic algorithm or particle swarm optimization algorithm. Based on changes in service type and network status, it dynamically calls a topology reconstruction strategy to transform the first network topology into a second network topology, including a ring topology and a star topology, thereby achieving adaptive optimization of the topology.

[0040] The industrial business topology graph construction module is used to select the first or second network topology structure according to actual network requirements, and combine graph theory algorithms to construct it into an industrial business topology graph.

[0041] The automatic response module is used to monitor the operating status of the device and the network connection status in real time through a combination of periodic polling and event triggering. When a new device is detected to be accessing the network, the topology update process is automatically triggered to collect the data of the new device and add it to the industrial business topology map to realize the topology update.

[0042] Thirdly, this application proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0043] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described above.

[0044] The beneficial effects of this invention are:

[0045] Based on real-time analysis of logical relationships between devices using neural networks and dynamic adjustment of node priorities, combined with a topology optimization mechanism using genetic algorithms / particle swarm optimization algorithms, the system reduces false alarm rates in scenarios such as device firmware upgrades and configuration changes. An event-triggered automatic topology update mechanism eliminates the need for manual maintenance of fingerprint and rule bases. By monitoring network status in real time and triggering adaptive reconstruction, the system achieves automated discovery and construction of industrial business topologies. Complete and accurate topology maps can be generated quickly without manual intervention. During the topology reconstruction process, configuration backup and gradual switching mechanisms are implemented to ensure business continuity. Attached Figure Description

[0046] Figure 1 This is the overall flowchart of the present invention.

[0047] Figure 2 This is a system principle block diagram of the present invention. Detailed Implementation

[0048] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein; rather, these embodiments are provided so that a more thorough understanding of the invention can be achieved and that the full scope of the invention can be conveyed to those skilled in the art.

[0049] Firstly, this application proposes an automated method for constructing business topologies based on industrial environments, such as... Figure 1 As shown, it includes the following steps:

[0050] S100: Collects sensor data, control system data and network communication data from industrial equipment, performs fusion processing on the collected data, and builds a real-time database of equipment nodes;

[0051] The system continuously captures operational status data using various sensors (temperature, pressure, position) installed on industrial equipment (CNC machine tools, robots, conveyor belts, etc.); extracts equipment information (equipment model, performance parameters, and work tasks) from the control system using industrial communication protocols (such as OPC UA, Modbus TCP, and Profinet); automatically identifies network connections between devices and obtains IP addresses, open port numbers, and communication protocols using network detection technologies (such as active Ping, port scanning, and ARP snooping); and fuses multi-source data to build a real-time database of equipment nodes for subsequent topology modeling.

[0052] S200: Based on the neural network, identify the logical relationship between devices in the real-time database of the device nodes, and dynamically sort the node priorities. Each device is a node, and each node has a corresponding weight value. The priorities are dynamically sorted according to the size of the weight values ​​to obtain the sorting result.

[0053] Based on the real-time database of device nodes, each device is abstracted as a graph node. A graph structure is constructed based on the collected multi-dimensional features (device type, communication frequency, data traffic, and criticality level). Graph neural networks (GCN) or long short-term memory networks (LSTM) are used to analyze the collaborative dependencies between devices, determine the communication strength between devices, and obtain the ranking results. Alternatively, after the sensors collect data and upload it to the platform, the platform is used to dynamically associate the pre-configured dataset with devices. In this case, each graph node has a corresponding weight value, and the nodes are dynamically prioritized according to the size of the weight value. The communication strength between devices is determined based on the priority, and the ranking results are obtained.

[0054] S300: A first network topology is generated by combining the sorting results with a genetic algorithm or particle swarm optimization algorithm. Based on changes in service type and network status, a topology reconstruction strategy is dynamically invoked to transform the first network topology into a second network topology, including a ring topology and a star topology, thereby achieving adaptive optimization of the topology.

[0055] In some embodiments, a first network topology is generated using a genetic algorithm combined with the sorting results, including:

[0056] Step A1: Randomly generate N binary matrices according to the encoding method to form the initial population. Each binary matrix represents an industrial network topology.

[0057] Step A2: Calculate the fitness value of each individual in the population according to the designed fitness function, where each individual represents a possible topology.

[0058] Step A3: Select individuals with high fitness from the current population to enter the next generation using roulette wheel selection or tournament selection.

[0059] Step A4: Perform crossover operations on the selected individuals according to the set crossover probabilities;

[0060] Step A5: Based on the mutation probability, mutate the individual's genes;

[0061] Step A6: Repeat steps A2-A5 until the set number of iterations is reached or the stopping condition is met. At this point, the topology corresponding to the individual with the highest fitness in the population is the first network topology.

[0062] In some embodiments, the encoding method of the genetic algorithm includes: representing the industrial network topology in binary encoding form. Assuming there are n nodes in the industrial network, an n×n binary matrix T can be constructed, where the elements T in the matrix... ij When T ij When T = 1, it indicates that there is a connection between node i and node j; when T ij When =0, it means there is no connection between node i and node j.

[0063] The parameters of the genetic algorithm are set as follows:

[0064] 1. Population size: Typically set to 50-200. A larger population size can increase population diversity and prevent the algorithm from getting trapped in local optima too early, but it will increase computational cost and time; a smaller population size is computationally efficient, but may not be able to fully search the solution space. In industrial network topology optimization scenarios, if the network is large and complex, a larger population size, such as 150, can be chosen; if the network is small, a population size of around 80 can be chosen.

[0065] 2. Number of iterations: Generally set between 100 and 500. Too few iterations may prevent the algorithm from finding a satisfactory solution; too many iterations will waste computational resources and time. For simple industrial network topology optimization, 200 iterations may be sufficient to find a good solution; while for complex networks, 300-400 iterations may be required. Crossover probability: Typically ranges from 0.6 to 0.9. A higher crossover probability can speed up the algorithm's search and promote the spread and combination of superior genes, but an excessively high crossover probability may destroy superior individuals; a lower crossover probability will slow down the algorithm's convergence speed. In practical applications, the crossover probability can be initially set to 0.8 and adjusted according to the algorithm's performance.

[0066] 3. Mutation probability: Generally between 0.001 and 0.01. Mutation operations can increase the diversity of the population and prevent the algorithm from getting trapped in local optima. Too high a mutation probability will make the algorithm resemble random search, while too low a probability will fail to effectively avoid local optima. You can initially set the mutation probability to 0.005 and observe the optimization effect of the algorithm.

[0067] In some embodiments, a first network topology is generated by combining the particle swarm optimization algorithm with the ranking results, including:

[0068] Step B1: Randomly generate M n×n real number matrices as the initial positions of the particles, and simultaneously initialize the particle velocity matrix V, with velocity matrix elements V ij The range of values ​​for is determined based on the actual situation, and is generally [−1, 1].

[0069] Step B2: Calculate the fitness value of each particle according to the designed fitness function;

[0070] Step B3: Compare the fitness value of each particle's current position with the fitness value of its own historical best position. If the current position is better, update the individual's best position. Compare the fitness values ​​of all particles, find the particle with the highest fitness, and set its position as the global best position.

[0071] Step B4: Update the particle's velocity and position according to the following formula:

[0072] V ij (t+1)=ωV ij (t)+c1r1 ij (t)(P ij (t)−X ij (t))+c2r2 ij (t)(G ij (t)−X ij (t))X ij (t+1)=X ij (t)+V ij (t+1);

[0073] Where t represents the current iteration number, r1 ij and r2 ij P is a random number in the interval [0,1]. ij G represents the individual historical best position of particle i. ij This represents the global optimal position. The updated position is constrained to ensure that the matrix elements are within the range [0,1].

[0074] Step B5: Repeat steps B2-B4 until the maximum number of iterations is reached or the stopping condition is met. At this point, the topology corresponding to the globally optimal position is the first network topology.

[0075] In some embodiments, the particle representation of the particle swarm optimization algorithm includes: representing each particle as an n×n real matrix X, where n is the number of industrial network nodes, and the matrix elements X... ij This represents the probability or weight of a connection between node i and node j, with a value range of [0,1]. ij When X is close to 1, it indicates a higher probability of a connection between node i and node j; when X... ij A value close to 0 indicates a lower probability of connection.

[0076] The parameters of the particle swarm optimization algorithm are set as follows:

[0077] 1. Particle swarm size: Typically set between 30 and 100. Similar to genetic algorithms, a larger swarm size results in stronger search capabilities, but also higher computational complexity. For medium-sized industrial networks, 50 particles can be selected.

[0078] 2. Maximum number of iterations: Generally 100-300. The number of iterations is selected based on the complexity of the network and available computational resources; for complex networks, the number of iterations can be increased appropriately. 3. Learning factors c1 and c2: Usually between [1, 2]. c1 and c2 represent the degree to which a particle learns from its own historical best position and the group's historical best position, respectively. Generally, c1 = c2 = 1.5, allowing the particle to fully utilize its own experience while also drawing on the group's experience. Inertia weight ω: Values ​​range from [0.4, 0.9]. A larger inertia weight is beneficial for global search, while a smaller inertia weight is beneficial for local search. In the early stages of the algorithm, a larger ω value, such as 0.8, can be set to accelerate the global search; in the later stages, the ω value can be gradually decreased, such as to 0.5, for more refined local search.

[0079] In some embodiments, the fitness function design of the genetic algorithm or particle swarm optimization algorithm includes: Let R represent the network reliability index, T represent the real-time index, S represent the scalability index, and C represent the cost-effectiveness index. The value range of each index is normalized to be within the interval [0,1]. Then the fitness function formula is: Fitness=w1*R−w2*T−w3*S+w4*C where w1, w2, w3, and w4 are weight coefficients, and w1+w2+w3+w4=1. The weight coefficients need to be adjusted according to the needs of the actual industrial network.

[0080] In some embodiments, the dynamic invocation of topology reconstruction strategy based on service type and network status changes includes: the network status-based conversion is as follows: real-time monitoring of the status parameters of the industrial network; when the bandwidth utilization of a certain link in the network continuously exceeds the threshold and the latency and packet loss rate increase significantly, it is determined that the current topology cannot meet the network requirements, and the first network topology is converted to the corresponding topology according to the preset conversion strategy.

[0081] During the conversion process, the current network configuration and data are backed up first, and then the connection relationships of network devices are gradually adjusted. At the same time, network tests are conducted to ensure the continuity and stability of network services during the conversion process.

[0082] In some embodiments, the topology transformation of the first network topology to obtain a second network topology including a ring topology and a star topology, thereby achieving adaptive optimization of the topology, includes:

[0083] The generation of the circular topology is as follows: In the binary encoding of the genetic algorithm, a continuous sequence of nodes is required, so that these nodes are connected in sequence to form a ring, and each node is only connected to the adjacent nodes on the ring. In the particle swarm optimization algorithm, the initial position and update rules of the particles are set so that the connection between the nodes tends to form a circular structure. When updating the particle position, the probability of increasing the connection weight of the adjacent nodes is increased.

[0084] The star topology is generated as follows: In the binary matrix encoding of the genetic algorithm, a central node is specified, and all other nodes are connected only to this central node. In the matrix, the connection element between the central node and other nodes is 1, and the other elements are 0. In the particle swarm optimization algorithm, when initializing particles, the connection weights of most nodes are concentrated on a specific node to form the prototype of the star structure, and the stability of this connection mode is maintained in subsequent iterations.

[0085] The tree topology is defined by hierarchical encoding to define the connection relationship between parent nodes and child nodes;

[0086] The mesh topology is characterized by relaxing connection constraints, allowing more nodes to establish connections, and using an index in the fitness function to guide the formation of the mesh structure that needs to be transformed.

[0087] Generation of polymorphic topologies:

[0088] 1. Circular Topology Generation: In genetic algorithms or particle swarm optimization algorithms, circular topologies are generated by constraining the encoding or particle representation. For example, in the binary encoding of a genetic algorithm, a continuous sequence of nodes is required, such that these nodes are connected sequentially to form a loop, and each node is only connected to its adjacent nodes on the loop (i.e., the corresponding element in the matrix is ​​1, and other non-adjacent elements are 0). In particle swarm optimization algorithms, the initial positions and update rules of the particles are set to make the connections between nodes tend to form a circular structure, such as increasing the probability of increasing the connection weight of adjacent nodes when updating particle positions.

[0089] 2. Star Topology Generation: Constraints are also added to the algorithm. In the binary matrix encoding of the genetic algorithm, a central node is specified, ensuring that all other nodes are connected only to this central node (the element connecting the central node to other nodes in the matrix is ​​1, and other elements are 0). In the particle swarm optimization algorithm, during particle initialization, the connection weights of most nodes are concentrated on a specific node, forming a preliminary star topology, and this connection pattern is maintained stable in subsequent iterations.

[0090] 3. Generation of other topologies: For tree topologies, mesh topologies, and other structures, more complex constraints and coding rules can be designed to achieve this.

[0091] For example, tree topology can define the connection relationship between parent and child nodes through hierarchical coding; mesh topology relaxes connection constraints, allowing more nodes to establish connections, but can guide the formation of a reasonable mesh structure through indicators such as reliability in the fitness function.

[0092] Transformation between topologies:

[0093] Network state-based transitions: Real-time monitoring of industrial network state parameters, such as bandwidth utilization, latency, and packet loss rate. When the bandwidth utilization of a network link consistently exceeds a threshold (e.g., 80%), and latency and packet loss rate increase significantly, it is determined that the current topology cannot meet network requirements. At this point, according to a preset transition strategy, the current topology is converted to a more suitable structure. For example, if the current topology is a star topology, and network performance degrades due to excessive load on the central node, it can be converted to a ring topology or a distributed mesh topology, increasing redundant paths to distribute traffic and improve network reliability and performance.

[0094] Transformation based on actual needs: When the actual needs of an industrial network change, a topology transformation is performed. For example, adding a new production line to a factory places higher demands on the network's scalability and real-time performance. If the current simple bus topology cannot meet the access and data transmission requirements of the new devices, a tree topology or hybrid topology can be generated based on algorithms to adapt to the new requirements. During the transformation process, the current network configuration and data are first backed up, then the connection relationships of network devices are gradually adjusted, and network testing is conducted simultaneously to ensure the continuity and stability of network services during the transformation.

[0095] S400: Select the first or second network topology based on actual network requirements, and construct it into an industrial business topology graph using graph theory algorithms;

[0096] Specifically, the system automatically matches the corresponding topology constraints based on the type of service carried by the current industrial network (such as real-time control, data acquisition, and remote debugging). For example, in real-time control scenarios, the system prioritizes low latency and high reliability, adopting a star topology; while in remote debugging scenarios with higher fault tolerance requirements, the system recommends a mesh topology.

[0097] Use a graph theory algorithm (Minimum Spanning Tree, MST) to generate a topology that meets specific business requirements. The goal of the MST algorithm is:

[0098]

[0099] in, The optimal MST structure is determined by Prim's algorithm or Kruskal's algorithm to calculate the communication cost between nodes u and v, and then presented to the operations and maintenance personnel in a graphical manner.

[0100] In addition, based on historical optimization experience and the current network status, topology switching strategies are provided. For example, when the system detects that the packet loss rate of a certain link exceeds the threshold of 10, it switches from a star topology to a partial mesh topology to enhance fault tolerance.

[0101] S500: By combining periodic polling and event triggering, it monitors the operating status of devices and network connectivity in real time. When a new device is detected to be accessing the network, it automatically triggers the topology update process, collects the data of the new device and adds it to the industrial business topology map to achieve topology update.

[0102] I. Polling Monitoring Mechanism Design

[0103] 1. Tiered polling cycle strategy

[0104] The three-level polling system is divided according to equipment importance and business sensitivity, as shown in Table 1:

[0105] Table 1

[0106] Equipment type Polling cycle Application scenarios Data sampling frequency adjustment rules Core production equipment 100ms-1s CNC machine tools, PLC controllers Automatically shortened to 50ms in case of runtime abnormality Critical network equipment 5s-30s Core switches and firewalls The peak network period has increased to 2 seconds. Auxiliary sensors 1min-5min Ambient temperature and humidity sensor Non-working hours extended to 10 minutes

[0107] Dynamic adjustment formula: T n =T0×(1+k×C th *C load -C th Where: T n : Adjusted cycle, T0: Initial cycle, k: Adjustment coefficient (k=0.5 for core equipment, k=1 for general equipment), C load Current system load, C th Load threshold;

[0108] 2. Polling Protocol and Data Acquisition Method

[0109] Industrial equipment: OPC UA protocol: 1s cycle, collects real-time parameters such as motor speed and temperature; Modbus RTU: 500ms cycle, polls PLC register status; Network equipment: SNMP v3: 30s cycle, obtains interface traffic (IF-MIB) and CPU utilization; ICMP ping: 2s cycle, monitors network connectivity (packet loss rate, latency).

[0110] Data caching strategy: Use a circular buffer to store data from the most recent 100 polling cycles.

[0111] II. Event Trigger Monitoring Mechanism

[0112] 1. Equipment status trigger condition design:

[0113] Threshold-type triggering (taking CNC machine tools as an example) is shown in Table 2:

[0114] Table 2

[0115] Monitoring parameters Warning threshold Fault trigger threshold Triggered Action Spindle temperature ≥60°C ≥75℃C Log out + activate cooling system vibration amplitude 20.5mm / s 21.2mm / s Trigger shutdown procedure + generate fault work order Current fluctuation Exceeding the rated value by ±15% Exceeding the rated value by ±30% Power off + send emergency alarm

[0116] State transition trigger:

[0117] Define a state machine model (e.g., running → standby → fault), which is triggered when the following state changes occur:

[0118] Abnormal transition: Operation → Failure (no normal shutdown procedure)

[0119] Timeout transition: Standby → Running timeout (exceeding the expected startup time by 30 seconds)

[0120] 2. Network connection triggering conditions

[0121] Link layer triggers: Sudden packet loss rate: The packet loss rate of three consecutive ICMP packets increases from <1% to >10%; Sudden bandwidth drop: The interface bandwidth utilization suddenly drops from 70% to 20% (possible link interruption);

[0122] Application layer triggers: Service response timeout: HTTP request does not respond for more than 500ms (triggering reconnection mechanism); Abnormal session termination: TCP connection RST packet frequency > 5 times / minute).

[0123] Secondly, this application proposes an automated business topology construction system based on an industrial environment, such as... Figure 2 As shown, it includes a data acquisition module, a node configuration module, an industrial network topology analysis module, an industrial business topology diagram construction module, and an automatic response module;

[0124] The data acquisition module is used to collect sensor data, control system data and network communication data of industrial equipment, perform fusion processing on the collected data, and build a real-time database of equipment nodes.

[0125] The node configuration module is used to identify the logical relationships between devices in the real-time database of device nodes based on a neural network, and dynamically sort the node priorities. Each device is a node, and each node has a corresponding weight value. The priorities are dynamically sorted according to the size of the weight values ​​to obtain the sorting result.

[0126] The industrial network topology analysis module is used to generate a first network topology by combining the sorting results with a genetic algorithm or particle swarm optimization algorithm. Based on changes in service type and network status, it dynamically calls a topology reconstruction strategy to transform the first network topology into a second network topology, including a ring topology and a star topology, thereby achieving adaptive optimization of the topology.

[0127] The industrial business topology graph construction module is used to select the first or second network topology structure according to actual network requirements, and combine graph theory algorithms to construct it into an industrial business topology graph.

[0128] The automatic response module is used to monitor the operating status of the device and the network connection status in real time through a combination of periodic polling and event triggering. When a new device is detected to be accessing the network, the topology update process is automatically triggered to collect the data of the new device and add it to the industrial business topology map to realize the topology update.

[0129] Thirdly, this application proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0130] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described above.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0132] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0133] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0134] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus / computer devices and methods can be implemented in other ways. For example, the apparatus / computer device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of apparatus or units may be electrical, mechanical, or other forms.

[0135] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0136] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0137] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in a computer-readable medium may be appropriately added to or subtracted according to the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0138] The above are merely preferred embodiments of the present invention. It should be noted that any modifications and improvements made by those skilled in the art without departing from the present technical solution should also be considered to fall within the scope of protection claimed by the present solution.

Claims

1. A method for automatic construction of a service topology based on an industrial environment, characterized in that: The method comprises the following steps: Collecting sensor data, control system data and network communication data of industrial equipment, fusing the collected data to construct a device node real-time database; Based on neural network, the logical relationship between devices in the device node real-time database is identified, and the node priority is dynamically sorted, wherein each device is a node, each node has a corresponding weight value, and the priority is dynamically sorted according to the size of the weight value to obtain a sorting result; A first network topology structure is generated by using a genetic algorithm combined with the sorting result, or by using a particle swarm optimization algorithm combined with the sorting result, a topology reconstruction strategy is dynamically called according to the business type and network state change, the first network topology structure is converted to obtain a second network topology structure including a ring topology structure and a star topology structure, and adaptive optimization of the topology structure is realized; According to actual network requirements, the first or second network topology structure is selected, and an industrial business topology graph is constructed by combining a graph theory algorithm; By combining regular polling and event triggering, the running state and network connection of the equipment are monitored in real time, when a new device is detected to access the network, a topology update process is automatically triggered, data of the new device is collected and added to the industrial business topology graph, and topology update is realized.

2. The method of claim 1, wherein: The first network topology structure is generated by using a genetic algorithm combined with the sorting result, comprising: Step A1: N binary matrices are randomly generated according to an encoding mode to form an initial population, and each binary matrix represents an industrial network topology structure; Step A2: According to a designed fitness function, the fitness value of each individual in the population is calculated, wherein each individual represents a possible topology structure; Step A3: A roulette selection method or a tournament selection method is used to select individuals with higher fitness from the current population to enter the next generation population; Step A4: According to a set crossover probability, the selected individuals are subjected to a crossover operation; Step A5: According to a mutation probability, the genes of the individuals are subjected to a mutation; Step A6: Steps A2-A5 are repeated until a set iteration number is reached or a stop condition is met, at this time, the topology structure corresponding to the individual with the highest fitness in the population is the first network topology structure.

3. The method of claim 2, wherein: The coding mode of the genetic algorithm comprises: adopting a binary coding form to represent the industrial network topology structure; when there are n nodes in the industrial network, an n*n binary matrix T is constructed, and an element T ij in the matrix represents that there is a connection between a node i and a node j when T ij =1; and represents that there is no connection between the node i and the node j when T ij =0.

4. The method of claim 2, wherein: The fitness function design of the genetic algorithm or the particle swarm optimization algorithm comprises: R represents a network reliability index, T represents a real-time index, S represents a scalability index, and C represents a cost benefit index; the value range of each index is normalized to be in the interval [0, 1]; and the fitness function formula is: wherein w1, w2, w3 and w4 are weight coefficients, and w1+w2+w3+w4=1; the setting of the weight coefficients needs to be adjusted according to the demand focus of an actual industrial network.

5. The method of claim 4, wherein: The topology reconstruction strategy is dynamically called according to the business type and network state change, comprising: based on network state conversion: real-time monitoring of state parameters of the industrial network, when the bandwidth utilization of a link of the network continuously exceeds a threshold value and the delay and packet loss rate significantly increase, it is judged that the current topology structure cannot meet the network requirements, and according to a preset conversion strategy, the first network topology structure is converted into a corresponding topology structure; In the conversion process, the current network configuration and data are first backed up, then the connection relationship of the network equipment is gradually adjusted, and network testing is performed at the same time to ensure the continuity and stability of network services during the conversion process.

6. An industrial environment based business topology automated building system, characterized by: The method comprises a data acquisition module, a node configuration module, an industrial network topology structure analysis module, an industrial business topology graph construction module and an automatic response module; The data collection module is configured to collect sensor data, control system data and network communication data of the industrial equipment, perform fusion processing on the collected data, and construct an equipment node real-time database; The node configuration module is configured to identify logical relationships between the equipment in the equipment node real-time database based on a neural network, dynamically sort node priorities, wherein each equipment is a node, each node has a corresponding weight value, and the priorities are dynamically sorted according to the weight values to obtain a sorting result; The industrial network topology structure analysis module is configured to generate a first network topology structure by using a genetic algorithm in combination with the sorting result or by using a particle swarm optimization algorithm in combination with the sorting result, dynamically call a topology reconstruction strategy according to a business type and a network state change, perform topology structure conversion on the first network topology structure to obtain a second network topology structure including a ring topology structure and a star topology structure, and realize adaptive optimization of the topology structure; The industrial business topology graph construction module is configured to select the first or second network topology structure according to actual network requirements, and construct the selected network topology structure into an industrial business topology graph by using a graph theory algorithm. The automatic response module is configured to monitor a running state of the equipment and a network connection in real time by using a combination of regular polling and event triggering, automatically trigger a topology update process when a new equipment is detected to access the network, collect data of the new equipment and add the new equipment to the industrial business topology graph, and realize topology update.

7. A computer-readable storage medium, characterized in that: The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the method in any one of claims 1-5.

Images