Discrete manufacturing-oriented edge intelligent network adaptive deployment method and system

Abstract

The invention relates to a discrete manufacturing-oriented edge intelligent network adaptive deployment method and system, and the method aims at maximizing the network throughput, minimizing the transmission time delay and maximizing the network reliability, builds a QoS-driven edge intelligent network deployment optimization model, and ensures that all devices in a factory can meet the requirements of network throughput maximization, transmission time delay minimization and network reliability maximization. Especially, the network transmission requirements of key equipment are met; an improved K-means algorithm and a particle swarm optimization algorithm are adopted, and a dynamic clustering algorithm and an adaptive weight adjustment mechanism are combined, so that the data rate and the network reliability of the intelligent equipment are effectively improved, and meanwhile, the transmission delay and the system interference are reduced. The method can adapt to factories of different sizes and complex environments, network performance is optimized, edge intelligent deployment efficiency is remarkably improved, and a stable and reliable communication solution is provided for discrete manufacturing intelligent factories and intelligent manufacturing in the industrial 5.0 era.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial communication networks, and in particular relates to an edge intelligent network adaptive deployment method and system for discrete manufacturing. Background Art

[0002] With the development of industrial automation and intelligent manufacturing, factory environments are becoming increasingly complex, and the demand for wireless connections between devices and systems is increasing. The discrete manufacturing industry particularly requires a network system that can not only adapt to frequent changes in layout and equipment configuration, but also meet increasingly stringent quality of service requirements, including low latency, high throughput, and high reliability. However, traditional network deployment strategies often have difficulty in meeting these requirements, especially in a continuously changing manufacturing environment.

[0003] In the existing technology, network deployment usually adopts static design solutions, which often ignore the dynamic changes within the factory, such as frequent adjustments to the factory layout and rapid movement of equipment, resulting in problems such as unstable connections, increased data transmission delays, and inability to update configurations in a timely manner in actual operations. In addition, traditional optimization algorithms often fail to fully consider the dynamic changes of real-time data, making network adjustments inflexible and unable to effectively respond to the impact of equipment movement and task changes within the factory, which may lead to reduced production efficiency and potential economic losses. Summary of the invention

[0004] In order to solve the above problems, the present invention has developed an edge intelligent network adaptive deployment method and system for discrete manufacturing, which can not only adapt to the dynamic changes and uncertainties within the factory, but also ensure that the network performance requirements of all equipment, especially key equipment, are met. In addition, the present invention combines modern communication technology and intelligent optimization algorithms to not only improve the data processing capabilities and response speed of factory equipment, but also greatly reduce production interruptions or quality problems caused by network problems, thereby ensuring efficient communication in the production process and stability of factory operations.

[0005] The purpose of the present invention is to provide an edge intelligent network adaptive deployment method and system for discrete manufacturing, which aims to cope with the complex dynamic changes and uncertainties in discrete manufacturing plants, and ensure that network reliability and service quality meet the requirements under any circumstances. By constructing an edge intelligent network deployment optimization model driven by Quality of Service (QoS) under limited spectrum resources, the edge intelligent network deployment has stronger flexibility and adaptability; by jointly using the improved K-means clustering algorithm and the particle swarm optimization (PSO) algorithm, the edge intelligent network deployment strategy can quickly converge to a high-quality network configuration solution in a shorter time, improving network deployment efficiency and network stability.

[0006] The purpose of the present invention is achieved through the following technical solutions:

[0007] An edge intelligent network adaptive system for discrete manufacturing, the system includes multiple factory equipment, multiple heterogeneous cellular network base stations, multiple mobile edge computing servers (Mobile Edge Computing, MEC) and cloud servers. The heterogeneous cellular network base stations include macro base stations and micro base stations. The micro base stations are deployed in areas where the macro base station signal coverage is weak or difficult to cover, and are connected to the macro base stations through wireless fronthaul technology, thereby enhancing the quality and coverage of network signals; each factory equipment is connected to an edge intelligent network node that can provide a maximum signal-to-interference-to-noise ratio according to the actual communication environment, ensuring that the equipment always maintains the best network connection quality; each macro base station is connected to a mobile edge server, which is responsible for local data processing and computing tasks, reducing data transmission delays and improving edge computing efficiency, and ultimately achieving global data coordination and optimization management through cloud servers.

[0008] An edge intelligent network adaptive deployment method for discrete manufacturing, the steps are as follows:

[0009] (1) For the above system, with the goal of reducing system interference and maximizing network service quality, the adaptive deployment optimization problem of edge intelligent network nodes is constructed according to the current network conditions and the network requirements of factory equipment;

[0010] (2) The improved K-means clustering algorithm and PSO algorithm are used to solve the adaptive deployment optimization problem, complete the intelligent deployment and dynamic adjustment of all heterogeneous cellular network base stations and all mobile edge servers, and improve the network service quality and the utilization of computing resources.

[0011] Preferably, the specific steps of constructing the adaptive deployment optimization problem of edge intelligent network nodes in the above step (1) are as follows:

[0012] (1-1) Assuming that the system adopts frequency division multiplexing technology, the macro base station and the micro base station work on different frequency bands and allocate the required network resources to the factory equipment in a collaborative way. ij =1 means that the i-th factory equipment is associated with the j-th edge intelligent network node, otherwise they are not associated, i∈[1,D]; x ik =1 means the kth subcarrier is allocated to the ith factory equipment, otherwise it is not allocated, k∈[1,K];

[0013] (1-2) Construct a QoS model for discrete manufacturing edge intelligent network systems. Considering the presence of a large number of obstacles such as metal surfaces, large equipment, and buildings in the factory, as well as the impact of machine operation and production activity changes on the transmission quality of wireless signals, the wireless channel between factory equipment i and edge intelligent network node j is described using the WINNER II B3 channel model. Assume that the transmission power of factory equipment i is The carrier frequency is f ik , the distance between device i and its associated edge intelligent network node j is d ij , then the path loss model is expressed as:

[0014]

[0015] Among them, LOS and NLOS represent line-of-sight propagation and non-line-of-sight propagation respectively, and X σ It is represented as the loss correction factor caused by obstacles. Therefore, the signal power received by edge intelligent network node j is expressed as Affected by shadow fading, is a random variable that obeys a log-normal distribution, and the signal reception failure rate is defined as The probability of is:

[0016]

[0017] in, is the average signal power received by base station j, σ is the standard deviation of shadow fading, and Q(z) is the probability that the random variable x that obeys the standard normal distribution is greater than z, that is:

[0018]

[0019] Considering the co-channel interference generated by other devices in the factory that use the same carrier frequency as device i, as well as the impact of mechanical vibration, electromagnetic interference, and random noise generated by equipment operation on the data transmission of factory device i, the uplink SINR of the i-th factory device is expressed as:

[0020]

[0021] in, Indicates the co-channel interference to device i. Represents the thermal noise of factory equipment, P ijk (t) = A∑ l δ(t-lT) represents the noise generated by equipment operation, switching action or other periodic activities, where A is the pulse amplitude, δ(t) is the Dirac delta function, T is the pulse period, and lT is the specific time point when the lth pulse occurs;

[0022] Assume that m ij represents the data size transmitted from the i-th factory equipment to the j-th edge intelligent network node, b ij is the network bandwidth allocated to the i-th factory equipment, then the sum of the transmission delays of all factory equipment within the coverage of the j-th edge intelligent network node is expressed as:

[0023]

[0024] Assuming that the jth edge intelligent network node has sufficient network capacity, the sum of the data sizes of all factory equipment within its coverage range is expressed as:

[0025]

[0026] The signal transmission in the system adopts quadrature phase shift keying modulation to effectively utilize the limited communication bandwidth. If the m ij The probability of successful transmission of bits of data is the product of the probability of successful transmission of each bit or data unit. The probability that the data of all factory equipment within the coverage area of ​​the jth edge intelligent network node is successfully transmitted is:

[0027]

[0028] (1-3) Construct the adaptive deployment optimization problem of discrete manufacturing edge intelligent network. If the key indicators of system network service quality include communication reachable rate, system delay and transmission reliability, the service quality of edge intelligent network nodes is defined as follows:

[0029]

[0030] Where z represents the deployment location of the edge intelligent network node, ω1, ω2, ω3∈[0,1] are the weighted coefficients of each key indicator in the system, and ω1+ω2+ω3=1, T j is the sum of the actual data upload delays of all factory equipment within the coverage area of ​​the jth edge intelligent network node, T req The minimum delay that should be met for data transmission of factory equipment within its coverage area; R j is the transmission capacity that the jth edge intelligent network node can provide, R req The average network capacity required to upload data from factory equipment within the coverage area; PSR j PSR is the reliability of data transmission of factory equipment within the coverage area of ​​the jth edge intelligent network node. req The minimum reliability that must be met for successful data transmission of factory equipment within the coverage area;

[0031] Considering the uneven distribution of factory equipment and communication requirements, reasonable and efficient deployment of edge intelligent network nodes can provide factory equipment with low latency, high reliability, and high data rate service quality, and meet the constraints of maximum system capacity and minimum transmission delay. Therefore, the edge intelligent network node adaptive deployment optimization problem is expressed as:

[0032]

[0033] C7:ω1+ω2+ω3=1

[0034] in, It is a service quality redundancy factor designed to cope with business fluctuations such as increased network load. Constraint C1 means that the device will only be connected to one base station at the same time. Constraint C2 means that each device only transmits data on a specific subchannel, and data transmission through multiple subchannels is not allowed at the same time, so as to avoid interference and data packet conflicts between channels. Constraint C3 means that the sum of the bandwidth allocated to all devices cannot exceed the total bandwidth of the base station, and the bandwidth allocated to each device must be non-negative. Constraints C4~C6 respectively represent that the uplink transmission rate, delay and reliability of each factory device i connected to the edge intelligent network node j can be basically guaranteed, where R min , T max and PSR min They represent the minimum transmission rate, maximum allowable transmission delay and minimum reliability of factory equipment i respectively.

[0035] Preferably, the specific steps of the above step (2) are as follows:

[0036] (2-1) Identify potential areas of insufficient QoS

[0037] By monitoring the real-time data transmission of each device in the discrete manufacturing smart factory, the preset QoS threshold standard R min 、T max and PSR min , detect the devices whose QoS is not met, and then analyze the number and spatial distribution of these devices whose QoS is not met, and quickly identify the potential underserved areas. The set of devices whose QoS is not met is

[0038] (2-2) Use the improved K-means algorithm to dynamically initialize clustering of factory equipment that does not meet QoS requirements, and use the cluster center as the candidate deployment location of the edge intelligent network node;

[0039] Assume x j is the decision variable, j indexes all possible micro base station locations, x j=1 means that a base station is deployed at position j, otherwise it is not deployed. For each factory equipment i∈U, a ij =1 means that the device i whose QoS is not satisfied is covered by the micro base station at position j, and vice versa, if it is 0, the QoS of device i is still not satisfied. If the deployment cost of each micro base station is C j , then the objective function of dynamic clustering is expressed as:

[0040]

[0041] For the optimization problem P2, the improved K-means algorithm is used to solve it. The number of cluster centers is dynamically adjusted in real time according to the number of devices whose QoS is not met on site and their spatial distribution, until an edge intelligent network node deployment solution is found that can ensure the service quality requirements of all devices in the set U and maximize cost-effectiveness.

[0042] (2-3) Use the adaptive weight PSO algorithm to accurately adjust the location and configuration of heterogeneous cellular network base stations and MEC servers;

[0043] In the PSO algorithm, each particle has a position vector X and a velocity vector V. X represents the position of the particle in the solution space, and V determines the flight direction and velocity. Assume that V i t represents the velocity of the ith particle at time t, represents the position of the ith particle at time t, represents the best position searched historically by the i-th particle, g best Represents the best position searched in the history of the entire group. In each iteration, each particle will be based on its own historical best position. and the global best position g in the population best To update its own speed and position, that is:

[0044]

[0045]

[0046] Among them, r1 and r2 are random numbers, c1 and c2 represent the acceleration weights that push particles to the individual optimal position and the group optimal position respectively. In order to maintain the balance between convergence speed and search effect, c1 and c2 are not 0; ω is the inertia weight coefficient at time t, which controls the continuity of particle velocity update. A higher weight coefficient is conducive to global search, while a lower weight coefficient is conducive to local search. Fixed inertia weight cannot effectively balance exploration and utilization in complex multi-objective optimization problems, resulting in particles converging to the local optimal solution too early and difficult to adapt to the dynamically changing factory environment. In order to effectively balance global and local searches, the inertia weight is defined as:

[0047] ω=ω max -(ω max -ω min )·(t / T) p

[0048] Among them, ω max and ω min They represent the maximum and minimum values ​​of the inertia weight coefficient respectively, p is a constant, T is the total number of iterations of the algorithm, and a larger inertia weight is used in the early stage of the search to help the algorithm explore a wider solution space. In the later stage of the iteration, the inertia weight is gradually reduced to enhance the local search ability of the particles, thereby improving the accuracy of the optimal solution and the convergence of the algorithm.

[0049] The beneficial effects of the present invention include:

[0050] 1. Improve network coverage quality and stability: By adopting a combination of macro base stations and micro base stations and using wireless fronthaul technology to connect micro base stations with macro base stations, the network service quality in areas with weak or difficult-to-cover signal coverage in discrete manufacturing smart factories can be effectively enhanced, ensuring that factory equipment always maintains a stable network connection;

[0051] 2. Optimize network resource utilization efficiency: Use the improved K-means clustering algorithm and PSO algorithm to dynamically adjust the deployment of heterogeneous cellular network base stations and MEC servers, optimize resource allocation according to the actual communication environment and equipment requirements, and improve computing resource utilization efficiency and network service quality.

[0052] 3. Reduce data transmission delay and improve edge computing efficiency: Through the collaboration between each macro base station and the local MEC server, the present invention can distribute data processing tasks to edge computing nodes, reduce data transmission distance and delay, improve real-time processing capabilities, and ultimately perform global optimization by the cloud server to ensure efficient operation of the entire system.

[0053] 4. Enhanced adaptive optimization capability: Through dynamic cluster initialization and adaptive iterative optimization mechanism, the present invention dynamically adjusts the deployment location and configuration of heterogeneous cellular network base stations and MEC servers according to the devices whose QoS is not met, ensuring that the system can quickly respond to changes in the factory environment, achieve a more efficient and low-interference operation state, and meet the dynamic needs of the factory. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] By referring to the following drawings, the principle of the present invention can be more completely understood. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, other drawings can be innovated based on these drawings without creative work:

[0055] Figure 1is a schematic diagram of a discrete manufacturing edge intelligent network architecture in an exemplary embodiment of the present invention;

[0056] Figure 2 It is a flowchart of solving the edge intelligent network deployment optimization problem by jointly using the improved K-means algorithm and the PSO algorithm in an exemplary embodiment of the present invention;

[0057] In the figure: 1. Factory equipment, 2. Micro base station, 3. Wireless fronthaul connection, 4. Macro base station, 5. MEC server, 6. Cloud server. DETAILED DESCRIPTION

[0058] The exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, the embodiments described with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention. In addition, the accompanying drawings are only schematic illustrations of the present invention and are not drawn to a certain scale; some of the block diagrams in the accompanying drawings are functional entities, but they do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processors and / or microcontrollers.

[0059] Embodiment 1:

[0060] Provides an edge intelligent network adaptive system for discrete manufacturing, such as Figure 1 As shown, it includes multiple factory equipment 1, multiple micro base stations 2, multiple macro base stations 4, multiple MEC servers 5 and cloud servers 6. The micro base station 2 is mainly deployed in areas where the signal coverage of the macro base station 4 is weak or difficult to cover, and is connected to the macro base station 4 through wireless fronthaul technology 3, thereby enhancing the quality and coverage of the overall network signal. The factory equipment 1 selects the edge intelligent network node that provides the maximum SINR for access to ensure that the equipment always maintains the best network connection quality. Each macro base station 4 is connected to a MEC server 5, which is responsible for sharing local data processing and computing tasks, reducing data transmission latency and improving edge computing efficiency. Ultimately, the cloud server 6 performs global data coordination and optimization management;

[0061] Embodiment 2:

[0062] With respect to the system architecture of Example 1, a method for adaptive deployment of edge intelligent networks for discrete manufacturing is provided, and the steps are as follows:

[0063] S1: With the goal of reducing system interference and maximizing network service quality, the adaptive deployment optimization problem of edge intelligent network nodes (including heterogeneous cellular network base stations and MEC servers) is constructed according to the current network conditions and the network requirements of factory equipment, including the following sub-steps:

[0064] Sub-step S11: Assuming that the system adopts frequency division multiple access technology, the macro base station 4 and the micro base station 2 work on different frequency bands and allocate the required network resources to the factory equipment in a collaborative manner. ij =1 means that the i-th factory equipment is associated with the j-th edge intelligent network node, otherwise they are not associated, i∈[1,D]; x ik =1 indicates that the kth subcarrier is allocated to the i-th factory equipment, otherwise it is not allocated, k∈[1,K].

[0065] Sub-step S12: Construct a QoS model for the discrete manufacturing edge intelligent network system. Considering the presence of a large number of obstacles such as metal surfaces, large equipment, and buildings in the factory, as well as the impact of machine operation and production activity changes on the transmission quality of wireless signals, the wireless channel between factory equipment i and edge intelligent network node j is described using the WINNER II B3 channel model. Assume that the transmission power of factory equipment i is The carrier frequency is f ik , the distance between device i and its associated edge intelligent network node j is d ij , then the path loss model is expressed as:

[0066]

[0067] Among them, LOS and NLOS represent line-of-sight propagation and non-line-of-sight propagation respectively, and X σ It is represented as the loss correction factor caused by obstacles such as walls. Therefore, the signal power received by edge intelligent network node j is expressed as Affected by shadow fading, is a random variable that obeys a log-normal distribution, and the signal reception failure rate is defined as The probability of is:

[0068]

[0069] in, is the average signal power received by base station j, σ is the standard deviation of shadow fading, and Q(z) is the probability that the random variable x that obeys the standard normal distribution is greater than z, that is:

[0070]

[0071] Consider the co-channel interference generated by other devices in the factory that use the same carrier frequency as device i, as well as the impact of mechanical vibration, electromagnetic interference, and random noise generated by device operation on the data transmission of factory device i. Therefore, the uplink SINR of the i-th factory device can be expressed as:

[0072]

[0073] in, Indicates the co-channel interference to device i. Represents the thermal noise of factory equipment, P ijk (t) = A∑ l δ(t-lT) represents the noise generated by device operation, switching action or other periodic activities, where A is the pulse amplitude, δ(t) is the Dirac delta function, T is the pulse period, and lT is the specific time point when the lth pulse occurs.

[0074] Assume that m ij represents the data size transmitted from the i-th factory equipment to the j-th edge intelligent network node, b ij is the network bandwidth allocated to the i-th factory equipment. Then the sum of the transmission delays of all factory equipment within the coverage of the j-th edge intelligent network node is expressed as:

[0075]

[0076] Assuming that the jth edge intelligent network node has sufficient network capacity, the sum of the data sizes of all factory equipment within its coverage range is expressed as:

[0077]

[0078] The signal transmission in the system adopts quadrature phase shift keying modulation to effectively utilize the limited communication bandwidth. ij The probability of successful transmission of bits of data is the product of the probability of successful transmission of each bit or data unit. The probability that the data of all factory equipment within the coverage area of ​​the jth edge intelligent network node is successfully transmitted is:

[0079]

[0080] Sub-step S13: Construct the adaptive deployment optimization problem of discrete manufacturing edge intelligent network. If the key indicators of system network service quality include communication reachable rate, system delay and transmission reliability, the service quality of edge intelligent network nodes is defined as follows:

[0081]

[0082] Where z represents the deployment location of the edge intelligent network node, ω1, ω2, ω3∈[0,1] are the weighted coefficients of each key indicator in the system, and ω1+ω2+ω3=1. j is the sum of the actual data upload delays of all factory equipment within the coverage area of ​​the jth edge intelligent network node, T req The minimum delay that should be met for data transmission of factory equipment within its coverage area; R jis the transmission capacity that the jth edge intelligent network node can provide, R req The average network capacity required to upload data from factory equipment within the coverage area; PSR j PSR is the reliability of data transmission of factory equipment within the coverage area of ​​the jth edge intelligent network node. req The minimum reliability that must be met for successful data transmission to factory equipment within the coverage area.

[0083] Considering the uneven distribution of factory equipment and communication needs, reasonable and efficient deployment of edge intelligent network nodes can provide factory equipment with low latency, high reliability, and high data rate service quality, and meet the constraints of maximum system capacity and minimum transmission latency. Therefore, the edge intelligent network node adaptive deployment optimization problem is expressed as:

[0084]

[0085] C7:ω1+ω2+ω3=1

[0086] in, It is a service quality redundancy factor designed to cope with business fluctuations such as increased network load. Constraint C1 means that the device will only be connected to one base station at the same time, and constraint C2 means that each device only transmits data on a specific sub-channel, and data transmission through multiple sub-channels at the same time is not allowed, so as to avoid interference between channels and data packet conflicts. Constraint C3 means that the sum of the bandwidth allocated to all devices cannot exceed the total bandwidth of the base station, and the bandwidth allocated to each device must be non-negative. Constraints C4 to C6 respectively represent that the uplink transmission rate, latency and reliability of each factory device i connected to the edge intelligent network node j can be basically guaranteed, where R min 、T max and PSR min They represent the minimum transmission rate, maximum allowable transmission delay and minimum reliability of factory equipment i respectively.

[0087] Step S2: Figure 2 As shown in the figure, the improved K-means clustering algorithm and PSO algorithm are used to solve the adaptive deployment optimization problem, complete the intelligent deployment and dynamic adjustment of all heterogeneous cellular network base stations and all MEC servers, and improve the network service quality and computing resource utilization. The following sub-steps are included:

[0088] Sub-step S21: Identify potential QoS insufficiency areas. By monitoring the real-time data transmission of each device in the discrete manufacturing smart factory, using the preset QoS threshold standard R min 、T max and PSR min, detect the devices whose QoS is not met. Then analyze the number and spatial distribution of these devices whose QoS is not met, and quickly identify the potential underserved areas. The set of devices whose QoS is not met is

[0089] Sub-step S22: Use the improved K-means algorithm to dynamically initialize clustering of factory equipment that does not meet QoS, and use the cluster center as the candidate deployment location of the edge intelligent network node. Assume x j is the decision variable, j indexes all possible micro base station locations, x j =1 means that a base station is deployed at location j, otherwise it is not deployed. For each factory equipment i∈U, a ij =1 means that the device i whose QoS is not satisfied is covered by the micro base station at position j, and vice versa, if it is 0, the QoS of device i is still not satisfied. If the deployment cost of each micro base station is C j , then the objective function of dynamic clustering is expressed as:

[0090]

[0091]

[0092] For the optimization problem P2, the improved K-means algorithm can be used to solve it. The number of cluster centers can be dynamically adjusted in real time according to the number of devices whose QoS is not met and their spatial distribution, until an edge intelligent network node deployment solution is found that can ensure the service quality requirements of all devices in the set U and maximize cost-effectiveness. The algorithm is shown in Table 1:

[0093]

[0094]

[0095] Table 1

[0096] Sub-step S23: Use the adaptive weight PSO algorithm to accurately adjust the location and configuration of heterogeneous cellular network base stations and MEC servers. In the PSO algorithm, each particle has a position vector X (the particle's position in the solution space) and a velocity vector V (determines the flight direction and speed). Assume that V i t represents the velocity of the ith particle at time t, represents the position of the ith particle at time t, represents the best position searched historically by the i-th particle, g best represents the best position searched in the history of the entire group. In each iteration, each particle will be based on its own historical best position. and the global best position g in the populationbest To update its own speed and position, that is:

[0097]

[0098]

[0099] Among them, r1 and r2 are random numbers, c1 and c2 represent the acceleration weights that push particles to the individual optimal position and the group optimal position respectively. In order to maintain the balance between convergence speed and search effect, c1 and c2 are not 0. ω is the inertia weight coefficient at time t, which controls the continuity of particle velocity update. A higher weight coefficient is conducive to global search, while a lower weight coefficient is conducive to local search. Fixed inertia weights may not be able to effectively balance exploration and utilization in complex multi-objective optimization problems, causing particles to converge to the local optimal solution too early and difficult to adapt to the dynamically changing factory environment. In order to effectively balance global and local searches, the inertia weight is defined as:

[0100] ω=ω max -(ω max -ω min )·(t / T) p

[0101] Among them, ω max and ω min Respectively represent the maximum and minimum values ​​of the inertia weight coefficient, p is a constant, and T is the total number of iterations of the algorithm. In the early stage of the search, a larger inertia weight is used to help the algorithm explore a wider solution space. In the later stage of the iteration, the inertia weight is gradually reduced to enhance the local search ability of the particles, improve the accuracy of the optimal solution and the convergence of the algorithm.

[0102] The edge intelligent network adaptive deployment optimization algorithm based on the improved K-means algorithm and PSO algorithm is shown in Table 2:

[0103]

[0104]

[0105]

[0106] Table 2

[0107] It should be noted that although several units of the system for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units described above can be concretized in one unit. Conversely, the features and functions of a unit described above can be further divided into multiple units for concretization. Some or all of the units can be selected according to actual needs to achieve the purpose of the present invention. Those of ordinary skill in the art can understand and implement without paying creative work.

[0108] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary technical means in the art that are not disclosed by the present invention.

Claims

1. An edge intelligent network adaptive system for discrete manufacturing, characterized in that: The system includes multiple factory equipment, multiple heterogeneous cellular network base stations, multiple mobile edge computing servers and cloud servers. The heterogeneous cellular network base stations include macro base stations and micro base stations. The micro base stations are deployed in areas where the macro base station signal coverage is weak or difficult to cover, and are connected to the macro base stations through wireless fronthaul technology, thereby enhancing the quality and coverage of network signals; each factory equipment is connected to an edge intelligent network node that can provide the maximum signal-to-interference-to-noise ratio according to the actual communication environment, ensuring that the equipment always maintains the best network connection quality; each macro base station is connected to a mobile edge server, which is responsible for local data processing and computing tasks, reducing data transmission delays and improving edge computing efficiency, and ultimately achieving global data coordination and optimization management through cloud servers.

2. An edge intelligent network adaptive deployment method for discrete manufacturing, characterized in that: Here are the steps: (1) For the system described in claim 1, with the goal of reducing system interference and maximizing network service quality, an adaptive deployment optimization problem of edge intelligent network nodes is constructed according to current network conditions and network requirements of factory equipment; (2) The improved K-means clustering algorithm and PSO algorithm are used to solve the adaptive deployment optimization problem, complete the intelligent deployment and dynamic adjustment of all heterogeneous cellular network base stations and all mobile edge servers, and improve the network service quality and the utilization of computing resources.

3. The method for adaptively deploying edge intelligent networks for discrete manufacturing according to claim 2, characterized in that: The specific steps of constructing the adaptive deployment optimization problem of edge intelligent network nodes in step (1) are as follows: (1-1) Assuming that the system adopts frequency division multiplexing technology, the macro base station and the micro base station work on different frequency bands and allocate the required network resources to the factory equipment in a collaborative way. ij =1 means that the i-th factory equipment is associated with the j-th edge intelligent network node, otherwise they are not associated, i∈[1,D]; x ik =1 means the kth subcarrier is allocated to the ith factory equipment, otherwise it is not allocated, k∈[1,K]; (1-2) Construct a QoS model for discrete manufacturing edge intelligent network systems. Considering the presence of a large number of obstacles such as metal surfaces, large equipment, and buildings in the factory, as well as the impact of machine operation and production activity changes on the transmission quality of wireless signals, the wireless channel between factory equipment i and edge intelligent network node j is described using the WINNER II B3 channel model. Assume that the transmission power of factory equipment i is The carrier frequency is f ik , the distance between device i and its associated edge intelligent network node j is d ij , then the path loss model is expressed as: Among them, LOS and NLOS represent line-of-sight propagation and non-line-of-sight propagation respectively, and X σ It is represented as the loss correction factor caused by obstacles. Therefore, the signal power received by edge intelligent network node j is expressed as Affected by shadow fading, is a random variable that obeys a log-normal distribution, and the signal reception failure rate is defined as The probability of is: in, is the average signal power received by base station j, σ is the standard deviation of shadow fading, and Q(z) is the probability that the random variable x that obeys the standard normal distribution is greater than z, that is: Considering the co-channel interference generated by other devices in the factory that use the same carrier frequency as device i, as well as the impact of mechanical vibration, electromagnetic interference, and random noise generated by equipment operation on the data transmission of factory device i, the uplink SINR of the i-th factory device is expressed as: in, Indicates the co-channel interference to device i. Represents the thermal noise of factory equipment, P ijk (t) = A∑ l δ(t-lT) represents the noise generated by equipment operation, switching action or other periodic activities, where A is the pulse amplitude, δ(t) is the Dirac delta function, T is the pulse period, and lT is the specific time point when the lth pulse occurs; Assume that m ij represents the data size transmitted from the i-th factory equipment to the j-th edge intelligent network node, b ij is the network bandwidth allocated to the i-th factory equipment, then the sum of the transmission delays of all factory equipment within the coverage of the j-th edge intelligent network node is expressed as: Assuming that the jth edge intelligent network node has sufficient network capacity, the sum of the data sizes of all factory equipment within its coverage range is expressed as: The signal transmission in the system adopts quadrature phase shift keying modulation to effectively utilize the limited communication bandwidth. If the m ij The probability of successful transmission of bits of data is the product of the probability of successful transmission of each bit or data unit. The probability that the data of all factory equipment within the coverage area of ​​the jth edge intelligent network node is successfully transmitted is: (1-3) Construct the adaptive deployment optimization problem of discrete manufacturing edge intelligent network. If the key indicators of system network service quality include communication reachable rate, system delay and transmission reliability, the service quality of edge intelligent network nodes is defined as follows: Where z represents the deployment location of the edge intelligent network node, ω1, ω2, ω3∈[0,1] are the weighted coefficients of each key indicator in the system, and ω1+ω2+ω3=1, T j is the sum of the actual data upload delays of all factory equipment within the coverage area of ​​the jth edge intelligent network node, T req The minimum delay that should be met for data transmission of factory equipment within its coverage area; R j is the transmission capacity that the jth edge intelligent network node can provide, R req The average network capacity required to upload data from factory equipment within the coverage area; PSR j PSR is the reliability of data transmission of factory equipment within the coverage area of ​​the jth edge intelligent network node. req The minimum reliability that must be met for successful data transmission of factory equipment within the coverage area; Considering the uneven distribution of factory equipment and communication requirements, reasonable and efficient deployment of edge intelligent network nodes can provide factory equipment with low latency, high reliability, and high data rate service quality, and meet the constraints of maximum system capacity and minimum transmission delay. Therefore, the edge intelligent network node adaptive deployment optimization problem is expressed as: C7:ω1+ω2+ω3=1 in, It is a service quality redundancy factor designed to cope with business fluctuations such as increased network load. Constraint C1 means that the device will only be connected to one base station at the same time. Constraint C2 means that each device only transmits data on a specific subchannel, and data transmission through multiple subchannels is not allowed at the same time, so as to avoid interference and data packet conflicts between channels. Constraint C3 means that the sum of the bandwidth allocated to all devices cannot exceed the total bandwidth of the base station, and the bandwidth allocated to each device must be non-negative. Constraints C4~C6 respectively represent that the uplink transmission rate, delay and reliability of each factory device i connected to the edge intelligent network node j can be basically guaranteed, where R min , T max and PSR min They represent the minimum transmission rate, maximum allowable transmission delay and minimum reliability of factory equipment i respectively.

4. The method for adaptively deploying edge intelligent networks for discrete manufacturing according to claim 2, characterized in that: The specific steps of step (2) are as follows: (2-1) Identify potential areas of insufficient QoS By monitoring the real-time data transmission of each device in the discrete manufacturing smart factory, the preset QoS threshold standard R min 、T max and PSR min , detect the devices whose QoS is not met, and then analyze the number and spatial distribution of these devices whose QoS is not met, and quickly identify the potential underserved areas. The set of devices whose QoS is not met is (2-2) Use the improved K-means algorithm to dynamically initialize clustering of factory equipment that does not meet QoS requirements, and use the cluster center as the candidate deployment location of the edge intelligent network node; Assume x j is the decision variable, j indexes all possible micro base station locations, x j =1 means that a base station is deployed at position j, otherwise it is not deployed. For each factory equipment i∈U, a ij =1 means that the device i whose QoS is not satisfied is covered by the micro base station at position j, and vice versa, if it is 0, the QoS of device i is still not satisfied. If the deployment cost of each micro base station is C j , then the objective function of dynamic clustering is expressed as: For the optimization problem P2, the improved K-means algorithm is used to solve it. The number of cluster centers is dynamically adjusted in real time according to the number of devices whose QoS is not met on site and their spatial distribution, until an edge intelligent network node deployment solution is found that can ensure the service quality requirements of all devices in the set U and maximize cost-effectiveness. (2-3) Use the adaptive weight PSO algorithm to accurately adjust the location and configuration of heterogeneous cellular network base stations and MEC servers; In the PSO algorithm, each particle has a position vector X and a velocity vector V. X represents the position of the particle in the solution space, and V determines the flight direction and velocity. Assume that V i t represents the velocity of the ith particle at time t, represents the position of the ith particle at time t, represents the best position searched historically by the i-th particle, g best Represents the best position searched in the history of the entire group. In each iteration, each particle will be based on its own historical best position. and the global best position g in the population best To update your speed and position, that is: Among them, r1 and r2 are random numbers, c1 and c2 represent the acceleration weights that push particles to the individual optimal position and the group optimal position respectively. In order to maintain the balance between convergence speed and search effect, c1 and c2 are not 0; ω is the inertia weight coefficient at time t, which controls the continuity of particle velocity update. A higher weight coefficient is conducive to global search, while a lower weight coefficient is conducive to local search. Fixed inertia weight cannot effectively balance exploration and utilization in complex multi-objective optimization problems, resulting in particles converging to the local optimal solution too early and difficult to adapt to the dynamically changing factory environment. In order to effectively balance global and local searches, the inertia weight is defined as: oh = oh max -(oh max -oh min )·(t / T) p Among them, ω max and ω min They represent the maximum and minimum values ​​of the inertia weight coefficient respectively, p is a constant, T is the total number of iterations of the algorithm, and a larger inertia weight is used in the early stage of the search to help the algorithm explore a wider solution space. In the later stage of the iteration, the inertia weight is gradually reduced to enhance the local search ability of the particles, thereby improving the accuracy of the optimal solution and the convergence of the algorithm.

Images