Reliability modeling and evaluation method for Internet of Things in intelligent manufacturing system

Abstract

The invention discloses a reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system. The method comprises the following steps: step 1, collecting the structure and operation information of an intelligent manufacturing system network; 2, establishing a production unit reliability model in the intelligent manufacturing physical system; 3, establishing an information unit reliability model in the intelligent manufacturing information system; 4, establishing an Internet of Things system reliability model according to the intelligent manufacturing system information physical system coupling relationship; and 5, evaluating the reliability of the Internet of Things system by adopting Monte Carlo simulation. According to the invention, a decision maker can be helped to carry out reliability modeling and evaluation on the Internet of Things in the intelligent manufacturing system, the reliability of the Internet of Things in the intelligent manufacturing system and the capability of meeting production requirements are improved, and the influence of the network on the production process is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing and industrial Internet of Things, and specifically relates to a reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system, which considers the system reliability modeling and evaluation under the tight coupling and cascading failures of information-physical systems. Background Art

[0002] The advent of the Industry 4.0 era, characterized by digitalization, automation, and connectivity, has triggered profound changes in manufacturing systems worldwide. As a pillar of the national economy, manufacturing is at a critical stage of transformation and upgrading from traditional manufacturing to high-end manufacturing. Intelligent manufacturing systems, at the core of the Industry 4.0 revolution, emphasize the deep integration of advanced information and communication technologies with manufacturing systems, aiming to create a highly efficient, automated, and intelligently decision-making production environment. However, the current application of the Internet of Things in intelligent manufacturing systems poses numerous risks and urgently requires further research.

[0003] In actual production, the widespread interconnection of IoT devices has brought increased flexibility and efficiency. However, due to the tightly coupled cyber-physical systems of IoT devices, which work together to complete complex production tasks, the system has become increasingly complex and fragile. Failure of some IoT devices can trigger a chain reaction, bringing the entire production line to a standstill and resulting in significant economic losses. For example, in the automotive manufacturing industry, data transmission errors in sensor networks can cause robots on the production line to malfunction, damaging parts, wasting resources and delaying production cycles, impacting product delivery. Given the serious impact these IoT reliability issues pose on the efficiency, security, and stability of intelligent manufacturing systems, in-depth research on IoT reliability and the development of effective methods for its evaluation and improvement are urgently needed. This paper aims to provide new insights and solutions to address the challenges of IoT reliability by establishing a scientific mathematical model and introducing innovative evaluation methods. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a reliability modeling and evaluation method that can describe the coupling relationship between the information and physical systems of intelligent manufacturing systems, and deal with the difficulties in reliability analysis and deduction caused by the mutual correlation of factors such as production decisions and information network data transmission performance, so as to improve the reliability of the Internet of Things in intelligent manufacturing systems, provide theoretical support and methodological guidance, and thus enhance the efficiency and safety of the manufacturing process.

[0005] The objective of the present invention is achieved through the following technical solution: A reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system, comprising the following steps:

[0006] Step 1: The structure and operation information of the Internet of Things in the intelligent manufacturing system are obtained, and the number of production units and information units in the manufacturing system, the routing paths between information units and production units, and the load capacity of the routing nodes therein are obtained; the production speed of each unit in the intelligent manufacturing system and the flow direction of materials in the production process are obtained;

[0007] Step 2: Establish a production unit reliability model based on the state characteristics, production capacity change patterns, and state transition mechanisms of the production unit in the intelligent manufacturing physical system. By analyzing this model, the reliability status of the production unit at different times is obtained, and the impact of the production unit degradation over time on production capacity is clarified.

[0008] Step 3: Based on the binary characteristics of information nodes in the intelligent manufacturing information system, the relationship between data flow density and capacity, and the load calculation logic of different types of nodes, such as terminals, routers, and gateways, an information unit reliability model is established. By solving this model, the accurate load of the information node in the fault-free state is obtained, and the degree to which the accuracy of data transmission is affected by abnormal load is determined.

[0009] Step 4: Based on the coupling relationship between cyber-physical systems in intelligent manufacturing systems, various cascading failure modes, such as overload, isolation, and interdependence, are comprehensively considered. The accuracy of information transmission between nodes is correlated with node load. By using factors such as the observation probability matrix and the production instruction release probability matrix, an IoT system reliability model is established. By solving and analyzing this model, the probability of the system meeting production requirements under different states is determined, thereby clarifying the reliability level of the IoT system.

[0010] Step 5: Use the Monte Carlo simulation method to repeat the simulation process multiple times and evaluate the reliability of the IoT system based on the simulation results.

[0011] 2. The reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system according to claim 1, characterized in that in the production unit model in the physical system in step 2, the reliability of the production unit is evaluated by analyzing the relationship between the production unit status, production capacity, state transition intensity and production rate;

[0012] In this step, the operating state of each physical production unit in the intelligent manufacturing system is first modeled. In order to accurately describe the performance degradation process of production equipment in long-term operation, the present invention divides the unit operating state into a finite number of discrete levels to form a state set To dynamically describe the evolution of unit states over time, a Markov process model is introduced to construct a state transition mechanism. This model assumes that state evolution depends only on the current state and the current production rate. The transition intensity between states is expressed as:

[0013]

[0014] i represents the initial state index of the production unit; j represents the target state index of the production unit; l represents the identification of the production unit; represents the standard state transition intensity, which represents the state transition intensity at the maximum production rate; α(γ l (t)) represents the production rate γ l (t) A coefficient related to the production rate, with a value of [0, 1], which increases monotonically with the production rate;

[0015] 3. The reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system according to claim 1 is characterized in that the specific implementation method of step 3 is as follows: by treating information nodes as binary components, describing their lifespans with probability density functions, representing data processing capabilities with normalization coefficients and capacities, calculating loads based on the characteristics of different types of nodes and taking load redistribution into account, and analyzing the impact of loads on data transmission accuracy; for different types of information nodes, the load calculation method is different, specifically expressed as follows:

[0016]

[0017] k represents the information node index; ω k (t) represents the load of information node k at time t; normalization coefficient z k Indicates the data flow density between information nodes; Represents a collection of terminal nodes in an information network. Terminal nodes are responsible for collecting production unit data and transmitting it to routing nodes. Represents a collection of routing nodes in an information network. Routing nodes must not only forward their own data but also process data from other nodes. Represents the collection of gateway nodes in the information network. Gateway nodes are responsible for uploading the data of routing nodes to the cloud. l，k (t) indicates whether the routing path from node l to the gateway passes through node k at time t. If it passes through h l，k (t)=1, otherwise h l，k (t)=0.

[0018] 4. The reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system according to claim 1 is characterized in that the specific implementation method of step 4 is: by analyzing factors such as fault type, data transmission accuracy, and production decision-making, an Internet of Things system reliability model is constructed; the specific operation process is as follows:

[0019] Step 41: Quantify the accuracy of data transmission of information nodes: According to the relationship between the information node load and its link capacity, use the piecewise function E k (ω k(t)) Calculate the accuracy of data transmission:

[0020]

[0021] Where, ω k (t) represents the information node load; Q k Indicates link capacity; The accuracy of data transmission when the load of information nodes does not exceed the capacity; b k It represents the minimum transmission accuracy of the information node in the tolerable overload stage; δ represents the tolerable load factor of the information node.

[0022] Step 42: Construct the state observation probability matrix: Based on the accuracy of data transmission, calculate the probability that the state of production unit l is observed to be o through node k when the actual state is i. When the observed state is consistent with the actual state, the probability is the transmission accuracy of the current node; when the observed state is inconsistent with the actual state, the probability is evenly distributed on other possible states, specifically expressed as:

[0023]

[0024] Step 43, calculate the cumulative observation probability of the path: put the production unit l on the path (V l path The observation probability matrix (B) of all nodes passed by (t) l，k ) and multiply them to get the comprehensive observation probability matrix B i , specifically expressed as:

[0025]

[0026]

[0027] Step 44: Calculate the state transition probability: Based on the Markov process assumption, the system state transition probability is equal to the product of the state transition probabilities of each production unit, which is specifically expressed as:

[0028]

[0029] 5. The reliability modeling and evaluation method of the Internet of Things in the intelligent manufacturing system according to claim 1 is characterized in that the specific implementation method of step 5 is: by facilitating all possible state combinations x i , observation combination o j and production strategy combinations Comprehensively consider the state-observation transition probability p i，j (t) and the probability of view-strategy transition q j，m (t), calculate to meet production demand The probability of , where , is specifically expressed as:

[0030]

[0031]

[0032]

[0033] Where, is the system productivity given the productivity of all units; D represents the production demand of the system; 1{.} represents the indicator function; The lth unit is at the kth node in the path and is observed to be o in the actual state i. j The probability of V l path (t) The set of path nodes of unit l at time t.

[0034] The beneficial effects of the present invention are: the method of the present invention proposes reliability modeling of the Internet of Things system taking into account the coupling relationship between information and physical networks. By collectively defining and probabilistically modeling the production unit status, observations, production rate combinations, etc. in the Internet of Things system, it helps to more accurately measure the reliability of the Internet of Things system in actual production, helps enterprises to discover potential risks of the system in advance, optimize production decisions and resource allocation, improve the efficiency, safety and profitability of the manufacturing process, and promote more reliable and efficient application of Internet of Things technology in actual scenarios such as intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the process of reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system of the present invention;

[0036] Figure 2 These are the system reliability curves under three different production scenarios output by the present invention. DETAILED DESCRIPTION

[0037] The technical solution of the present invention is further described below with reference to the accompanying drawings.

[0038] like Figure 1 As shown, a reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system of the present invention includes the following steps:

[0039] Step 1: Collect the structure and operation information of the Internet of Things in the intelligent manufacturing system, and obtain the connection relationship between the source node (terminal information node that provides production task control data or status collection data), sink node (intermediate communication node that receives or relays information), transfer node (data path node that is only responsible for information forwarding and does not directly participate in terminal collection or edge control), and link (data path that forwards information between different Internet of Things nodes) of the unit Internet of Things system.

[0040] By calculating each node's data upload rate, forwarding channel capacity, and the maximum achievable communication capacity between links, we construct the physical information coupling path structure of the "terminal-relay-edge" system. Each path consists of a unique multi-hop forwarding link between a terminal node and an edge node. The node with the smallest capacity in the path determines the maximum communication bandwidth that the path can support, which is the path's actual effective information capacity.

[0041] If any information node on any path loses communication due to excessive load, network fluctuations, or failure, the production units along that path will face interruptions in state observation, misdirected control commands, or packet loss, impacting their reliable operation. Therefore, based on statistics on the terminal-bound path information topology, combined with node capability parameters and path link structure information, we establish a node set s and a path set L at the system information layer, and obtain their state capacity and connection reliability indicators for subsequent reliability analysis. Furthermore, we collect abnormal state data on information nodes or links caused by sudden disturbances to assess the robustness of system operation when node-link combinations fail, and to obtain failure probability parameters for each network subcomponent.

[0042] This embodiment uses a typical workshop-level intelligent manufacturing Internet of Things system structure composed of an industrial equipment network and an information communication network as the analysis object. Each production unit node has a maximum output capacity, which corresponds to the maximum processing rate of material processing in the process; each information unit node includes a terminal perception unit, a communication relay node and an edge processing unit, and has certain data perception and forwarding capabilities.

[0043] The industrial equipment network consists of eight major production unit nodes, including automatic placement machines, laser marking machines, inspection robots, and packaging units. Table 1 shows the capacity status of these production unit nodes. The information communication network comprises 12 information nodes, organized into three levels: edge sensors, repeaters, and gateways. Information flows are collected by terminals and then sequentially transmitted through multiple levels of relay nodes to the control end, where control signals are fed back to guide production scheduling for each device. Table 2 shows the communication capacity status of these information unit nodes.

[0044] Table 1

[0045]

[0046] Table 2

[0047]

[0048] The parameters used in the present invention and their meanings are shown in Table 3.

[0049] Table 3

[0050]

[0051] This case sets three production scenarios:

[0052] Scenario 1: Production scenario with degraded cyber-physical system coupling

[0053] Scenario 2: Assume the information network always operates perfectly, with only production units deteriorating during operation. In this scenario, the true state of the production units is fully accessible, allowing optimal production instructions to be executed. This scenario is indeed a common reliability assessment model for manufacturing systems without an information network.

[0054] Scenario 3 assumes that the production unit always operates perfectly, with only information nodes deteriorating during operation. In this scenario, the decline in manufacturing system performance is entirely due to information network degradation. Although this scenario does not exist in practice, it is intended to illustrate the impact of information network degradation on manufacturing system performance.

[0055] Step 2: Establish a production unit reliability model based on the state characteristics, production capacity change rules, and state transition mechanism of the production unit in the intelligent manufacturing physical system; by analyzing the model, obtain the reliability state of the production unit at different times, and then clarify the impact of the production unit degradation over time on the production capacity. The reliability of the production unit is evaluated by analyzing the relationship between the production unit state, production capacity, state transition intensity, and production rate; in order to accurately characterize the performance degradation process of production equipment in long-term operation, the present invention divides the unit operation state into a finite number of discrete levels to form a state set To dynamically describe the evolution of unit states over time, a Markov process model is introduced to construct a state transition mechanism. This model assumes that state evolution depends only on the current state and the current production rate. The transition intensity between states is expressed as:

[0056] Specifically expressed as:

[0057]

[0058] i represents the initial state index of the production unit; j represents the target state index of the production unit; l represents the identification of the production unit; represents the standard state transition intensity, which represents the state transition intensity at the maximum production rate; α(γ l (t)) represents the production rate γ l (t) A coefficient related to the production rate, with a value of [0, 1], which increases monotonically with the production rate;

[0059] This case study includes eight production units, namely, patching, marking, testing, packaging, dispensing, assembly, sorting, and aging. Each production unit is divided into five reliability states, corresponding to production capacity levels of 100%, 90%, 80%, 70%, and 0%. During the simulation, the state of unit l is related to its current load and is subject to the influence of information path error feedback, thereby dynamically evolving its degradation trend and obtaining the reliability time-varying evolution sequence γ. l (t).

[0060] Step 3: Based on the binary characteristics of information nodes in the intelligent manufacturing information system, the relationship between data flow density and capacity, and the load calculation logic of different types of nodes, such as terminals, routers, and gateways, an information unit reliability model is established; by solving this model, the accurate load situation of the information node in the fault-free state is obtained, and then the degree to which the accuracy of data transmission is affected by load anomalies is determined.

[0061] By treating information nodes as binary components, describing their lifespans with probability density functions, and expressing data processing capabilities with normalization coefficients and capacity, we calculate loads based on the characteristics of different types of nodes and consider load redistribution to analyze the impact of loads on data transmission accuracy. Load calculation methods differ for different types of information nodes, as shown in the following table:

[0062]

[0063] k represents the information node index; ω k (t) represents the load of information node k at time t; normalization coefficient z k Indicates the data flow density between information nodes; Represents a collection of terminal nodes in an information network. Terminal nodes are responsible for collecting production unit data and transmitting it to routing nodes. Represents a collection of routing nodes in an information network. Routing nodes must not only forward their own data but also process data from other nodes. Represents the collection of gateway nodes in the information network. Gateway nodes are responsible for uploading the data of routing nodes to the cloud. l，k (t) indicates whether the routing path from node l to the gateway passes through node k at time t. If it passes through h l，k (t)=1, otherwise h l，k (t)=0.

[0064] In this case, the system includes 12 information nodes (S1 to S12), including 8 terminal nodes, 3 relay nodes, and 1 gateway node. Each production unit is bound to a three-segment information path (terminal-relay-gateway), generating 20 data per minute. At any time, the total load of each node can be calculated based on the path structure. For example, node S2 needs to forward path data from P1 and P2 at the same time, and the total load is This model is further used in subsequent steps to construct the accuracy function b of the information path k (ω k (t)) and the system reliability transfer matrix, which reflect the indirect impact of information layer performance degradation on the physical system capacity assessment results.

[0065] Step 4: Based on the coupling relationship between cyber-physical systems in intelligent manufacturing systems, various cascading failure modes, such as overload, isolation, and interdependence, are comprehensively considered. The accuracy of information transmission between nodes is correlated with node load. By using factors such as the observation probability matrix and the production instruction release probability matrix, an IoT system reliability model is established. By solving and analyzing this model, the probability of the system meeting production requirements under different states is determined, thereby clarifying the reliability level of the IoT system.

[0066] Step 41: Quantify the accuracy of data transmission of information nodes: According to the relationship between the information node load and its link capacity, use the piecewise function E k (ω k (t)) Calculate the accuracy of data transmission:

[0067]

[0068] Where, ω k (t) represents the information node load; Q k Indicates link capacity; The accuracy of data transmission when the load of information nodes does not exceed the capacity; b k It represents the minimum transmission accuracy of the information node in the tolerable overload stage; δ represents the tolerable load factor of the information node.

[0069] In this case, the system includes 12 information nodes, including 8 terminal nodes, 3 routing nodes, and 1 edge gateway. The capacity and tolerance multiples of each node are shown in Table 1. The upload rate of all terminals is . The load calculation results of the path relay node are as follows: S2 load comes only from P1 and P2, passing through S5 and S6. S3 load comes only from P3, and S4 load comes only from P4. The S1 gateway node accepts all 8 paths, with a total load of 160. The node accuracy parameter is b k =0.80,δ=1.5.

[0070] Step 42: Construct the state observation probability matrix: Based on the accuracy of data transmission, calculate the probability that the state of production unit l is observed to be o through node k when the actual state is i. When the observed state is consistent with the actual state, the probability is the transmission accuracy of the current node; when the observed state is inconsistent with the actual state, the probability is evenly distributed on other possible states, specifically expressed as:

[0071]

[0072] In this intelligent manufacturing IoT system, eight production units are configured, each bound to eight information paths. Each path consists of a terminal sensor, a relay router, and an edge gateway. For example, the path corresponding to P1 is S5→S2→S1. The observation performance of an information node in a path degrades due to changes in its load, which in turn affects the accuracy of the final control instructions. Each production unit state is divided into five levels (full load-failure), resulting in a corresponding observation error matrix of 5×5. When the load of an information node increases, its accuracy decreases and the observation error increases.

[0073] For example, if the path corresponding to P1 is S5→S2→S1, and the node load of node S2 is is less than its capacity, then Observation is accurate; if the load rises to is greater than its capacity, At this time, the matrix diagonal is 0, and the remaining items are evenly distributed, which is a random misjudgment.

[0074] Step 43, calculate the cumulative observation probability of the path: put the production unit l on the path (V l path The observation probability matrix (B) of all nodes passed by (t) l，k ) and multiply them to get the comprehensive observation probability matrix B i , specifically expressed as:

[0075]

[0076]

[0077] Taking the P1 unit as an example, its path is S5-S2-S1, and the corresponding path observation matrix is:

[0078]

[0079] This matrix is ​​used to calculate the cumulative error of the production unit's true state being misjudged as other states, which is used for subsequent decision-making transmission.

[0080] Step 44, calculate the state transition probability: Based on the Markov process assumption, the system state transition probability is equal to the product of the state transition probabilities of each production unit. The system state space consists of 8 production units, each unit has 5 states, so the state combination space is 390625. For controllable modeling, assuming that the equipment unit states are independent, the state transition probability from any time t to t+Δt can be written as: Specifically, it is:

[0081]

[0082] Each transition probability is controlled by the rate function defined in step 2:

[0083]

[0084] Taking P6 as an example, its maximum production capacity is 85 pieces / hour. If the current production rate γ6(t) = 68, then α = 0.8, the degradation rate from state 1 to state 2 increases, and the probability of entering a low state increases.

[0085] Step 5: Use the Monte Carlo simulation method to repeat the simulation process multiple times and evaluate the reliability of the IoT system based on the simulation results. i , observation combination o j and production strategy combinations Comprehensively consider the state-observation transition probability p i，j (t) and the probability of view-strategy transition q j，m (t), calculate to meet production demand The probability of , where , is specifically expressed as:

[0086]

[0087]

[0088]

[0089] Where, is the system productivity given the productivity of all units; D represents the production demand of the system; 1{.} represents the indicator function; The lth unit is at the kth node in the path and is observed to be o in the actual state i. j The probability of V l path (t) The set of path nodes of unit l at time t.

[0090] The term "disturbance scenario" in this method refers to a combination of abnormal cyber-physical system operations caused by non-structural factors such as network delays, node overloads, relay link disconnections, and information feedback failures or distortions. This combination affects the status of production units in three categories of variables, observation errors in information paths, and execution errors in control strategies.

[0091] To obtain the expected value of reliability, a Monte Carlo simulation method is used to evaluate a large number of disturbance scenarios, and path compression technology is introduced to improve computational efficiency. The specific settings are as follows:

[0092] Simulation step Δt 1 Total period T 36 Total number of disturbance scenarios N 500 Typical combination extraction number M 30 The number of repeated samplings per combination simulation Q 500

[0093] The simulation evaluation mechanism constructed in this step, combining factors such as information layer structural state degradation, changes in path load accuracy, strategy misjudgment, and feedback accuracy, effectively characterizes the impact of "observation-control-capacity" link interruptions in the IoT system on system robustness, and can provide basic reliability quantitative indicators for production line scheduling.

[0094] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A reliability modeling and evaluation method for the Internet of Things in an intelligent manufacturing system, characterized by: The steps include: Step 1: The structure and operation information of the Internet of Things in the intelligent manufacturing system are obtained, and the number of production units and information units in the manufacturing system, the routing paths between information units and production units, and the load capacity of the routing nodes therein are obtained; the production speed of each unit in the intelligent manufacturing system and the flow direction of materials in the production process are obtained; Step 2: Establish a production unit reliability model based on the state characteristics, production capacity change patterns, and state transition mechanisms of the production unit in the intelligent manufacturing physical system. By analyzing this model, the reliability status of the production unit at different times is obtained, and the impact of the production unit degradation over time on production capacity is clarified. Step 3: Based on the binary characteristics of information nodes in the intelligent manufacturing information system, the relationship between data flow density and capacity, and the load calculation logic of different types of nodes, such as terminals, routers, and gateways, an information unit reliability model is established. By solving this model, the accurate load of the information node in the fault-free state is obtained, and the degree to which the accuracy of data transmission is affected by abnormal load is determined. Step 4: Based on the coupling relationship between cyber-physical systems in intelligent manufacturing systems, various cascading failure modes, such as overload, isolation, and interdependence, are comprehensively considered. The accuracy of information transmission between nodes is correlated with node load. By using factors such as the observation probability matrix and the production instruction release probability matrix, an IoT system reliability model is established. By solving and analyzing this model, the probability of the system meeting production requirements under different states is determined, thereby clarifying the reliability level of the IoT system. Step 5: Use the Monte Carlo simulation method to repeat the simulation process multiple times and evaluate the reliability of the IoT system based on the simulation results.

2. The reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system according to claim 1 is characterized in that: In the production unit model of the physical system in step 2, the reliability of the production unit is evaluated by analyzing the relationship between the production unit state, production capacity, state transition intensity and production rate; in order to accurately describe the performance degradation process of the production equipment in long-term operation, the present invention divides the unit operation state into a finite number of discrete levels to form a state set In order to dynamically describe the evolution of unit states over time, a Markov process model is introduced to construct a state transition mechanism. This model assumes that state evolution depends only on the current state and the current production rate. The transition intensity between states is expressed as: Where i represents the initial state index of the production unit; j represents the target state index of the production unit; l represents the identifier of the production unit; represents the standard state transition intensity, which represents the state transition intensity at the maximum production rate; α(γ l (t)) represents the production rate γ l (t) is a coefficient related to the production rate, with a value of [0, 1] and monotonically increasing with the production rate.

3. The reliability modeling and evaluation method of the Internet of Things in the intelligent manufacturing system according to claim 1 is characterized in that: The specific implementation method of step 3 is as follows: by treating information nodes as binary components, describing their lifespans with probability density functions, expressing data processing capabilities with normalization coefficients and capacity, calculating loads based on the characteristics of different types of nodes and taking load redistribution into account, and analyzing the impact of loads on data transmission accuracy; different types of information nodes use different load calculation methods, specifically expressed as follows: Where k represents the information node index; ω k (t) represents the load of information node k at time t; normalization coefficient z k Indicates the data flow density between information nodes; Represents a collection of terminal nodes in an information network. Terminal nodes are responsible for collecting production unit data and transmitting it to routing nodes. Represents a collection of routing nodes in an information network. Routing nodes must not only forward their own data but also process data from other nodes. Represents the collection of gateway nodes in the information network. Gateway nodes are responsible for uploading the data of routing nodes to the cloud. l，k (t) indicates whether the routing path from node l to the gateway passes through node k at time t. If it passes through h i，k (t)=1, otherwise h l，k (t)=0.

4. The reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system according to claim 1 is characterized in that: The specific implementation method of step 4 is: by analyzing factors such as fault type, data transmission accuracy, and production decision-making, an IoT system reliability model is constructed; the specific operation process is as follows: Step 41: Quantify the accuracy of data transmission of information nodes: According to the relationship between the information node load and its link capacity, use the piecewise function E k (ω k (t)) Calculate the accuracy of data transmission: Where, ω k (t) represents the information node load; Q k Indicates link capacity; The accuracy of data transmission when the load of information nodes does not exceed the capacity; b k represents the minimum transmission accuracy of the information node in the tolerable overload stage; δ represents the tolerable load factor of the information node; Step 42: Construct the state observation probability matrix: Based on the accuracy of data transmission, calculate the probability that the state of production unit l is observed to be o through node k when the actual state is i. When the observed state is consistent with the actual state, the probability is the transmission accuracy of the current node; when the observed state is inconsistent with the actual state, the probability is evenly distributed on other possible states, specifically expressed as: Step 43, calculate the cumulative observation probability of the path: put the production unit l on the path (V l path The observation probability matrix (B) of all nodes passed by (t) l，k ) and multiply them to get the comprehensive observation probability matrix B i , specifically expressed as: Step 44: Calculate the state transition probability: Based on the Markov process assumption, the system state transition probability is equal to the product of the state transition probabilities of each production unit, which is specifically expressed as:

5. The reliability modeling and evaluation method of the Internet of Things in an intelligent manufacturing system according to claim 1 is characterized in that: The specific implementation method of step 5 is: by facilitating all possible state combinations x i , observation combination o j and production strategy combinations Comprehensively consider the state-observation transition probability p i，j (t) and the probability of view-strategy transition q j，m (t), calculate to meet production demand The probability of , where , is specifically expressed as: Where, is the system productivity given the productivity of all units; D represents the production demand of the system; 1{.} represents the indicator function; The lth unit is at the kth node in the path and is observed to be o in the actual state i. j The probability of V l path (t) The set of path nodes of unit l at time t.

Images