Industrial internet-based device maintenance strategy determination method, device, equipment and medium

Abstract

The application discloses a device maintenance strategy determination method and device based on an industrial internet, and relates to the field of intelligent production and manufacturing. The method comprises the following steps: dividing a plurality of production devices and a plurality of buffer zones in a production line into a plurality of sub-device units; determining an initial maintenance strategy and a first state transition probability matrix of each sub-device unit, determining a second state transition probability matrix of a corresponding virtual device of each sub-device unit, and calculating a first internal blocking rate of each sub-device unit; updating the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix; and iteratively updating the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate and the second state transition probability matrix to obtain a target maintenance strategy of each production device. The application can effectively solve the maintenance decision problem of a multi-machine production line system and improve the feasibility and efficiency of the solving process.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing technology, and in particular to a method, apparatus, equipment and medium for determining equipment maintenance strategies based on the Industrial Internet. Background Technology

[0002] In modern manufacturing, predictive maintenance can predict the remaining service life of a machine based on its real-time operating status and determine the most cost-effective time to intervene in maintenance before a failure occurs.

[0003] Existing technologies mostly address maintenance decisions for single or two machines. However, for multi-machine production lines, as the number of machines increases, it's necessary to consider the direct impact of machine degradation on product quality and the complex interconnected effects. Current technologies struggle to effectively address predictive maintenance decisions in multi-machine production lines, particularly considering the overall output performance of the line, limiting their effectiveness in practical applications. Therefore, achieving globally optimal maintenance decisions in multi-machine systems requires new modeling and solution methods to address the computational complexity and decision coordination issues. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and medium for determining equipment maintenance strategies based on the Industrial Internet, which can effectively handle maintenance decision-making problems of large-scale multi-machine production line systems and improve the feasibility and efficiency of the solution process.

[0005] Firstly, this application provides a method for determining equipment maintenance strategies based on the Industrial Internet, the method comprising: The production line is divided into multiple sub-equipment units according to the upstream and downstream sequence of the production line. Determine the initial maintenance strategy for each of the sub-device units; Determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit; The first state transition probability matrix is ​​updated based on the first internal blocking rate to obtain a new first state transition probability matrix. The initial maintenance strategy is iteratively updated based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device.

[0006] Furthermore, the first sub-device unit among the plurality of sub-device units includes the first production device among the plurality of production devices, the first buffer corresponding to the first production device, and the second production device. The non-first sub-device units among the plurality of sub-device units include the previous virtual device, the previous buffer corresponding to the previous production device, and the current production device. The previous virtual device is a virtual device composed of all production devices located upstream of the current production device.

[0007] Furthermore, determining the second state transition probability matrix of the virtual device corresponding to the sub-device unit includes: determining the steady-state probability distribution of the sub-device unit; calculating the failure probability of the sub-device unit being in a fault state and the repair probability of the sub-device unit transitioning from a fault state to an operating state based on the steady-state probability distribution and the product quality data of the current production equipment; and constructing the second state transition probability matrix of the virtual device based on the failure probability and the repair probability.

[0008] Furthermore, the first internal blocking rate is generated by the upstream production equipment on the sub-device unit, and the first state transition probability matrix is ​​composed of the state transition probabilities corresponding to different equipment maintenance methods under the initial maintenance strategy; updating the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix includes: obtaining the second internal blocking rate of the sub-device unit caused by the downstream production equipment from the first internal blocking rate corresponding to each sub-device unit; updating the state transition probability corresponding to the preset equipment maintenance method based on the second internal blocking rate to obtain a new first state transition probability matrix.

[0009] Furthermore, the iterative update of the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device includes: updating the initial maintenance strategy based on the new first state transition probability for each sub-device unit to obtain an intermediate maintenance strategy; obtaining a new first internal blocking rate during the update of the second state transition probability matrix based on the intermediate maintenance strategy and the new first state transition probability; if the difference between the new first internal blocking rate and the first internal blocking rate for each sub-device unit does not meet a preset convergence threshold, then continuing to update the initial maintenance strategy based on the new first state transition probability, the first internal blocking rate, and the second state transition probability matrix for each sub-device unit to obtain an intermediate maintenance strategy and a new first internal blocking rate for each sub-device unit; if the difference between the new first internal blocking rate and the first internal blocking rate meets the preset convergence threshold, then using the intermediate maintenance strategy as the target maintenance strategy.

[0010] Furthermore, for the first sub-device unit, determining the initial maintenance strategy corresponding to the sub-device unit includes: constructing a Markov decision process model for the first production equipment, the first buffer, and the second production equipment; and calculating the initial maintenance strategy for the first production equipment and the second production equipment based on the Markov decision process model.

[0011] Furthermore, for the non-first sub-device unit, determining the initial maintenance strategy corresponding to the sub-device unit includes: constructing a Markov decision process model for the previous virtual device, the previous buffer, and the current production device; and calculating the initial maintenance strategy for the current production device based on the Markov decision process model.

[0012] Secondly, this application provides an equipment maintenance strategy determination device based on the Industrial Internet, the device comprising: The equipment unit division module is used to divide multiple production equipment and multiple buffers in the production line into multiple sub-equipment units according to the upstream and downstream sequence of the production line. An initial strategy determination module is used to determine the initial maintenance strategy corresponding to each of the sub-device units; The data processing module is used to determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit. The data update module is used to update the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix. The target strategy determination module is used to iteratively update the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device.

[0013] Thirdly, this application provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to execute the device maintenance strategy determination method based on the Industrial Internet as described in any embodiment of this application.

[0014] Fourthly, this application provides a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute the device maintenance strategy determination method based on the Industrial Internet described in any embodiment of this application.

[0015] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the device maintenance strategy determination method based on the Industrial Internet described in any embodiment of this application.

[0016] To address the shortcomings of existing technologies, this application provides a method for determining equipment maintenance strategies based on the Industrial Internet. This method offers the following advantages: It divides a multi-machine pipeline into multiple two-machine sub-units and independently solves for the optimal maintenance strategy for each unit. Then, by iteratively updating the state transition probabilities and maintenance strategies of each sub-unit, it effectively approximates the predictive maintenance strategy for the multi-machine pipeline. This application solves the problem of the "curse of dimensionality" in MDP solutions caused by the explosive growth of the system's state space and action space as the number of production devices and buffer capacity increases, leading to difficulties in solving dynamic programming. This application effectively handles the maintenance decision-making problem of large-scale multi-machine pipeline systems, improving the feasibility and efficiency of the solution process.

[0017] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the industrial internet-based equipment maintenance strategy determination device, or it may be packaged separately from the processor of the industrial internet-based equipment maintenance strategy determination device; this application does not impose any limitations on this.

[0018] The descriptions of the second, third, ... and fifth aspects in this application can be referred to the detailed description of the first aspect; and the beneficial effects of the descriptions of the second, third, ... and fifth aspects can be referred to the analysis of the beneficial effects of the first aspect, which will not be repeated here.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description.

[0020] It is understood that before using the technical solutions disclosed in the various embodiments of this application, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this application in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This application provides a schematic diagram of the composition of a production line according to an embodiment of the present application. Figure 2 A first flowchart illustrating a method for determining equipment maintenance strategies based on the Industrial Internet, provided in an embodiment of this application; Figure 3 This application provides a second flowchart illustrating a method for determining equipment maintenance strategies based on the Industrial Internet. Figure 4 Comparison results of four maintenance methods provided in the embodiments of this application; Figure 5 A schematic diagram of a device for determining equipment maintenance strategies based on the Industrial Internet, provided in an embodiment of this application; Figure 6 This is a block diagram of an electronic device used to implement a method for determining equipment maintenance strategies based on the Industrial Internet, as described in the embodiments of this application. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0024] It should be noted that the terms "first," "second," "target," and "original," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein. Furthermore, the terms "comprising," "having," and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] Figure 2 This is a first flowchart illustrating a method for determining equipment maintenance strategies based on the Industrial Internet, provided in an embodiment of this application. This embodiment is applicable to scenarios where predictive maintenance decisions are made for production lines with multiple machines (i.e., production equipment). The method for determining equipment maintenance strategies based on the Industrial Internet provided in this embodiment can be executed by an device for determining equipment maintenance strategies based on the Industrial Internet, as provided in this application. This device can be implemented through software and / or hardware and integrated into the electronic device executing the method.

[0026] See Figure 2 The method in this embodiment includes, but is not limited to, the following steps: S110. Divide multiple production equipment and multiple buffer zones in the production line into multiple sub-equipment units according to the upstream and downstream sequence of the production line.

[0027] In this embodiment, a production line can refer to a production organization that processes raw materials into finished products step by step through a series of production equipment and operations in a specific sequence. For example, the production of electronic products uses production lines, from the assembly of components to the completion of the appliance, involving multiple stages and equipment operations.

[0028] In this embodiment, "multiple production devices" can refer to various devices with different functions existing on a production line, each undertaking a specific production task. For example, on an electronic product production line, there may be welding equipment, testing equipment, assembly equipment, etc., and these devices work together to complete the product manufacturing process.

[0029] In this embodiment, the buffer can be a temporary storage area set up in the production line. Its function is to adjust the production rhythm differences between different stages of the production process. When the production speed of upstream production equipment is inconsistent with that of downstream production equipment, the buffer can temporarily store the intermediate products produced to avoid interruption of the production process. For example, when the upstream equipment produces faster, but the downstream equipment has limited processing capacity, the intermediate products can be stored in the buffer and wait for the downstream equipment to process them.

[0030] In this embodiment, the upstream and downstream sequence refers to the production sequence of a production line, where raw materials enter from one end, pass through various production equipment and buffer zones in sequence, and are gradually processed into finished products that are output from the other end. In this process, the equipment and buffer zones closer to the raw material entry end are in the upstream position, while those closer to the finished product output end are in the downstream position.

[0031] In this embodiment of the application, multiple production devices and buffers are grouped according to the upstream and downstream sequence of the production line, and each group constitutes a sub-device unit. Each sub-device unit includes at least one buffer and one production device, and each sub-device unit also includes a virtual device composed of all the production devices located upstream of this production device or another production device.

[0032] Understandably, the first sub-device unit among multiple sub-device units includes the first production device, the first buffer corresponding to the first production device, and the second production device. Sub-device units that are not the first among multiple sub-device units include the previous virtual device, the previous buffer corresponding to the previous production device, and the current production device. The previous virtual device is a virtual device composed of all production devices upstream of the current production device.

[0033] like Figure 1 The diagram shows the composition of a production line. Taiwan production equipment and It consists of several buffers, and the buffers are used The buffer capacity is also expressed as... Production equipment Indicated by the reference numerals in the figure. to This indicates M production equipment, as shown in the attached diagram. to This represents M-1 buffers, decomposing the production pipeline into M-1 two-machine sub-equipment units. Indicate. For example: the first production equipment. The first buffer zone corresponding to the first production equipment and the second production equipment Divided into a sub-device unit, denoted as It can be Treat it as a virtual device, denoted as Then the virtual device Buffer and production equipment Sub-device units By analogy, we can obtain the M-1 sub-device unit.

[0034] S120. Determine the initial maintenance strategy for each sub-device unit.

[0035] Production equipment in this embodiment have A series of progressively deteriorating operating states can be represented by... Indicates production equipment The running status, that is . Specifically, Indicates production equipment For a healthy state, Indicates production equipment This is a fault condition. If the production equipment... In a non-faulty state, it is The probability of producing a qualified product is [a certain percentage]. The probability of producing a defective product is [missing information]. When the production equipment [missing information]... When in a faulty state, no workpieces can be produced.

[0036] In this embodiment, for the first sub-equipment unit, determining the initial maintenance strategy corresponding to the sub-equipment unit includes: constructing a Markov Decision Process (MDP) model for the first production equipment, the first buffer, and the second production equipment; and calculating the initial maintenance strategies for the first and second production equipment based on the MDP model. The maintenance strategy in this application is a predictive maintenance strategy in production.

[0037] In this embodiment, for a non-first sub-device unit, determining the initial maintenance strategy corresponding to the sub-device unit includes: constructing an MDP model for the previous virtual device, the previous buffer, and the current production device; and calculating the initial maintenance strategy for the current production device based on the MDP model. Similarly, based on the two-machine pipeline MDP problem, the initial maintenance strategy is sequentially implemented from the production device... To production equipment Analysis was conducted to obtain the initial maintenance strategy for all production equipment.

[0038] For example, first, all possible operating states of the production equipment and buffer within the sub-device unit are determined. For instance, the production equipment might be in states such as "normal operation," "minor fault," or "major fault"; the buffer might be in states such as "zero storage," "half storage," or "full storage." Then, actions that can be taken against them are defined, such as performing preventative maintenance on the production equipment, adjusting its production speed, or performing product input or output operations on the buffer. Next, the reward for each action is determined, such as the production revenue from normal equipment operation or the cost of maintaining the equipment. Finally, the state transition probability is calculated, i.e., the likelihood of transitioning from one operating state to another after taking a certain action. By determining these factors, an MDP model for the first sub-device unit is constructed.

[0039] Initialize the state transition matrix and reward function for each production machine. The reward function consists of three parts: output reward, work-in-process inventory cost, and maintenance cost. Considering output revenue, work-in-process cost, and maintenance cost, construct the reward function to maximize system revenue. Analyze the impact of factors such as machine condition deterioration, product quality scrap, and maintenance activities on the system state transition process to construct an MDP model. Using the constructed MDP model, calculate the optimal value function for each state using specific algorithms (such as value iteration algorithms and policy iteration algorithms) until convergence yields the optimal maintenance strategy, denoted as the initial maintenance strategy. The maintenance strategy refers to the plan for determining when and how to maintain the equipment under different equipment states. The goal of solving the predictive maintenance dynamic decision problem is to find the maximum cumulative expected discount reward and the optimal maintenance strategy. For example, calculate whether it is more advantageous to perform preventative maintenance immediately or continue operation for a period of time before maintenance when the production equipment is in normal operation (considering long-term reward maximization).

[0040] It should be noted that for the first sub-device unit Production equipment Buffer and production equipment Instead of using aggregation methods, the optimal initial maintenance strategy can be directly solved using two-machine MDP. For non-first sub-device units Due to virtual machines The decision space of the MDP is limited by using only two states, working and fault, to characterize the dynamic process. and maintenance strategies It only includes production equipment. Maintenance decisions. The initial maintenance strategy is expressed by the following formula (1). The solution process: (1) In the formula, This indicates the initial maintenance strategy for the first and second production devices within the first sub-equipment unit. This represents the MDP model corresponding to the first sub-device unit. This represents the decision space regarding the first and second production equipment. This represents the first state transition probability matrix corresponding to the first sub-device unit. This represents the initial maintenance strategy for the (m+1)th production equipment in the m-th sub-equipment unit. This represents the MDP model corresponding to the m-th sub-device unit. This represents the decision space regarding the (m+1)th production equipment. Let represent the first state transition probability matrix corresponding to the (m+1)th sub-device unit.

[0041] This step uses Markov Decision Process (MDP) to model the maintenance decisions of a multi-machine pipeline system. The multi-machine pipeline is broken down into multiple two-machine sub-units, and a dynamic programming algorithm is used to solve for the initial maintenance strategy. Specifically, for a two-machine pipeline, a predictive maintenance decision model is established based on Markov Decision Process, and the optimal initial maintenance strategy is obtained using dynamic programming. For a multi-machine pipeline, an approximate solution for the predictive maintenance strategy is obtained based on an aggregation iterative method of the two-machine pipeline maintenance strategy.

[0042] S130. Determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit.

[0043] In this embodiment of the application, the process of determining the first state transition probability matrix is ​​specifically as follows: for production equipment Buffer and production equipment Sub-device units Initial maintenance strategy is obtained by utilizing the MDP problem in a two-machine pipeline. Based on strategy The first state transition probability matrix can be obtained using existing calculation methods, denoted as... For virtual machines Buffer and production equipment Sub-device units Initial maintenance strategy is obtained by utilizing the MDP problem in a two-machine pipeline. Based on strategy The first state transition probability matrix can be obtained using existing calculation methods, denoted as... .for This pattern continues until the production equipment... The process ends, yielding the first state transition probability matrix for each sub-device unit under the corresponding initial maintenance strategy.

[0044] Specifically, determining the second state transition probability matrix of the virtual device corresponding to the sub-device unit includes: determining the steady-state probability distribution of the sub-device unit; calculating the failure probability of the sub-device unit being in a fault state and the repair probability of the sub-device unit transitioning from a fault state to an operating state based on the steady-state probability distribution and the product quality data of the current production equipment; and constructing the second state transition probability matrix of the virtual device based on the failure probability and the repair probability.

[0045] In one embodiment, the idea behind algorithmic state aggregation is to represent the aggregation performance of the aggregation block using a running state and a repair state, with probability... This represents the failure probability of the aggregate block. This represents the repair probability of the aggregated block. Through such aggregation, the sub-device unit can be modeled using a geometric reliability model, and the second state transition probability matrix of the virtual machine is represented by the following formula (2). Represented as: (2) For failure probability and repair probability In this embodiment, the solution involves first solving for the system state transition matrix of the subsystem to be aggregated, and then obtaining the steady-state probability distribution through the equilibrium equations. Then, the failure probability is calculated using the steady-state probability and the previously derived formula. Repair probability and the internal blocking rate of the aggregate block The solution successfully aggregated the states of the subsystem.

[0046] In this embodiment, the determination of the second state transition probability matrix takes into account the product quality data of the current production equipment, so that the state deterioration of the production equipment and the quality of the products can be comprehensively considered.

[0047] S140. Update the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix.

[0048] Specifically, updating the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix includes: obtaining the second internal blocking rate of each sub-device unit caused by downstream production equipment from the first internal blocking rate corresponding to each sub-device unit; updating the state transition probability corresponding to the preset equipment maintenance method based on the second internal blocking rate to obtain a new first state transition probability matrix.

[0049] The first internal blocking rate is generated by upstream production equipment to the sub-equipment unit, and the second internal blocking rate is generated by downstream production equipment to the sub-equipment unit. The determination process for the second internal blocking rate can be as follows: for the current sub-equipment unit (e.g., ... In this regard, the first internal blocking rate can be obtained through step S130. The next sub-device unit (such as The corresponding first internal blocking rate As the current sub-device unit The second internal blocking rate.

[0050] The first state transition probability matrix consists of the state transition probabilities corresponding to different equipment maintenance methods under the initial maintenance strategy. The preset equipment maintenance methods include no maintenance activities and predictive maintenance.

[0051] The above step S120, which determines the initial maintenance strategy, only considers the starvation effect of upstream machines on downstream machines, and does not consider the blocking effect of downstream machines on upstream machines. This step considers the blocking effect and uses the machine blocking rate to update the state transition probabilities of machines under no maintenance activity and predictive maintenance.

[0052] When maintenance decisions are not implemented, production equipment From a healthy state to a faulty state, Representing state Transition to state The probability of, where When predictive maintenance is performed, production equipment To recover from a deteriorated state to a healthier state. Representing state Transition to state The probability of, where .

[0053] Since updating the machine state transition probabilities will cause a change in the state transition probability matrix of the MDP, the new first state transition probability matrix after the change is represented by the following formula (3): (3) In the formula, This represents the new first state transition probability matrix corresponding to the m-th sub-device unit. This represents the update function that updates the state transition probabilities. This represents the first state transition probability matrix corresponding to the m-th sub-device unit. This represents the second internal blocking rate corresponding to the (m+1)th sub-device unit.

[0054] S150. Based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix, the initial maintenance strategy is iteratively updated to obtain the target maintenance strategy for each production device.

[0055] In this embodiment, updating the first state transition probability matrix of the production equipment causes updates to the machine's blocking rate and the MDP state transition probability matrix, thereby changing the optimal predictive maintenance strategy. Simultaneously, when the optimal predictive maintenance strategy changes, the blocking rate of the production equipment will also change. Therefore, this embodiment, based on the initial strategy, uses an aggregation iterative algorithm to satisfy the iterative update requirements of the predictive maintenance strategy and the blocking rate, obtaining the target maintenance strategy for each production equipment.

[0056] In practical applications, the operation process of this application can be as follows: Before the start of assembly line production, system initialization is performed, including sensor installation, database setup, and central processing system configuration. During production, sensors monitor the machine's operating status and product quality in real time and transmit the data to the central processing system. Based on the collected data, the central processing system dynamically formulates the optimal predictive maintenance strategy using an MDP model and dynamic programming algorithm. Maintenance personnel or robots execute corresponding maintenance activities, such as predictive maintenance or reactive repair, based on the maintenance decisions. After the maintenance activities are completed, the maintenance results are fed back to the system to evaluate the effectiveness of the maintenance strategy and provide a reference for subsequent maintenance decisions. Through this cyclical process of real-time monitoring, dynamic decision-making, and feedback adjustment, this invention can ensure the efficient operation of the assembly line system, minimize unexpected downtime, and improve operational efficiency and product quality.

[0057] The technical solution provided in this embodiment divides multiple production devices and buffers in a production pipeline into multiple sub-device units according to the upstream and downstream sequence of the production pipeline; determines the initial maintenance strategy corresponding to each sub-device unit; determines the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy; determines the second state transition probability matrix of the virtual device corresponding to each sub-device unit; and calculates the first internal blocking rate of each sub-device unit; updates the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix; and iteratively updates the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device. For multi-machine pipelines, the increase in the number of production devices and buffer capacity will cause an explosive growth in the system state space and action space, making the solution of MDP encounter the "curse of dimensionality" problem, resulting in difficulties in solving dynamic programming. To this end, this application proposes an aggregation iteration method to decompose the multi-machine pipeline into multiple two-machine sub-device units and solve the optimal maintenance strategy independently for each unit. Then, by iteratively updating the state transition probabilities and maintenance strategies of each sub-unit, an effective approximate solution for the predictive maintenance strategy of a multi-machine pipeline is obtained. The innovation of this method lies in its ability to effectively handle the maintenance decision-making problem of large-scale multi-machine pipeline systems, improving the feasibility and efficiency of the solution process.

[0058] The method for determining equipment maintenance strategies based on the Industrial Internet, provided in the embodiments of this application, is further described below. Figure 3 This is a second flowchart illustrating a method for determining equipment maintenance strategies based on the Industrial Internet, provided in an embodiment of this application. This embodiment optimizes the above embodiments, specifically by providing a detailed explanation of the process of iteratively updating the initial maintenance strategy to obtain the target maintenance strategy.

[0059] See Figure 3 The method in this embodiment includes, but is not limited to, the following steps: S210. For each sub-device unit, update the initial maintenance strategy based on the new first state transition probability to obtain the intermediate maintenance strategy.

[0060] In this embodiment, updating the first state transition probability matrix of the production equipment will cause a change in the optimal predictive maintenance strategy. Based on the new first state transition probability, the initial maintenance strategy can be updated using the above formula (1) to obtain the intermediate maintenance strategy.

[0061] S220. During the process of updating the second state transition probability matrix based on the intermediate maintenance strategy and the new first state transition probability, a new first internal blocking rate is obtained.

[0062] In this embodiment of the application, the focus is on production equipment. Buffer and production equipment Sub-device units The intermediate maintenance strategy is obtained by updating the initial maintenance strategy. Based on strategy and the new first state transition probability The virtual machine is obtained based on the following formula (4). The new second state transition probability matrix For virtual machines Buffer and production equipment Sub-device units The intermediate maintenance strategy is obtained by updating the initial maintenance strategy. Based on strategy and the new first state transition probability The virtual machine is obtained based on the following formula (4). The new second state transition probability matrix Production equipment The new first internal blocking rate ..for This pattern continues until the production equipment... End, and obtain the new first internal blocking rate corresponding to each sub-device unit.

[0063] (4) In the formula, This is the new second state transition probability matrix for the first virtual machine. This indicates the intermediate maintenance strategy for the first and second production devices within the first sub-equipment unit. This represents the new first state transition probability matrix corresponding to the first sub-device unit. Let the new second state transition probability matrix be the m-th virtual machine. The new first internal blocking rate corresponds to the m-th sub-device unit. A function to calculate the second state transition probability matrix for the m-th sub-device unit. This represents the intermediate maintenance strategy for the (m+1)th production equipment in the m-th sub-equipment unit. This represents the new first state transition probability matrix corresponding to the m-th sub-device unit.

[0064] S230. If the difference between the new first internal blocking rate and the first internal blocking rate corresponding to each sub-device unit does not meet the preset convergence threshold, then continue to update the initial maintenance strategy for each sub-device unit based on the new first state transition probability, the first internal blocking rate and the second state transition probability matrix to obtain the intermediate maintenance strategy and the new first internal blocking rate for each sub-device unit.

[0065] In this embodiment, the value of the preset convergence threshold can be determined according to the actual application. The number of iterations can also be set.

[0066] S240. If the difference between the new first internal blocking rate and the first internal blocking rate satisfies the preset convergence threshold, then the intermediate maintenance strategy is taken as the target maintenance strategy.

[0067] The technical solution provided in this embodiment addresses the "curse of dimensionality" problem encountered in solving Multi-Machine Pipelines (MDPs) due to the explosive growth of the system's state space and action space as the number of production devices and buffer capacity increase, leading to difficulties in dynamic programming solutions. To address this, this application proposes an aggregation iterative method that decomposes the multi-machine pipeline into multiple two-machine sub-units and independently solves the optimal maintenance strategy for each unit. Then, by iteratively updating the state transition probabilities and maintenance strategies of each sub-unit, an effective approximate solution for the predictive maintenance strategy of the multi-machine pipeline is obtained. The innovation of this method lies in its ability to effectively handle maintenance decision-making problems in large-scale complex systems while maintaining the feasibility and efficiency of the solution process.

[0068] This application verifies the effectiveness of the proposed predictive maintenance decision-making method through numerical experiments comparing the following three maintenance methods. For ease of description, the pipeline that only uses reactive maintenance is defined as the baseline pipeline, denoted by BL. The proposed predictive maintenance decision-making method is abbreviated as Predictive Maintenance (PM) method, denoted by PM. The three other maintenance methods are described in detail below: (1) Time-based maintenance (TM): This is a traditional maintenance method that performs maintenance at fixed time intervals without considering the actual condition of the machine. This method is simple and easy to implement, but it may not be able to adapt to dynamic changes in the production process.

[0069] (2) Machine State-based Maintenance (MSM): This method determines the timing of maintenance based on the machine's operating status, and uses simulation experiments to maximize system benefits and determine the state threshold for each machine to carry out maintenance activities.

[0070] (3) Buffer State-based Maintenance (BSM): This method adjusts maintenance strategies based on the work-in-process level of the buffer, taking into account the balance of the production process. Simulation experiments are used to determine the buffer threshold for implementing maintenance activities on each machine with the goal of maximizing system benefits.

[0071] To conduct a comparative experiment, 5000 production lines were randomly generated, and the corresponding parameter selection ranges are shown in Tables 1 and 2: Table 1 Relevant parameters of the production line Table 2. Machine state transition probability parameters under different maintenance decisions. For the 5000 generated pipelines, the proposed PM method converged, with a maximum convergence time of 4 minutes and a maximum number of iterations of 22. Each pipeline operated on a three-shift system, with each shift lasting 8 hours. Under each maintenance method, the pipeline underwent a 10-day warm-up period, followed by 100 days of operation. Throughput (TH), work-in-process (WIP), and maintenance cost (MC) were statistically analyzed. For each maintenance method, the deviations of performance indicators from those obtained from the BL (Boolean Process) method were calculated, along with the deviations in output, WIP levels, and maintenance costs for each pipeline (using...). The calculation formula (represented by the formula) is as follows: , , , in, and These represent the performance index values ​​obtained from different maintenance methods and BL, respectively.

[0072] like Figure 4 The figure shows a comparison of four maintenance methods, specifically the average improvement of the pipeline by the PM, TM, MSM, and BSM methods compared to BL. Figure 4 As shown. The specific experimental results are as follows: (1) In terms of output revenue, all four maintenance methods can effectively improve the output revenue of the production line. Among them, the PM method brings the largest output revenue, which is 34.88%. (2) In terms of work-in-process cost, compared with the other three methods, the PM method achieves the largest percentage reduction in work-in-process cost, which is 37.12%. (3) In terms of maintenance cost, the PM method also brings the lowest maintenance cost, with a reduction percentage of 36.35%.

[0073] Experimental results show that the method of this application demonstrates superiority in terms of output revenue, work-in-process cost, and maintenance cost. A comprehensive comparison of four maintenance methods reveals that the method of this invention, based on system status, can more effectively identify opportunities for maintenance interventions in the production line, thereby effectively increasing system output while reducing maintenance and work-in-process costs.

[0074] Figure 5 A schematic diagram of a device for determining equipment maintenance strategies based on the Industrial Internet, provided in an embodiment of this application, is shown below. Figure 5 As shown, the device 500 may include: The equipment unit division module 510 is used to divide multiple production equipment and multiple buffers in the production line into multiple sub-equipment units according to the upstream and downstream sequence of the production line. The initial strategy determination module 520 is used to determine the initial maintenance strategy corresponding to each of the sub-device units; The data processing module 530 is used to determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit. The data update module 540 is used to update the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix. The target strategy determination module 550 is used to iteratively update the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate and the second state transition probability matrix to obtain the target maintenance strategy for each production equipment.

[0075] In one embodiment, the first sub-device unit among the plurality of sub-device units includes the first production device among the plurality of production devices, the first buffer corresponding to the first production device, and the second production device. The non-first sub-device units among the plurality of sub-device units include the previous virtual device, the previous buffer corresponding to the previous production device, and the current production device. The previous virtual device is a virtual device composed of all production devices located upstream of the current production device.

[0076] In one embodiment, the data processing module 530 described above can be specifically used to: determine the steady-state probability distribution of the sub-device unit; calculate the failure probability of the sub-device unit being in a fault state and the repair probability of the sub-device unit transitioning from a fault state to an operating state based on the steady-state probability distribution and the product quality data of the current production equipment; and construct a second state transition probability matrix of the virtual device based on the failure probability and the repair probability.

[0077] In one embodiment, the first internal blocking rate is generated by the upstream production equipment to the sub-equipment unit, and the first state transition probability matrix is ​​composed of the state transition probabilities corresponding to different equipment maintenance methods under the initial maintenance strategy. In one embodiment, the data update module 540 described above can be specifically used to: obtain a second internal blocking rate of the sub-device unit caused by downstream production equipment from the first internal blocking rate corresponding to each sub-device unit; update the state transition probability corresponding to the preset equipment maintenance method based on the second internal blocking rate to obtain a new first state transition probability matrix.

[0078] In one embodiment, the target strategy determination module 550 can be specifically used to: update the initial maintenance strategy based on the new first state transition probability for each sub-device unit to obtain an intermediate maintenance strategy; obtain a new first internal blocking rate during the process of updating the second state transition probability matrix based on the intermediate maintenance strategy and the new first state transition probability; if the difference between the new first internal blocking rate and the first internal blocking rate for each sub-device unit does not meet a preset convergence threshold, then continue to update the initial maintenance strategy based on the new first state transition probability, the first internal blocking rate, and the second state transition probability matrix for each sub-device unit to obtain an intermediate maintenance strategy and a new first internal blocking rate for each sub-device unit; if the difference between the new first internal blocking rate and the first internal blocking rate meets the preset convergence threshold, then use the intermediate maintenance strategy as the target maintenance strategy.

[0079] In one embodiment, for the first sub-device unit, the initial strategy determination module 520 can be specifically used to: construct a Markov decision process model for the first production device, the first buffer, and the second production device; and calculate the initial maintenance strategy for the first production device and the second production device based on the Markov decision process model.

[0080] In one embodiment, for the non-first sub-device unit, the initial strategy determination module 520 can be specifically used to: construct a Markov decision process model for the previous virtual device, the previous buffer, and the current production device; and calculate the initial maintenance strategy for the current production device based on the Markov decision process model.

[0081] The device for determining equipment maintenance strategies based on the Industrial Internet provided in this embodiment can be applied to the device maintenance strategy determination method based on the Industrial Internet provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0082] Figure 6 This is a block diagram of an electronic device used to implement an industrial internet-based device maintenance strategy determination method according to embodiments of this application. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0083] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0084] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0085] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for determining equipment maintenance strategies based on the Industrial Internet.

[0086] In some embodiments, the Industrial Internet-based equipment maintenance strategy determination method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the Industrial Internet-based equipment maintenance strategy determination method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the Industrial Internet-based equipment maintenance strategy determination method by any other suitable means (e.g., by means of firmware).

[0087] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0088] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0089] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0090] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0091] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0092] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0093] Note that the above are merely preferred embodiments and technical principles applied in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. For example, those skilled in the art can use the various forms of processes shown above to reorder, add, or delete steps; the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solution of this application can be achieved, and no limitations are imposed herein.

[0094] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for determining equipment maintenance strategies based on the Industrial Internet, characterized in that, The method includes: The production line is divided into multiple sub-equipment units according to the upstream and downstream sequence of the production line. Determine the initial maintenance strategy for each of the sub-device units; Determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit; The first state transition probability matrix is ​​updated based on the first internal blocking rate to obtain a new first state transition probability matrix, including: obtaining a second internal blocking rate of the sub-device unit caused by the downstream production equipment from the first internal blocking rate corresponding to each sub-device unit; updating the state transition probability corresponding to the preset equipment maintenance method based on the second internal blocking rate to obtain a new first state transition probability matrix; wherein, the first internal blocking rate is caused by the upstream production equipment to the sub-device unit, and the second internal blocking rate is caused by the downstream production equipment to the sub-device unit; The initial maintenance strategy is iteratively updated based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device.

2. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 1, characterized in that, The first sub-device unit among the plurality of sub-device units includes the first production equipment among the plurality of production equipment, the first buffer corresponding to the first production equipment, and the second production equipment. The non-first sub-device units among the plurality of sub-device units include the previous virtual device, the previous buffer corresponding to the previous production equipment, and the current production equipment. The previous virtual device is a virtual device composed of all production equipment located upstream of the current production equipment.

3. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 2, characterized in that, Determining the second state transition probability matrix of the virtual device corresponding to the sub-device unit includes: Determine the steady-state probability distribution of the sub-device unit; Based on the steady-state probability distribution and the product quality data of the current production equipment, calculate the failure probability of the sub-equipment unit being in a fault state and the repair probability of the sub-equipment unit transitioning from a fault state to an operating state. The second state transition probability matrix of the virtual device is constructed based on the failure probability and the repair probability.

4. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 1, characterized in that, The first internal blocking rate is generated by the upstream production equipment to the sub-equipment unit, and the first state transition probability matrix is ​​composed of the state transition probabilities corresponding to different equipment maintenance methods under the initial maintenance strategy.

5. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 1, characterized in that, The initial maintenance strategy is iteratively updated based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device, including: For each of the sub-device units, the initial maintenance strategy is updated based on the new first state transition probability to obtain an intermediate maintenance strategy; During the process of updating the second state transition probability matrix based on the intermediate maintenance strategy and the new first state transition probability, a new first internal blocking rate is obtained; If the difference between the new first internal blocking rate and the first internal blocking rate corresponding to each of the sub-device units does not meet the preset convergence threshold, then the process of updating the initial maintenance strategy for each of the sub-device units based on the new first state transition probability, the first internal blocking rate and the second state transition probability matrix continues to be executed, so as to obtain the intermediate maintenance strategy and the new first internal blocking rate for each of the sub-device units. If the difference between the new first internal blocking rate and the first internal blocking rate satisfies the preset convergence threshold, then the intermediate maintenance strategy is taken as the target maintenance strategy.

6. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 2, characterized in that, For the first sub-device unit, determining the initial maintenance strategy corresponding to the sub-device unit includes: A Markov decision process model is constructed for the first production equipment, the first buffer, and the second production equipment. The initial maintenance strategies for the first and second production equipment are calculated based on the Markov decision process model.

7. The method for determining equipment maintenance strategies based on the Industrial Internet according to claim 2, characterized in that, For the non-first sub-device unit, determining the initial maintenance strategy corresponding to the sub-device unit includes: A Markov decision process model is constructed for the previous virtual device, the previous buffer, and the current production device; The initial maintenance strategy for the current production equipment is calculated based on the Markov decision process model.

8. A device for determining equipment maintenance strategies based on the Industrial Internet, characterized in that, The apparatus for implementing the equipment maintenance strategy determination method based on the Industrial Internet as described in claim 1, the apparatus comprising: The equipment unit division module is used to divide multiple production equipment and multiple buffers in the production line into multiple sub-equipment units according to the upstream and downstream sequence of the production line. An initial strategy determination module is used to determine the initial maintenance strategy corresponding to each of the sub-device units; The data processing module is used to determine the first state transition probability matrix of each sub-device unit under the corresponding initial maintenance strategy, determine the second state transition probability matrix of the virtual device corresponding to each sub-device unit, and calculate the first internal blocking rate of each sub-device unit. The data update module is used to update the first state transition probability matrix based on the first internal blocking rate to obtain a new first state transition probability matrix. The target strategy determination module is used to iteratively update the initial maintenance strategy based on the new first state transition probability matrix, the first internal blocking rate, and the second state transition probability matrix to obtain the target maintenance strategy for each production device.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores a computer program that is executed by the at least one processor, which enables the at least one processor to perform the industrial internet-based equipment maintenance strategy determination method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method for determining an industrial internet-based equipment maintenance strategy as described in any one of claims 1 to 7.

Images