A maintenance task sequencing optimization method for distributed cluster equipment

Abstract

The present application discloses a maintenance task sorting optimization method for distributed cluster equipment, which belongs to the field of intelligent operation and maintenance technology of complex systems. A general generating function model of a distributed cluster equipment system is established, and the node importance is calculated; then, a maintenance time matrix of an equipment path duration function and a maintenance level is constructed, and operation constraints and resource constraints are configured; then, a feasible decision variable solution set is constructed using the total maintenance time and node importance of the distributed cluster equipment, and the system steady-state availability is calculated as the objective function. The model is solved using an ant colony algorithm, and finally, the optimal distributed cluster equipment maintenance level and maintenance sequence decision information are obtained; the present application fully considers the type and layout location of the distributed cluster equipment, as well as the operation scenario and maintenance resource constraints of the cluster equipment, to provide decision support for on-site operation and maintenance management personnel to formulate the optimal maintenance plan.

Description

Technical Field

[0001] The present application belongs to the field of intelligent operation and maintenance technology for complex systems, and specifically relates to a maintenance task sorting optimization method for distributed cluster equipment. Background Art

[0002] With the rapid advancement of industrial technology, the structure of equipment and systems has become increasingly complex. In specialized industrial scenarios, large-scale equipment clusters exist, forming complex systems with diverse topologies. The daily maintenance and management of such complex equipment presents significant challenges, especially in distributed clusters, where the number of devices is large, the devices are widely distributed, and the interconnectedness and dependencies between different devices are strong. Traditional maintenance methods typically rely on predetermined periodic maintenance plans or perform maintenance only after a failure occurs. This model struggles to adapt to the dynamic needs of distributed clusters, potentially leading to wasted maintenance resources and delayed maintenance of critical equipment, increasing uncertainty and potential risks in system operation. Maintenance plans for distributed clusters require traversing all device nodes within a single cycle, and maintenance tasks are often subject to multiple constraints, including time, resources, and task dependencies. Furthermore, the wide distribution of cluster devices introduces significant uncertainty into the execution time of maintenance tasks, placing higher demands on both the formulation and execution of maintenance plans.

[0003] Current research mostly focuses on the maintenance optimization of single equipment, and there is still a lack of a systematic solution for the overall maintenance task sequencing and maintenance level of distributed cluster equipment. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, this application proposes a maintenance task sorting optimization method for distributed cluster equipment, which solves the sorting optimization problem of distributed cluster equipment in periodic maintenance tasks and provides auxiliary decision-making for the maintenance optimization of complex systems composed of distributed cluster equipment.

[0005] This application is implemented through the following technical solutions:

[0006] A maintenance task sequencing optimization method for distributed cluster equipment, comprising:

[0007] Obtain node type and location information based on the system topology structure composed of distributed cluster devices;

[0008] Establishing a general generating function model of the system topology structure and calculating the importance of each node in the system topology structure;

[0009] Construct the maintenance time matrix of equipment maintenance level and the path duration function of location path;

[0010] Configure operational and resource constraints for distributed cluster equipment maintenance activities;

[0011] The maintenance level and maintenance sequence are encoded as decision variables, and the total maintenance time and node importance of distributed cluster equipment are used to construct feasible solutions for the decision variables.

[0012] Calculate the availability of the system composed of distributed cluster equipment in a maintenance cycle as the objective function;

[0013] The optimization model is iteratively solved using an ant colony algorithm to obtain the optimal solution for the maintenance task sorting of distributed cluster equipment, and the maintenance level and maintenance sequence are output after decoding; wherein the optimization model is composed of the objective function, the feasible solution set and the constraint conditions.

[0014] In some embodiments, obtaining the type and location information of the node according to the system topology structure formed by the distributed cluster devices includes:

[0015] According to the system topology structure composed of distributed cluster devices, input ports and output ports are divided, and input and output ports are matched to form an input and output port set of distributed cluster devices;

[0016] Obtain the type and location information of the distributed cluster device, using latitude and longitude coordinates or fixed reference system coordinates to form a location set with node coding.

[0017] In some embodiments, the step of establishing a universal generating function model of the system topology structure and calculating the importance of each node in the system topology structure includes:

[0018] Based on the input and output port sets of the distributed cluster devices, the devices in each combination within the set are divided into multi-state degradation performance categories, a universal generating function model for each combination is established, and based on the universal generating function model for each combination, a universal generating function model for the system topology is obtained;

[0019] Calculating the overall performance state of the system topology structure according to a universal generating function model of the system topology structure;

[0020] The importance value of each node is calculated based on the overall performance status of the system topology.

[0021] In some implementations, the importance value of each node is calculated based on the overall performance status of the system topology structure, and the Fussel-Vesely algorithm is used to calculate the importance value of each node.

[0022] In some embodiments, constructing a maintenance duration matrix of equipment maintenance levels and a path duration function of location paths includes:

[0023] Each distributed cluster device corresponds to multiple maintenance levels, and each maintenance level corresponds to a maintenance duration. The maintenance durations of all maintenance levels of all distributed cluster devices constitute the maintenance level duration matrix.

[0024] The ratio of the travel time from when the maintenance personnel performs the current task to when they perform the next task during the system operation interval to the movement speed of the maintenance personnel is calculated as the path duration function.

[0025] In some embodiments, the operational constraints and resource constraints for configuring distributed cluster equipment maintenance activities include:

[0026] Based on the operational requirements of distributed cluster equipment in different industrial scenarios, configure resource constraints for maintenance activities, including: maintenance costs should be less than or equal to the maximum cost resource;

[0027] Based on the operational requirements of distributed cluster equipment in different industrial scenarios, operational constraints for maintenance activities are configured, including: the time for each maintenance task combination should be less than or equal to the system task interval time.

[0028] In some embodiments, encoding the maintenance level and maintenance order as decision variables and constructing a feasible solution set of the decision variables using the total maintenance duration and node importance of the distributed cluster devices may include:

[0029] According to the number of distributed cluster devices, maintenance level and system task interval, binary coding technology is used to encode the decision variables of maintenance level and maintenance sequence;

[0030] Based on the distribution location and constraints of the distributed cluster devices, the first maintenance task in the task list is set as the current task;

[0031] Get the maintenance duration corresponding to the current task. Starting from the current task location, get the maintenance duration and path duration of all remaining tasks and calculate the total duration required for each remaining task.

[0032] Calculate the difference between the maintenance time of the current task and the system task interval time;

[0033] Compare the total duration of all remaining tasks with the difference, sort the tasks that are smaller than the difference according to the importance of the nodes, and extract the most important tasks from the sorting to combine them;

[0034] Calculate the difference between the total duration of this combined task and the system task interval time, and return to the previous step until the total duration of the remaining tasks does not meet the condition of being less than the difference. This completes the combination of the first maintenance task and other tasks.

[0035] Delete the combined tasks from the task list, then read the task with the earliest maintenance order among the remaining tasks in the task list as the current task, return to the step of obtaining the maintenance time corresponding to the current task, and perform the next task combination until there are no tasks in the task list, that is, the combination sorting of all tasks is completed. The decision variable encoding after the combination sorting is a feasible solution constructed.

[0036] In some embodiments, the availability of the system composed of the computing distributed cluster devices in a maintenance cycle is used as an objective function, including:

[0037] Calculate the total maintenance time of distributed cluster equipment based on the encoding of feasible solutions;

[0038] The availability index of the system composed of distributed cluster devices is calculated based on the total maintenance time of the distributed cluster devices. The availability index is equal to the ratio of the difference between the total maintenance time and the total maintenance time to the total maintenance time.

[0039] In some embodiments, the iteratively solving the optimization model includes:

[0040] In the initialization phase, a real number of the task number is randomly generated, then converted into a binary code for initialization, and an initial feasible solution set is generated;

[0041] Based on the initial feasible solution set, iterative optimization is performed. Each iterative optimization generates a new feasible solution set based on the optimization result of the previous iterative optimization. Then, using the minimum unavailability as the optimization goal, the optimal solution of the current iteration is searched until the iteration stopping condition is met, and the final decision optimal solution is obtained. The unavailability is equal to 1 minus the availability index.

[0042] In some embodiments, the decision variable is coded from 1 to N Groups of binary numbers, each group of binary numbers exists Position, front The digits are the maintenance level code of the equipment, and the The bits are the maintenance sequence codes for the equipment; w The highest maintenance level number. N is the number of devices, is the rounding function.

[0043] The present application proposes a maintenance task sorting optimization method for distributed cluster equipment. Based on the system topology and equipment type information of the distributed cluster equipment, and according to the multi-state degradation performance characteristics of the node equipment, a general generating function model of the system topology is established, and the node importance is calculated; then, based on the node position information of the distributed cluster equipment, the equipment path duration function and the maintenance level maintenance duration matrix are constructed, and the operation constraints and resource constraints are configured according to the specific usage scenarios of the distributed cluster equipment; the total maintenance time and node importance of the distributed cluster equipment are used to construct a feasible decision variable solution set, and the system steady-state availability is calculated as the objective function. The ant colony algorithm is used to solve the model, and finally the optimal distributed cluster equipment maintenance level and maintenance sequence decision information are obtained; the present application fully considers the type and layout location of the distributed cluster equipment, as well as the operation scenario and maintenance resource constraints of the cluster equipment, to provide decision support for on-site operation and maintenance management personnel to formulate the optimal maintenance plan, and to achieve the maximum operation availability of the distributed cluster equipment, thereby maximizing the feasibility of the task. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation of the embodiments of the present application. In the drawings:

[0045] Figure 1 A schematic flow chart of the optimization method proposed in the embodiment of this application;

[0046] Figure 2 A schematic diagram of the system principle structure proposed in the embodiment of this application;

[0047] Figure 3 This is a schematic diagram of the electronic device structure proposed in an embodiment of the present application;

[0048] Figure 4 A schematic diagram of a computer-readable storage medium proposed in an embodiment of the present application;

[0049] Reference numerals and corresponding component names:

[0050] 200-Maintenance task sorting optimization system; 201-Information acquisition module; 202-Importance calculation module; 203-Duration module; 204-Constraint configuration module; 205-Feasible solution construction module; 206-Objective function construction module; 207-Iterative optimization module; 300-Electronic device; 310-Memory; 320-Processor; 311-Computer program A; 400-Computer-readable storage medium; 411-Computer program B. DETAILED DESCRIPTION

[0051] In order to make the objectives, technical solutions and advantages of this application more clear, the present application is further described in detail below in conjunction with examples and drawings. The schematic implementation methods of this application and their descriptions are only used to explain this application and are not intended to limit this application.

[0052] Example: Current research focuses on the maintenance optimization of single devices, lacking technology for optimizing the overall maintenance of distributed cluster devices. To address this, this example proposes a method for optimizing the sequencing of maintenance tasks for distributed cluster devices. Based on the system topology composed of distributed cluster devices, a general generating function model for the overall performance status of distributed cluster devices is established, and the importance of each node in the system topology is calculated. A maintenance duration matrix for device maintenance levels and a path duration function for location paths are then constructed. Operational constraints and maintenance resource constraints for distributed cluster devices are configured. The maintenance levels and maintenance sequences are encoded as decision variables, and a feasible set of decision variable solutions is constructed using the total maintenance duration and importance of distributed cluster devices. The system steady-state availability probability is then calculated as the objective function, completing the construction of the optimization model. An ant colony algorithm is used to iteratively solve the optimization model, ultimately obtaining the optimal solution for sequencing maintenance tasks for distributed cluster devices. After decoding, the maintenance levels and maintenance sequences are output. This method solves the problem of optimizing the sequencing of periodic maintenance tasks for distributed cluster devices, providing auxiliary decision support for the maintenance sequencing of complex systems composed of distributed cluster devices.

[0053] like Figure 1 As shown, the method proposed in this embodiment includes the following steps:

[0054] Step 110: Obtain the type and location information of the node according to the system topology structure formed by the distributed cluster devices.

[0055] Furthermore, the implementation process of step 110 is as follows:

[0056] Step 111: divide the input ports and output ports according to the system topology structure of the distributed cluster devices, perform input and output port matching, and form an input and output port set of the distributed cluster devices. Io :

[0057] Io ={( I 1, O 1),( I 1, O 2),…,( I 1, O n ),( I 2, O 1),( I 2, O 2),…,( I 2,O n ),...,( I m , O 1),( I m , O 2),…,( I m , O n )};

[0058] in, I is the input port; m is the number of input ports; O is the output port; n is the number of output ports; ( I i , O j ) is the i ( i =1,2,…, m ) Enter the j ( j =1,2, n ) a combination of output ports; Io It represents the combination of all input and output ports, which can be used to draw the topological connection of distributed cluster devices.

[0059] This embodiment further illustrates this step using a railway station signal distributed cluster device as an example. Railway station signal devices are arranged in a station yard topology within the station yard layout, exhibiting the characteristics of distributed cluster devices. Step 111 divides station entrances and exits based on the station yard node topology structure formed by the railway station signal distributed cluster device, performs entry and exit port matching, and forms a set of input and output ports for the railway station signal cluster device.

[0060] It should be noted that, in this embodiment, the normal station train entry situation is considered, and the special situation of changing the line running direction is not considered for the time being, that is, it is assumed that the properties of the input and output ports of the distributed cluster device will not change.

[0061] Step 112: Obtain the type and location information of the distributed cluster device, using latitude and longitude coordinates or fixed reference coordinates to form a location set with node numbers. Local:

[0062] Local ={( x 1, y 1),( x 2, y 2),…,( x N, y N )}

[0063] in, x is the horizontal coordinate of the device; y is the vertical coordinate of the device; N is the number of devices; ( x i , y i ) indicates the i ( i =1,2,… N ) device's location coordinates.

[0064] Specifically, for distributed clustered railway station signal equipment, the equipment types acquired in step 112 include track circuits, signal machines, and switch machines. These are treated as equivalent signal node devices. The horizontal and vertical coordinates of the signal equipment are measured using the railway station coordinate system, with the signal tower as the origin.

[0065] This embodiment obtains different input and output port combinations based on distributed cluster devices, implements hierarchical topological connection of cluster devices, obtains device node location information, forms corresponding sets, and provides basic information support for performance modeling and maintenance optimization of distributed cluster devices.

[0066] Step 120: Establish a general generating function model of the system topology structure and calculate the importance of each node in the system topology structure.

[0067] Furthermore, the implementation process of step 120 is as follows:

[0068] Step 121 , based on the input and output port sets of the distributed cluster devices, perform multi-state degradation performance division on the devices in each combination within the set, and establish a general generating function model for each combination, as shown below:

[0069]

[0070] in, For the collection ( I i , O j ) General generating function model under combined device topology connection, ; operators between models of universal generating functions composed of different combinations; A general generating function model for representing system topology.

[0071] Step 122: Calculate the overall performance status of the system topology structure based on the general generating function model of the system topology structure:

[0072]

[0073] in, The performance conversion function is used. According to the system's universal generating function model, the lowest state at which the system can maintain normal operation is selected as the performance threshold. The state probabilities after the expansion of all universal generating function models greater than this state are added together to obtain the current performance state value of the system. It is the overall performance status value of the system topology.

[0074] Step 123: Calculate the importance value of each device node based on the overall performance status of the system topology. This embodiment uses the Fussel-Vesely algorithm to calculate the importance value of each device node, as shown below:

[0075]

[0076] in, For the k Devices in is the importance value of the performance threshold, For the k The performance status value of the device when the performance is normal (assuming that the k When the performance of each device is normal, the overall performance status value of the system topology calculated using the above steps 121 and 122) For the k Equipment performance < The performance status value at the time (assuming the k Equipment performance < When the overall performance status value of the system topology structure is calculated using the above steps 121 and 122); the importance value of each node in the system topology structure is obtained in the range of [0,1].

[0077] This embodiment can use a general generating function model of the system topology structure of distributed cluster equipment to calculate the Fussel-Vesely importance of the node, provide a decision basis for establishing the maintenance optimization objective function and decision variable optimization process information, and provide decision information for optimizing the fitness function and optimization algorithm.

[0078] Step 130: construct a maintenance duration matrix of equipment maintenance levels and a path duration function of location paths.

[0079] Furthermore, the implementation process of step 130 is as follows:

[0080] Step 131: Different types of equipment have different status and performance. When performing maintenance operations on equipment, you can choose from the set maintenance levels and corresponding maintenance duration. The maintenance methods of all distributed cluster equipment constitute the maintenance level duration matrix. T m , expressed as:

[0081]

[0082] in, For the i ( i= 1,2,… N ) device select j ( j =1,2,… w ) level maintenance tasks, For the i ( i= 1,2,… N ) device select j ( j =1,2,… w ) The duration of the maintenance task, w The highest maintenance level number (the larger the maintenance level number, the higher the maintenance level).

[0083] In this embodiment, due to the existence of different types of devices, the status division of various devices is uneven. The matrix can be configured according to the device with the largest number of states, and the state values ​​of other types of devices that are insufficient are in T m The matrix is ​​replaced by 0, and the maintenance level selection constraints of the corresponding type of equipment are added.

[0084] Specifically, for distributed clustered signal equipment at railway stations, although the three types of equipment are considered equivalent signal equipment, different types of equipment have different maintenance times. For example, a signal machine can be configured with a minimum maintenance level of 0.3 hours, a track circuit requires 0.2 hours, and a switch machine requires 1 hour.

[0085] Step 132: Since distributed cluster devices are deployed in different locations, when maintenance personnel perform the first a task (current task) and go to execute the a When adding 1 task (next task), travel time is required, which can be expressed as:

[0086]

[0087] in, The path time required to execute the next task, is the horizontal coordinate of the current task, is the vertical coordinate of the current task, The horizontal coordinate for the next task, For the next task vertical coordinate, The movement speed of maintenance personnel. The feasible path and required time calculation function of the distributed cluster equipment (i.e., the ratio of the path distance to the moving speed is calculated as the path duration) can be used to calculate the path duration occupied by the maintenance task sequence and to optimize the more concentrated equipment points.

[0088] Specifically, for distributed signal cluster equipment at railway stations, the horizontal and vertical coordinates of the equipment are measured using the railway station coordinate system, with the signal tower as the origin. On-track maintenance of distributed signal cluster equipment at railway stations must be performed during the designated time window, thus subjecting it to strict time constraints. Furthermore, maintenance personnel can move at a speed of 10 km / h.

[0089] This embodiment constructs a maintenance duration matrix corresponding to the equipment maintenance level and uses the location information of the equipment nodes to obtain the path duration between tasks, thereby obtaining the total time required for maintenance personnel to perform tasks, that is, the system downtime. The steady-state availability index of the system can be indirectly obtained to perform task sorting optimization.

[0090] Step 140: Configure operational constraints and resource constraints for distributed cluster equipment maintenance activities.

[0091] Furthermore, the implementation process of step 140 is as follows:

[0092] Step 141: Configure resource constraints for maintenance activities based on the operational requirements of distributed cluster equipment in different industrial scenarios. ST z , as shown below:

[0093]

[0094] in, ST z It is a resource constraint item for the maintenance activity of distributed cluster equipment. f z The constraints of system operation must be met f Z For example, maintenance costs c z Must be less than the maximum cost resource c total , and other resource constraints, etc., can be configured differently according to different distributed cluster devices.

[0095] Specifically, for railway station signal distributed cluster equipment, according to its operational requirements, the configured resource constraint condition is that the maintenance cost of the railway station signal distributed cluster equipment should be less than or equal to the maximum cost resource, that is:

[0096]

[0097] in, is the maintenance cost of distributed cluster equipment, is the maximum value of cost resources.

[0098] Step 142: Configure operational constraints for maintenance activities based on the operational requirements of distributed cluster equipment in different industrial scenarios. ST y , as shown below:

[0099]

[0100] in, ST y Represents the operational constraints of distributed cluster equipment maintenance activities. f y The constraints of system operation must be met f Y For example, the time for each maintenance task combination t a Must be less than the system task interval t lim (i.e., window period), each device must be maintained at least once, and each maintenance task must be less than or equal to the number of maintenance work groups. Indicator constraints such as system reliability and availability, as well as other operational constraints, can be configured differently according to different distributed cluster devices.

[0101] Specifically, for distributed cluster equipment of railway station signals, according to its operational requirements, the configured operational constraints are that the time for each maintenance task combination should be less than or equal to the system task interval time, and each device must be maintained at least once.

[0102]

[0103] in, The time for each maintenance task combination, is the system task interval time, The maintenance frequency of any signal equipment.

[0104] This embodiment configures operational constraints and resource constraints for distributed cluster equipment maintenance activities, making the maintenance task sequencing optimization model more adaptable to industrial equipment in different scenarios and improving the generalization capability of the model.

[0105] In step 150 , the maintenance level and maintenance sequence are encoded as decision variables, and a feasible solution of the decision variables is constructed using the total maintenance time and node importance of the distributed cluster equipment.

[0106] Furthermore, the implementation process of step 150 is as follows:

[0107] Step 151: Based on the number of distributed cluster devices, maintenance level, and system task interval, binary coding technology is used to encode the maintenance level and maintenance sequence as decision variables. dec , as shown below:

[0108]

[0109] in, dec From 1 to N Groups of binary numbers, each group of binary numbers exists Position, front The digits are the maintenance level code of the equipment, and the The bits are the maintenance sequence codes for the equipment; All the above codes constitute the basic maintenance decision information for all equipment.

[0110] Specifically, for the distributed cluster equipment of railway station signals, assuming that there are 30 signal devices of three types, namely track circuits, signal machines, and switch machines, and each device has four states, then each decision variable has an 8-bit binary code.

[0111] Step 152, grouping the maintenance tasks according to the distribution location and time constraints of the distributed cluster equipment, so that as many equipment as possible can be maintained in a system skylight, therefore, the maintenance tasks in step 151 are grouped according to the distribution location and time constraints of the distributed cluster equipment, so as to maintain as many equipment as possible in a system skylight, thereby dec Further grouping optimization is performed to construct a feasible algorithm solution that meets actual operational requirements. The first maintenance task in the task list is set as the current task. The maintenance duration corresponding to the current task is obtained. Starting from the current task location, the maintenance duration (obtained from the maintenance duration matrix) and path duration (calculated using the path duration function) of all remaining tasks are obtained. The total duration required is calculated (the sum of the maintenance duration and path duration).

[0112] Step 153, calculate the difference between the current task duration and the system task interval time, compare the total duration required for all remaining tasks with the difference, and then sort the tasks that meet the conditions (less than the difference) according to the node importance, and then read the task with the greatest importance in the sorting and combine it (that is, sort the task after the current task).

[0113] Step 154 ​​calculates the difference between the total duration of the combined tasks and the system task interval time, and executes step 153 again until the total duration of the remaining tasks does not meet the conditions. At this point, the combination of the first task and the others is completed (that is, the optimal sorted combination of tasks that can be executed within the system task interval is completed).

[0114] In step 155, the combined tasks are deleted from the task list. The task with the earliest maintenance order among the remaining tasks in the task list is then read as the current task. The maintenance duration corresponding to the current task is obtained. Starting from the current task's location, the maintenance duration and path duration of all remaining tasks are calculated, and the total required duration is calculated. Steps 153 and 154 are repeated to complete the combination of the current task with the remaining tasks. Step 155 is repeated until there are no tasks in the task list, completing the combined sorting of the tasks. The decision variable encoding after this combined sorting constitutes a constructed feasible solution. Using this method, several feasible solutions can be constructed in the subsequent solution process, generating a feasible solution set.

[0115] Specifically, for the distributed cluster equipment of railway station signals, the difference should also be configured with a corresponding protection time, because protective measures need to be taken when going on the track in the railway station. Therefore, about 0.2 hours of protection time needs to be added during one on-track operation.

[0116] This embodiment uses decision variable encoding, including the maintenance level and maintenance sequence encoding of each device, to construct a feasible solution using the maintenance duration and node importance of distributed cluster devices. This allows the maintenance process of distributed cluster devices to execute the most important and optimal task combination within a limited system operation interval, providing an effective feasible solution combination scheme for the algorithm to search for the global optimal solution.

[0117] Step 160 , calculating the availability of the system composed of distributed cluster devices in a maintenance cycle as an objective function.

[0118] Furthermore, the implementation process of step 160 is as follows:

[0119] Step 161: Calculate the total maintenance time of the distributed cluster equipment based on the encoding of the feasible solution. T d :

[0120]

[0121] in, t cp The time required for the common path to start each maintenance task can be configured differently according to the distribution of distributed cluster devices. .

[0122] Step 162: Calculate the availability index of the system composed of distributed cluster devices.

[0123]

[0124] in, A It is the availability index, the larger the better; T t The total duration of a maintenance cycle. Specifically, the railway station signal distributed cluster equipment, T t Generally 180 days.

[0125] This embodiment uses the availability of the system composed of distributed cluster devices to directly reflect the steady-state status of the system operation. According to the availability index, the maintenance level and maintenance sequence of the distributed cluster devices can be optimized, thereby searching for the optimal maintenance strategy to achieve the maximum steady-state availability value of the system. A It can be used directly as the fitness function of the optimization algorithm, or it can be used to calculate the fitness function of (1- A )Search for the minimum value to cooperate with the optimization algorithm for optimization.

[0126] Step 170: Iterate and solve the optimization model to obtain the optimal solution for the maintenance task sorting of the distributed cluster equipment. After decoding, the maintenance level and maintenance sequence are output. The optimization model consists of an objective function, a feasible solution set, and constraints.

[0127] Specifically, the implementation process of step 170 is as follows:

[0128] Step 171, the objective function is established, which is a nonlinear combinatorial sorting optimization problem. This embodiment uses the ant colony algorithm to solve it. In the initialization phase of the algorithm, a real number of randomly generated task numbers is used, which is then converted into a binary code for initialization to ensure that the tasks are non-repetitive. Then, the steady-state availability value is used as the fitness function for optimization, specifically using the unavailability (i.e., 1- A ) is minimized as the optimization goal to adapt to the algorithm's minimization search.

[0129] Step 172, configure the algorithm parameters such as the number of ant colonies and the number of iterations, and finally obtain the encoded optimal decision solution. After decoding, the maintenance task sorting decision result of the distributed cluster equipment is output, including the maintenance level and maintenance order. During the iterative optimization process, each iteration generates a new feasible solution set based on the results of the previous iteration, and then uses the unavailability (1- A ) is minimized as the optimization goal, and the optimal solution of the iteration is searched until the iteration stopping condition is met.

[0130] In this embodiment, the taboo table of the ant colony algorithm can be configured as the task deletion list in step 155, and the remaining task list can be implemented by creating a new feasible task list as an optional task.

[0131] This embodiment employs an ant colony algorithm to solve the established optimization model. During the optimization process, the ant colony algorithm can fully utilize pheromones within the constraints to guide the ant colony's search direction, allowing for a more rapid determination of the optimal solution for sorting maintenance tasks for distributed cluster equipment. This optimal solution fully considers the topology, location, and operational constraints of the distributed cluster equipment, and uses steady-state reliability as an optimization metric to maximize system reliability. The resulting optimal solution provides decision support for maintenance activities for distributed cluster equipment, reducing subjectivity in actual on-site maintenance decisions and effectively improving system availability.

[0132] It can be understood that the above exemplary description of the distributed cluster equipment of railway station signals does not limit the scope of application of the maintenance task sorting optimization method proposed in this embodiment. The maintenance task sorting optimization method proposed in this embodiment can be applied to various distributed cluster equipment scenarios, such as: wind turbine groups, mobile communication base station groups, etc.; according to the topological structure and equipment type of the distributed cluster equipment, effective maintenance task sorting optimization can be performed on different distributed cluster equipment. When the types and topological structures of distributed cluster equipment in different industrial scenarios change, the state division and topological relationship reconstruction can be set according to the category, and the changed operating constraints can be adaptively configured, which can be applicable to the maintenance task sorting optimization of distributed cluster equipment in different industrial scenarios.

[0133] This embodiment also proposes a maintenance task sorting optimization system for distributed cluster equipment, such as Figure 2 As shown, the maintenance task sequencing optimization system 200 proposed in this embodiment includes:

[0134] The information acquisition module 201 is configured to: acquire the type and location information of the node according to the system topology structure composed of the distributed cluster devices;

[0135] The importance calculation module 202 is configured to: establish a general generating function model of the system topology structure and calculate the importance of each node in the system topology structure;

[0136] Duration module 203, the duration module 203 is configured to: construct a maintenance duration matrix of equipment maintenance levels and a path duration function of location paths;

[0137] A constraint configuration module 204 is configured to: configure operational constraints and resource constraints for distributed cluster equipment maintenance activities;

[0138] A feasible solution construction module 205 is configured to: encode the maintenance level and maintenance order as decision variables, and construct a feasible solution for the decision variables using the total maintenance time and node importance of the distributed cluster equipment;

[0139] An objective function construction module 206 is configured to: calculate the availability of a system composed of distributed cluster devices in a maintenance cycle as an objective function;

[0140] The iterative optimization module 207 is configured to iteratively solve the optimization model to obtain the optimal solution for the maintenance task ranking of the distributed cluster equipment, and output the maintenance level and maintenance sequence after decoding. The optimization model is composed of an objective function, a set of feasible solutions, and constraints.

[0141] It is understandable that the specific implementation process of each functional module of the system is as described in steps 1 to 7 above, and will not be elaborated here.

[0142] This embodiment also provides an electronic device 300, such as Figure 3 As shown, the electronic device 300 includes: a memory 310, a processor 320, and a computer program A311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program A311, the following steps are implemented:

[0143] Obtain node type and location information based on the system topology structure composed of distributed cluster devices;

[0144] Establish a general generating function model of the system topology structure and calculate the importance of each node in the system topology structure;

[0145] Construct the maintenance time matrix of equipment maintenance level and the path duration function of location path;

[0146] Configure operational and resource constraints for distributed cluster equipment maintenance activities;

[0147] The maintenance level and maintenance sequence are encoded as decision variables, and the total maintenance time and node importance of distributed cluster equipment are used to construct a feasible solution set.

[0148] Calculate the availability of the system composed of distributed cluster equipment in a maintenance cycle as the objective function;

[0149] The optimization model is iteratively solved to obtain the optimal solution for the maintenance task sorting of distributed cluster equipment. The maintenance level and maintenance sequence are output after decoding. The optimization model consists of an objective function, a feasible solution set, and constraints.

[0150] Optionally, when the processor 320 executes the computer program A311, any implementation method in the corresponding embodiment of the above-mentioned optimization method can be implemented.

[0151] It should be noted that the electronic device proposed in this embodiment is a device used to implement the above-mentioned optimization method. Therefore, based on the above-mentioned optimization method proposed in this embodiment, technical personnel in this field can understand the specific implementation methods of the electronic device of this embodiment and its various variations. Therefore, how the electronic device specifically implements the above-mentioned early warning method will not be introduced in detail here. As long as the electronic device used by technical personnel in this field to implement the above-mentioned optimization method falls within the scope of protection to be protected by this application.

[0152] This embodiment also provides a computer-readable storage medium 400, such as Figure 4 As shown, the computer readable storage medium 400 stores a computer program B411. When the computer program B411 is executed by the processor, the following steps are implemented:

[0153] Obtain node type and location information based on the system topology structure composed of distributed cluster devices;

[0154] Establish a general generating function model of the system topology structure and calculate the importance of each node in the system topology structure;

[0155] Construct the maintenance time matrix of equipment maintenance level and the path duration function of location path;

[0156] Configure operational and resource constraints for distributed cluster equipment maintenance activities;

[0157] The maintenance level and maintenance sequence are encoded as decision variables, and the total maintenance time and node importance of distributed cluster equipment are used to construct a feasible solution set.

[0158] Calculate the availability of the system composed of distributed cluster equipment in a maintenance cycle as the objective function;

[0159] The optimization model is iteratively solved to obtain the optimal solution for the maintenance task sorting of distributed cluster equipment. The maintenance level and maintenance sequence are output after decoding. The optimization model consists of an objective function, a feasible solution set, and constraints.

[0160] Optionally, when the computer program B411 is executed by a processor, it can implement any implementation method in the embodiments corresponding to the above-mentioned optimization method.

[0161] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0162] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0164] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0166] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of this application. It should be understood that the above description is only the specific implementation methods of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this application should be included in the scope of protection of this application.

Claims

1. A maintenance task sorting optimization method for distributed cluster equipment, characterized in that: include: According to the system topology structure composed of distributed cluster devices, the input ports and output ports are divided, and the input and output ports are matched to form the input and output port set of the distributed cluster devices, and the type and location information of the nodes are obtained; Establishing a universal generating function model of the system topology structure and calculating the importance of each node in the system topology structure, including: performing multi-state degradation performance division on the devices in each combination within the set according to the input and output port sets of the distributed cluster devices, establishing a universal generating function model for each combination, and obtaining a universal generating function model of the system topology structure according to the universal generating function model of each combination; calculating the overall performance status of the system topology structure according to the universal generating function model of the system topology structure; and obtaining the importance value of each node according to the calculation of the overall performance status of the system topology structure; Construct the maintenance time matrix of equipment maintenance level and the path duration function of location path; Configure operational and resource constraints for distributed cluster equipment maintenance activities; The maintenance level and maintenance sequence are encoded as decision variables, and the total maintenance time and node importance of distributed cluster equipment are used to construct feasible solutions for the decision variables. Calculating the availability of a system composed of distributed cluster devices in a maintenance cycle as an objective function, including: calculating the total maintenance time of the distributed cluster devices according to the codes of the feasible solutions; calculating the availability index of the system composed of the distributed cluster devices according to the total maintenance time of the distributed cluster devices, where the availability index is equal to the ratio of the difference between the total maintenance time and the total maintenance time of the maintenance cycle to the total maintenance time of the maintenance cycle; The optimization model is iteratively solved using an ant colony algorithm to obtain the optimal solution for the maintenance task sorting of distributed cluster equipment, and the maintenance level and maintenance sequence are output after decoding; wherein the optimization model is composed of the objective function, the feasible solution set and the constraint conditions.

2. A maintenance task sorting optimization method for distributed cluster equipment according to claim 1, characterized in that: The obtaining of node type and location information includes: Obtain the type and location information of the distributed cluster device, using latitude and longitude coordinates or fixed reference system coordinates to form a location set with node coding.

3. The maintenance task sorting optimization method for distributed cluster equipment according to claim 1, characterized in that: The importance value of each node is calculated based on the overall performance status of the system topology structure, and the Fussel-Vesely algorithm is used to calculate the importance value of each node.

4. The method for optimizing the maintenance task sequencing of distributed cluster equipment according to claim 1, characterized in that: The construction of the maintenance duration matrix of the equipment maintenance level and the path duration function of the location path includes: Each distributed cluster device corresponds to multiple maintenance levels, and each maintenance level corresponds to a maintenance duration. The maintenance durations of all maintenance levels of all distributed cluster devices constitute the maintenance level duration matrix. The ratio of the travel time from when the maintenance personnel performs the current task to when they perform the next task during the system operation interval to the movement speed of the maintenance personnel is calculated as the path duration function.

5. The method for optimizing the maintenance task sequencing of distributed cluster equipment according to claim 1, characterized in that: The operational constraints and resource constraints for configuring distributed cluster equipment maintenance activities include: Based on the operational requirements of distributed cluster equipment in different industrial scenarios, configure resource constraints for maintenance activities, including: maintenance costs should be less than or equal to the maximum cost resource; Based on the operational requirements of distributed cluster equipment in different industrial scenarios, operational constraints for maintenance activities are configured, including: the time for each maintenance task combination should be less than or equal to the system task interval time.

6. A maintenance task sorting optimization method for distributed cluster equipment according to any one of claims 1 to 5, characterized in that: The aforementioned encoding of the maintenance level and maintenance sequence as decision variables and constructing feasible solutions for the decision variables using the total maintenance time and node importance of distributed cluster equipment include: According to the number of distributed cluster devices, maintenance level and system task interval, binary coding technology is used to encode the decision variables of maintenance level and maintenance sequence; Based on the distribution location and constraints of the distributed cluster devices, the first maintenance task in the task list is set as the current task; Get the maintenance duration corresponding to the current task. Starting from the current task location, get the maintenance duration and path duration of all remaining tasks and calculate the total duration required for each remaining task. Calculate the difference between the maintenance time of the current task and the system task interval time; Compare the total duration of all remaining tasks with the difference, sort the tasks that are smaller than the difference according to the importance of the nodes, and extract the most important tasks from the sorting to combine them; Calculate the difference between the total duration of this combined task and the system task interval time, and return to the previous step until the total duration of the remaining tasks does not meet the condition of being less than the difference. This completes the combination of the first maintenance task and other tasks. Delete the combined tasks from the task list, then read the task with the earliest maintenance order among the remaining tasks in the task list as the current task, return to the step of obtaining the maintenance time corresponding to the current task, and perform the next task combination until there are no tasks in the task list, that is, the combination sorting of all tasks is completed. The decision variable encoding after the combination sorting is a feasible solution constructed.

7. A maintenance task sorting optimization method for distributed cluster equipment according to any one of claims 1 to 5, characterized in that: The iterative solution of the optimization model includes: In the initialization phase, a real number of the task number is randomly generated, then converted into a binary code for initialization, and an initial feasible solution set is generated; Based on the initial feasible solution set, iterative optimization is performed. Each iterative optimization generates a new feasible solution set based on the optimization result of the previous iterative optimization. Then, using the minimum unavailability as the optimization goal, the optimal solution of the current iteration is searched until the iteration stopping condition is met, and the final decision optimal solution is obtained. The unavailability is equal to 1 minus the availability index.

8. The method for optimizing the maintenance task sequencing of distributed cluster equipment according to claim 6, characterized in that: The decision variables are coded from 1 to N Groups of binary numbers, each group of binary numbers exists Position, front The digits are the maintenance level code of the equipment, and the The bits are the maintenance sequence codes for the equipment; w The highest maintenance level number. N is the number of devices, is the rounding function.

Images