CN115860435A - Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV - Google Patents
Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV Download PDFInfo
- Publication number
- CN115860435A CN115860435A CN202310135943.4A CN202310135943A CN115860435A CN 115860435 A CN115860435 A CN 115860435A CN 202310135943 A CN202310135943 A CN 202310135943A CN 115860435 A CN115860435 A CN 115860435A
- Authority
- CN
- China
- Prior art keywords
- maintenance
- agv
- equipment
- time
- power equipment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000003449 preventive effect Effects 0.000 title claims description 17
- 238000003745 diagnosis Methods 0.000 claims abstract description 60
- 230000008569 process Effects 0.000 claims abstract description 22
- 230000035772 mutation Effects 0.000 claims description 22
- 125000004432 carbon atom Chemical group C* 0.000 claims description 11
- 210000000349 chromosome Anatomy 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 9
- VKYKSIONXSXAKP-UHFFFAOYSA-N hexamethylenetetramine Chemical compound C1N(C2)CN3CN1CN2C3 VKYKSIONXSXAKP-UHFFFAOYSA-N 0.000 claims description 9
- 238000013136 deep learning model Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 230000002068 genetic effect Effects 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 229910002056 binary alloy Inorganic materials 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims 1
- 230000032258 transport Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000000192 social effect Effects 0.000 description 1
Images
Landscapes
- General Factory Administration (AREA)
Abstract
The method comprises the steps of firstly constructing a knowledge map based on historical fault data of the electric power equipment to form a knowledge map fault diagnosis system, then carrying out fault diagnosis on the equipment in a selected diagnosis equipment range by adopting the knowledge map fault diagnosis system to obtain information of the electric power equipment to be maintained, then constructing a single-target dynamic flexible job shop scheduling model with the AGV and taking the maximum maintenance time as the minimum target according to the information of the electric power equipment to be maintained, solving the scheduling model to obtain an optimal scheduling scheme, and finally allocating the AGV to transport a maintenance tool to the equipment to be maintained according to the optimal scheduling scheme based on the constructed AGV space-time transport model. The invention solves the problems of unpredictability of equipment faults and subjectivity of maintenance sequences in the existing maintenance process, and improves the automation degree of power equipment maintenance.
Description
Technical Field
The invention belongs to the technical field of flexible job shop scheduling, and particularly relates to a dynamic flexible scheduling method and system for preventive maintenance of power equipment with an AGV.
Background
At present, various power companies have three types of maintenance opportunities for power equipment, namely, the power equipment exceeds the service life and is directly scrapped or parts are replaced; secondly, the power equipment is maintained after being detected; and thirdly, the power equipment is subjected to fault repair or rush repair. The maintenance of the power equipment is carried out after the fault is known, and when a plurality of equipment needs to be maintained, the maintenance sequence is determined by a maintenance person. Unpredictability of equipment failure makes equipment maintenance random; the subjectivity of the maintenance sequence of the maintenance personnel prevents the maintenance efficiency from being maximized; the detection efficiency of the set of process is difficult to reflect the fine management of electric power materials, and is also difficult to meet the increasing maintenance requirements of electric power equipment, so that adverse social effects and economic losses are caused.
Disclosure of Invention
The invention aims to provide a knowledge graph-based power equipment preventive maintenance dynamic flexible scheduling method and system with an AGV, aiming at the problems in the prior art.
In order to realize the purpose, the technical scheme of the invention is as follows:
the method for dynamically and flexibly scheduling the preventive maintenance of the power equipment with the AGV sequentially comprises the following steps of:
a, constructing a knowledge graph based on historical fault data of the power equipment to form a knowledge graph fault diagnosis system;
b, adopting a knowledge graph fault diagnosis system to carry out fault diagnosis on equipment in the selected diagnosis equipment range to obtain the information of the electric power equipment to be maintained;
step C, constructing a single-target dynamic flexible job shop scheduling model with the AGV and the maximum maintenance time as the minimum target according to the information of the power equipment to be maintained, and solving the scheduling model to obtain an optimal scheduling scheme;
and D, allocating the AGV trolley based on the constructed AGV trolley transportation space-time model to transport the maintenance tool to the equipment to be maintained according to the optimal scheduling scheme for maintenance.
In step C, the objective function of the single-target dynamic flexible job shop scheduling model with the AGV is as follows:
in the above formula, the first and second carbon atoms are,
for a total maintenance completion time>
A maintenance status variable for the ith maintenance tool for the jth part of the kth apparatus to be maintained>
For the end time of the jth part of the ith service tool for the kth device to be serviced, the decision unit decides>
For the maintenance time of the ith maintenance tool on the f-th part>
And I is the total number of parts of all the devices,
for the maintenance start time of the ith maintenance tool for the jth part of the kth apparatus to be maintained, a decision is made as to whether the maintenance start time is greater than or equal to>
For the maintenance start time of the xth part of the kth apparatus to be maintained for the xth maintenance tool, the->
Is a very large positive number, and>
is a 0-1 variable, if>
Early in->
Then>
=1, otherwise->
=0，/>
For the maintenance time of the jth part of the kth apparatus to be maintained for the ith maintenance tool, ->
A transport process transported by the AGV car after servicing the f part for the ith service tool>
For a conveying process->
Is started, is greater than or equal to>
For a conveying process->
Is greater than or equal to>
、/>
Respectively, loading time and unloading time, m is the total number of equipment to be maintained, n is the total number of parts in the equipment, and d is the total number of maintenance tools.
In the step C, a hybrid algorithm is adopted for solving the single-target dynamic flexible job shop scheduling model with the AGV, and the method sequentially comprises the following steps:
step C1, coding the information of the equipment to be maintained, initializing a genetic algorithm, and randomly generating a group of initial populations;
c2, decoding and calculating a fitness value of the population, wherein the fitness value is an objective function value of the model;
c3, mutation is carried out on individuals with low fitness values by using tabu search as a mutation operator, and first tabu search mutation is realized;
step C4, screening individuals with high fitness in the population as parent individuals of subsequent evolution by using a binary system tournament method;
step C5, randomly selecting two chromosomes from the population obtained by screening as parent chromosomes, and uniformly crossing and crossing GOX respectively on the process station part and the process sequence part in the parent chromosomes to generate new individuals;
c6, mutating the new individuals with low fitness values by using tabu search as a mutation operator to realize second tabu search mutation;
and C7, outputting the optimal scheduling scheme.
In steps C3 and C6, the mutating an individual with a low fitness value by using tabu search as a mutation operator includes: if the fitness value of the individual is low, entering a tabu search process, putting the considered solution into a tabu table, removing the neighborhood candidate solution which is the same as that in the tabu table when neighborhood search is carried out, and stopping the tabu search when the solution is not improved or the repeated iteration times of all the neighborhood candidate solutions which are tabu reach the maximum iteration times;
the step C4 comprises the following steps: and selecting a certain number of individuals from the group, wherein two individuals form a group to play the game, and winning the match with a high fitness value to participate in the next round of game until the individual with the best fitness is determined.
The AGV dolly transportation space-time model is based on the spatial layout condition construction of waiting to maintain equipment and obtains, specifically is:
AGV Car
In a static state:
in the above formula, the first and second carbon atoms are,
is->
In a position of (4), (v) is greater than or equal to>
Is a device to be serviced>
Is taken up and taken off>
Is->
Stay on the device to be serviced>
A time period of (d);
AGV Car
In the no-load state: />
In the above formula, the first and second carbon atoms are,
is->
Is selected from the device to be serviced>
Come to the device to be repaired>
Based on the driving speed of the vehicle>
For devices to be serviced>
Is taken up and taken off>
Is in an idle state>
Slave->
Back to the end time of the X-axis of the main rail, <' > or>
Is->
Is moved along the main track to be matched with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Move to the device to be serviced>
The time of (a) is,
、/>
are respectively the fifth->
K devices to be serviced are located in spatially distributed positions->
Is in an idle state>
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The time required for treatment;
AGV Car
In the conveying state:
in the above formula, the first and second carbon atoms are,
is a device to be serviced>
Is taken up and taken off>
Is->
Is selected from the device to be serviced>
Is in the device to be repaired>
Based on the driving speed of the vehicle, is adjusted>
、/>
Respectively a loading time, an unloading time->
For the loading time of the maintenance tool, is>
Is in a conveying state>
Is selected from the device to be serviced>
Is returned to the end time of the X-axis of the main track and is taken>
Is->
Is moved along the main track to be matched with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The end time of (c).
The step A sequentially comprises the following steps:
a1, training an AI deep learning model for power equipment fault diagnosis based on historical fault data of the power equipment;
a2, constructing a knowledge graph based on an AI (advanced learning) model for power equipment fault diagnosis;
and A3, dynamically updating the knowledge graph to form a knowledge graph fault diagnosis system.
The step A2 comprises the following steps in sequence:
a21, collecting fault maintenance data of each part and relevant performance index parameters during maintenance, and inputting the data into a power equipment fault diagnosis AI deep learning model to obtain an expert knowledge set, wherein the performance indexes comprise pressure, temperature, vibration frequency, rotating speed and acceleration;
step A22, putting an expert knowledge set into an expert knowledge base;
and A23, correcting the expert knowledge base based on the professional knowledge to form a knowledge graph.
The step A3 comprises the following steps in sequence:
a31, processing the acquired latest power equipment data to form a standardized data set;
step A32, taking the standardized data set as the input of an inference machine, and finishing fault diagnosis inference based on a knowledge graph;
and step A33, correcting the fault diagnosis result by the user, adding the corrected fault diagnosis result into historical fault data, and finishing the updating of the knowledge graph.
The power equipment preventive maintenance dynamic flexible scheduling system with the AGV comprises a knowledge map fault diagnosis system building module, a device to be maintained determining module, a scheduling model building and solving module, an AGV trolley transportation space-time model building module and an AGV trolley allocation module;
the knowledge map fault diagnosis system construction module is used for constructing a knowledge map based on historical fault data of the power equipment to form a knowledge map fault diagnosis system;
the to-be-maintained equipment determining module is used for performing fault diagnosis on equipment in a selected diagnosis equipment range by adopting a knowledge graph fault diagnosis system to obtain to-be-maintained electric power equipment information;
the scheduling model constructing and solving module is used for constructing a single-target dynamic flexible job shop scheduling model with the AGV and the maximum maintenance time as the target according to the power information of the equipment to be maintained, and solving the scheduling model to obtain an optimal scheduling scheme;
the AGV trolley transportation space-time model building module is used for building an AGV trolley transportation space-time model;
the AGV dolly allotment module allots the AGV dolly and transports the maintenance tool to waiting to maintain equipment department and maintain according to the optimal scheduling scheme based on the AGV dolly transportation space-time model that founds.
Compared with the prior art, the invention has the beneficial effects that:
1. the method comprises the steps of firstly constructing a knowledge map based on historical fault data of the electric power equipment to form a knowledge map fault diagnosis system, then carrying out fault diagnosis on the equipment in a selected diagnosis equipment range by adopting the knowledge map fault diagnosis system to obtain information of the electric power equipment to be maintained, then constructing a single-target dynamic flexible job shop scheduling model with the AGV and aiming at the minimum maintenance time according to the information of the electric power equipment to be maintained, solving the scheduling model to obtain an optimal scheduling scheme, finally allocating the AGV to transport a maintenance tool to the equipment to be maintained according to the optimal scheduling scheme based on the constructed AGV transportation space-time model, diagnosing the health state of the electric power equipment by the knowledge map fault diagnosis system, inputting the diagnosed result as the dynamic flexible scheduling system, calculating the optimal scheduling scheme through the dynamic flexible scheduling model, finally transporting materials such as maintenance parts to corresponding positions by the AGV according to the optimal scheduling scheme, and only needing a user to define the range of the equipment to be scheduled in the knowledge map fault diagnosis system, completing fault location, maintaining and scheduling, solving the problem of the maintenance of the existing equipment in the maintenance sequence automatically.
2. The power equipment preventive maintenance dynamic flexible scheduling method with the AGV solves the scheduling model by adopting a hybrid algorithm, and the algorithm utilizes tabu search as a mutation operator to mutate individuals with low fitness value, so that the first tabu search mutation is realized, and the initial population quality is further improved; screening individuals with high fitness in the population as parent individuals for subsequent evolution by using a binary tournament method, eliminating the individuals with low fitness, avoiding the influence of super individuals by using a parallel mechanism of the algorithm, selecting the individuals with better fitness in a short time, avoiding the premature phenomenon to a certain extent and ensuring the stability of the algorithm; and the second tabu search mutation can effectively avoid roundabout search and prevent the algorithm from falling into local optimum.
Drawings
FIG. 1 is a flow chart of the present invention.
Fig. 2 is a gantt chart of the optimal scheduling scheme obtained in example 1.
Fig. 3 is a spatial layout coordinate diagram of the equipment to be repaired in example 1.
Fig. 4 is a schematic structural diagram of the system described in embodiment 2.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings.
Example 1:
referring to fig. 1, a dynamic flexible scheduling method for preventive maintenance of power equipment with AGVs is sequentially performed according to the following steps:
1. training an AI deep learning model for power equipment fault diagnosis based on historical fault data of the power equipment, wherein the AI deep learning model for power equipment fault diagnosis comprises a power equipment service life prediction model, a fault occurrence time model and a fault prediction model.
2. Collecting fault maintenance data of each part and performance index parameters related to maintenance, inputting the data into an AI (advanced fault diagnosis) deep learning model of power equipment to obtain an expert knowledge set, putting the expert knowledge set into an expert knowledge base to update and expand the expert knowledge base, and then correcting the expert knowledge base by a knowledge engineer based on professional knowledge to form a knowledge graph, wherein the performance indexes comprise pressure, temperature, vibration frequency, rotating speed and acceleration;
3. the method comprises the steps of carrying out data processing on collected latest power equipment data to form a standardized data set, using the standardized data set as input of a reasoning machine, completing fault diagnosis reasoning based on a knowledge graph, carrying out semantic interpretation on a diagnosis result for a user to check, correcting the fault diagnosis result by the user, adding the corrected fault diagnosis result into historical fault data, completing dynamic updating of the knowledge graph, and finally forming a knowledge graph fault diagnosis system.
4. The knowledge map fault diagnosis system collects equipment performance index data in the selected diagnostic equipment range in real time according to the selected diagnostic equipment range, carries out fault diagnosis through the collected performance index data and obtains the information of the electric power equipment to be maintained, wherein the information of the electric power equipment to be maintained comprises the name and the work area position of the electric power equipment to be maintained and the maintenance duration of different parts on the electric power equipment to be maintained under different maintenance tools.
5. Constructing a single-target dynamic flexible job shop scheduling model with the maximum maintenance time as the target and the minimum maintenance time as the target according to the information of the power equipment to be maintained:
In the above formula, the first and second carbon atoms are,
for a total maintenance completion time>
A maintenance status variable for the ith maintenance tool for the jth part of the kth apparatus to be maintained>
For the end time of the jth part of the ith service tool for the kth device to be serviced, the decision unit decides>
For the maintenance time of the ith maintenance tool on the f th part>
And I is the total number of parts of all the devices,
for the ith maintenance tool to the kth equipment to be maintainedMaintenance start time of j parts>
For the maintenance start time of the xth part of the kth apparatus to be maintained for the xth maintenance tool, the->
Is a very large positive number, and>
is a 0-1 variable, if>
Early in->
Then->
=1, otherwise->
=0，/>
For the maintenance time of the jth part of the kth apparatus to be maintained for the ith maintenance tool, ->
A transport process transported by the AGV car after servicing the f part for the ith service tool>
For a conveying process->
Is started, is greater than or equal to>
For a conveying process->
Is greater than or equal to>
、/>
Respectively, loading time and unloading time, m is the total number of equipment to be maintained, n is the total number of parts in the equipment, and d is the total number of maintenance tools.
6. Solving the scheduling model by adopting a hybrid algorithm to obtain an optimal scheduling scheme shown in fig. 2, wherein the method specifically comprises the following steps:
6-1, coding the equipment and parts to be maintained by adopting a double-layer chromosome coding mode, initializing a genetic algorithm, wherein the genetic algorithm comprises an initial population scale N, a current iteration number t and a maximum iteration number t max A set of initial populations is randomly generated.
And 6-2, decoding and calculating a fitness value of the population, wherein the fitness value is an objective function value of the model.
The decoding process is equal to the inverse process of the coding, a decoding mode based on active scheduling is adopted, the maintenance time of the part is converted according to the part to be maintained and the machine coding, and the load and the termination time of each machine are calculated according to the part, the equipment to be maintained and the maintenance time. The maximum time-out is the end time when the last equipment is repaired.
6-3, mutation is carried out on individuals with low fitness values by using tabu search as a mutation operator, and first tabu search mutation is realized.
Tabu search starts from an initial feasible solution, a series of specific search directions are selected as heuristics, and the movement that allows the most change in the value of a specific objective function is selected. The mutation process is as follows: if the fitness value of the individual is low, entering a tabu search process, putting the considered solution into a tabu table, removing the neighborhood candidate solution which is the same as that in the tabu table when neighborhood search is carried out, and stopping the tabu search when the solution is not improved or the repeated iteration times of all the neighborhood candidate solutions which are tabu reach the maximum iteration times.
6-4, screening the individuals with high fitness in the population as parent individuals of subsequent evolution by using a binary tournament method.
The selection mechanism is to select the individuals with higher fitness as parent individuals of the subsequent evolution, so as to further improve the average fitness of the population. The present embodiment selects a binary tournament method, which is applicable to most problems. The binary tournament method is characterized in that a certain number of individuals are selected from a group, two individuals form a group to be subjected to competition, and the individuals with better fitness win the next round of competition until the individuals with the best fitness are obtained. The binary tournament method is used for selecting the offspring individuals with excellent fitness as the parent individuals in the later cross mutation process, the parallel mechanism of the algorithm is used, the influence of super individuals is avoided, the individuals with excellent fitness are selected in a short time, and the premature phenomenon is avoided to a certain extent.
And 6-5, randomly selecting two chromosomes from the population obtained by screening as parent chromosomes, and uniformly crossing and crossing GOX respectively on the process station part and the process sequence part in the parent chromosomes to generate new individuals and improve the population diversity.
And 6-6, mutating the new individuals with low fitness values by using tabu search as a mutation operator, realizing second tabu search mutation, and then outputting an optimal scheduling scheme.
7. And constructing an AGV trolley transportation space-time model based on the distribution condition of the equipment to be maintained.
In order to simulate a real maintenance scene, the layout of the specific equipment to be maintained is analyzed, a space position condition is considered on the basis of a mathematical model of AGV single-target dynamic flexible maintenance operation scheduling, a space-time model of AGV trolley transportation is established, and a space layout coordinate system of the equipment to be maintained is shown in FIG. 3.
The coordinate axis is set by taking the starting point of the primary track of the AGV trolley as a far point, the position change direction of the primary track as an X axis and the position change aspect of the secondary track as a Y axis, so that the coordinate value of any equipment to be maintained in the coordinate system can be represented as the coordinate value
。
Setting a time variable t in the execution process of the maintenance task, and dividing the motion state of the AGV into 3 types: static, no-load and carrying. Suppose in
Is reached at moment>
At that moment, the AGV dolly>
Stays always on the device to be serviced>
Is located, is slave->
The moment begins to receive a conveying task, wherein the conveying task is to judge whether the maintenance tool is from the equipment to be maintained>
Is conveyed to the device to be repaired>
At, is->
Need to be driven from
At a moment starting from the position previously parked>
Go to the device to be serviced>
To (3). The carrying process is generally divided into three steps:
in the first step, the first step is that,
after the maintenance tool completes the maintenance task of a part of the previous equipment, the equipment to be maintained
Takes the time to load the service tool and the loading time->
；
In the second step, the first step is that,
taking the service tool out of the device to be serviced>
Is conveyed to>
The concrete same as the no-load process;
thirdly, unloading the maintenance tool to the equipment to be maintained
Where the elapsed time is the unload time>
。
Therefore, there are:
the AGV dolly transportation space-time model is based on the spatial layout condition construction of waiting to maintain equipment and obtains, specifically is:
AGV Car
In a static state:
in the above formula, the first and second carbon atoms are,
is->
In a position of (4), (v) is greater than or equal to>
Is a device to be serviced>
Is taken up and taken off>
Is->
Stay on the device to be serviced>
A time period of (d);
AGV Car
In the no-load state:
in the above formula, the first and second carbon atoms are,
is->
Is selected from the device to be serviced>
Is in the device to be repaired>
Based on the driving speed of the vehicle>
Is a device to be serviced>
Is taken up and taken off>
Is in an idle state>
Slave->
Back to the end time of the X-axis of the main rail, <' > or>
Is->
Is moved along the main track to be matched with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Move to the device to be serviced>
The time of (a) is,
、/>
are respectively the fifth->
K devices to be serviced are located in spatially distributed positions->
Is in an idle state>
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The time required for treatment;
AGV Car
In the conveying state: />
In the above formula, the first and second carbon atoms are,
is a device to be serviced>
Is taken up and taken off>
Is->
Is selected from the device to be serviced>
Is in the device to be repaired>
Based on the driving speed, is greater or less than>
、/>
Respectively the loading timeUnload time,. Based on the measured value>
For the loading time of the maintenance tool, is>
In a conveying state>
Slave device to be repaired>
Is returned to the end time of the X-axis of the main track and is taken>
Is->
Is moved along the main track to be matched with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The end time of (c).
8. And allocating the AGV to transport the maintenance tool to the equipment to be maintained according to the optimal scheduling scheme for maintenance based on the constructed AGV transporting space-time model.
Example 2:
referring to fig. 4, the power equipment preventive maintenance dynamic flexible scheduling system with AGV includes a knowledge map fault diagnosis system building module 1, a device to be maintained determining module 2, a scheduling model building and solving module 3, an AGV trolley transportation space-time model building module 4, and an AGV trolley allocation module 5;
the knowledge map fault diagnosis system construction module 1 is used for constructing a knowledge map based on historical fault data of the power equipment to form a knowledge map fault diagnosis system;
the to-be-maintained equipment determining module 2 is used for performing fault diagnosis on equipment in a selected diagnosis equipment range by adopting a knowledge-graph fault diagnosis system to obtain to-be-maintained electric equipment information;
the scheduling model constructing and solving module 3 is used for constructing a single-target dynamic flexible job shop scheduling model with AGV according to the power information of the equipment to be maintained, wherein the single-target dynamic flexible job shop scheduling model takes the minimum maximum maintenance time as a target, and solving the scheduling model to obtain an optimal scheduling scheme;
the AGV trolley transportation space-time model building module 4 is used for building an AGV trolley transportation space-time model;
the AGV dolly allotment module 5 allots the AGV dolly and transports the maintenance tool to the equipment department of waiting to maintain according to the optimal scheduling scheme based on the AGV dolly transportation space-time model that founds and maintains.
Claims (9)
1. The dynamic flexible scheduling method for preventive maintenance of the power equipment with the AGV is characterized by comprising the following steps of:
the scheduling method sequentially comprises the following steps:
a, constructing a knowledge graph based on historical fault data of the power equipment to form a knowledge graph fault diagnosis system;
b, adopting a knowledge graph fault diagnosis system to carry out fault diagnosis on equipment in the selected diagnosis equipment range to obtain the information of the electric power equipment to be maintained;
step C, constructing a single-target dynamic flexible job shop scheduling model with the AGV and the maximum maintenance time as the minimum target according to the information of the power equipment to be maintained, and solving the scheduling model to obtain an optimal scheduling scheme;
and D, allocating the AGV trolley based on the constructed AGV trolley transportation space-time model to transport the maintenance tool to the equipment to be maintained according to the optimal scheduling scheme for maintenance.
2. The method of claim 1, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
in step C, the objective function of the single-target dynamic flexible job shop scheduling model with the AGV is as follows:
in the above formula, the first and second carbon atoms are,
for a total maintenance completion time>
A maintenance status variable for the ith maintenance tool for the jth part of the kth apparatus to be maintained>
For the end time of the jth part of the ith service tool for the kth device to be serviced, the decision unit decides>
For the maintenance time of the ith maintenance tool on the f-th part>
I is the total number of parts of all the devices>
For the maintenance start time of the ith maintenance tool for the jth part of the kth apparatus to be maintained, a decision is made as to whether the maintenance start time is greater than or equal to>
For the maintenance start time of the xth part of the kth apparatus to be maintained for the xth maintenance tool, the->
Is a very large positive number, and>
is a 0-1 variable, if>
Early in->
Then->
=1, otherwise->
=0，/>
For the maintenance time of the jth part of the kth apparatus to be maintained for the ith maintenance tool, ->
A transport process transported by the AGV car after servicing the f part for the ith service tool>
For a transportation process>
In conjunction with a start time of (a)>
For a conveying process->
Is greater than or equal to>
、/>
Respectively, loading time and unloading time, m is the total number of equipment to be maintained, n is the total number of parts in the equipment, and d is the total number of maintenance tools.
3. The method according to claim 2, wherein said method comprises:
in the step C, a hybrid algorithm is adopted for solving the single-target dynamic flexible job shop scheduling model with the AGV, and the method sequentially comprises the following steps of:
c1, coding information of equipment to be maintained, initializing a genetic algorithm, and randomly generating a group of initial populations;
c2, decoding and calculating a fitness value of the population, wherein the fitness value is an objective function value of the model;
c3, mutation is carried out on individuals with low fitness values by using tabu search as a mutation operator, and first tabu search mutation is realized;
step C4, screening individuals with high fitness in the population as parent individuals of subsequent evolution by using a binary system tournament method;
step C5, randomly selecting two chromosomes from the population obtained by screening as parent chromosomes, and uniformly crossing and crossing GOX respectively on the process station part and the process sequence part in the parent chromosomes to generate new individuals;
c6, mutating a new individual with low fitness value by using tabu search as a mutation operator to realize second tabu search mutation;
and C7, outputting the optimal scheduling scheme.
4. The method of claim 3, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
in steps C3 and C6, the mutating an individual with a low fitness value by using tabu search as a mutation operator includes: if the individual fitness value is low, entering a tabu search process, putting the considered solution into a tabu table, removing the neighborhood candidate solution which is the same as that in the tabu table when neighborhood search is carried out, and stopping the tabu search when the solution is not improved or the repeated iteration times of all the neighborhood candidate solutions which are tabu reach the maximum iteration times;
the step C4 comprises the following steps: and selecting a certain number of individuals from the group, wherein two individuals form a group to play the game, and winning the match with a high fitness value to participate in the next round of game until the individual with the best fitness is determined.
5. The method of claim 1, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
the AGV dolly transportation space-time model is based on the spatial layout condition construction of waiting to maintain equipment and obtains, specifically is:
AGV Car
In a static state:
in the above-mentioned formula, the compound has the following structure,
is->
In a position of (4), (v) is greater than or equal to>
Is a device to be serviced>
Is taken up and taken off>
Is->
Stay on the device to be serviced>
A time period of (d);
AGV Car
In the no-load state:
in the above formula, the first and second carbon atoms are,
is->
Is selected from the device to be serviced>
Is in the device to be repaired>
Based on the driving speed of the vehicle>
Is a device to be serviced>
Is taken up and taken off>
Is in an unloaded stateLower->
Slave->
Back to the end time of the X-axis of the main rail, <' > or>
Is->
Is moved along the main track to be matched with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Move to the device to be serviced>
The time of (a) is,
、/>
are respectively the fifth->
K devices to be serviced are located in spatially distributed positions->
Is in an idle state>
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The time required for treatment;
AGV Car
In the conveying state: />
In the above formula, the first and second carbon atoms are,
is a device to be serviced>
Is taken up and taken off>
Is->
Is selected from the device to be serviced>
Come to the device to be repaired>
Based on the driving speed, is greater or less than>
、/>
Respectively a loading time, an unloading time->
For the loading time of a maintenance tool>
Is in a conveying state>
Is selected from the device to be serviced>
Is returned to the end time of the X-axis of the main track and is taken>
Is->
Move along the main rail into a position in which it is associated with a device to be serviced>
Is equal in the X-axis coordinate, an end time, is greater than or equal to>
Is->
Is selected by the device to be serviced>
Is moved to the device to be serviced>
The end time of (c).
6. The method of claim 1, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
the step A sequentially comprises the following steps:
a1, training an AI deep learning model for power equipment fault diagnosis based on historical fault data of the power equipment;
a2, constructing a knowledge graph based on an AI (advanced learning) model for power equipment fault diagnosis;
and A3, dynamically updating the knowledge graph to form a knowledge graph fault diagnosis system.
7. The method of claim 6, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
the step A2 comprises the following steps in sequence:
a21, collecting fault maintenance data of each part and relevant performance index parameters during maintenance, and inputting the data into a power equipment fault diagnosis AI deep learning model to obtain an expert knowledge set, wherein the performance indexes comprise pressure, temperature, vibration frequency, rotating speed and acceleration;
step A22, putting an expert knowledge set into an expert knowledge base;
and A23, correcting the expert knowledge base based on the professional knowledge to form a knowledge graph.
8. The method of claim 6, wherein the AGV-equipped power equipment preventive maintenance dynamic flexible scheduling method comprises:
the step A3 comprises the following steps in sequence:
a31, processing the acquired latest power equipment data to form a standardized data set;
step A32, taking the standardized data set as the input of an inference machine, and finishing fault diagnosis inference based on a knowledge graph;
and step A33, the user corrects the fault diagnosis result, and adds the corrected fault diagnosis result into historical fault data to finish the updating of the knowledge graph.
9. Take power equipment preventive maintenance dynamic flexible dispatch system of AGV, its characterized in that:
the system comprises a knowledge map fault diagnosis system construction module (1), a to-be-maintained equipment determination module (2), a scheduling model construction and solving module (3), an AGV trolley transportation space-time model construction module (4) and an AGV trolley allocation module (5);
the knowledge map fault diagnosis system construction module (1) is used for constructing a knowledge map based on historical fault data of the power equipment to form a knowledge map fault diagnosis system;
the to-be-maintained equipment determining module (2) is used for carrying out fault diagnosis on equipment in a selected diagnosis equipment range by adopting a knowledge map fault diagnosis system to obtain to-be-maintained electric equipment information;
the scheduling model constructing and solving module (3) is used for constructing a single-target dynamic flexible job shop scheduling model with the maximum maintenance time and the minimum maintenance time as targets according to the power information of the equipment to be maintained, and solving the scheduling model to obtain an optimal scheduling scheme;
the AGV trolley transportation space-time model building module (4) is used for building an AGV trolley transportation space-time model;
and the AGV trolley allocation module (5) allocates the AGV trolley based on the constructed AGV trolley transportation space-time model to transport the maintenance tool to the position of the equipment to be maintained according to the optimal scheduling scheme for maintenance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310135943.4A CN115860435A (en) | 2023-02-20 | 2023-02-20 | Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310135943.4A CN115860435A (en) | 2023-02-20 | 2023-02-20 | Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115860435A true CN115860435A (en) | 2023-03-28 |
Family
ID=85658437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310135943.4A Pending CN115860435A (en) | 2023-02-20 | 2023-02-20 | Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115860435A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829577A (en) * | 2024-03-06 | 2024-04-05 | 江苏益捷思信息科技有限公司 | Operation management method and system of business operation and maintenance platform |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509483A (en) * | 2018-01-31 | 2018-09-07 | 北京化工大学 | The mechanical fault diagnosis construction of knowledge base method of knowledge based collection of illustrative plates |
| CN113723632A (en) * | 2021-08-27 | 2021-11-30 | 北京邮电大学 | Industrial equipment fault diagnosis method based on knowledge graph |
| CN113870046A (en) * | 2021-09-07 | 2021-12-31 | 国网河北省电力有限公司电力科学研究院 | Power equipment fault diagnosis method and equipment |
| CN114579875A (en) * | 2022-03-03 | 2022-06-03 | 中国科学院大学 | Equipment fault diagnosis and maintenance knowledge recommendation system based on knowledge graph |
| CN115310630A (en) * | 2022-08-03 | 2022-11-08 | 彩虹无线（北京）新技术有限公司 | Vehicle fault maintenance method based on knowledge graph |
-
2023
- 2023-02-20 CN CN202310135943.4A patent/CN115860435A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509483A (en) * | 2018-01-31 | 2018-09-07 | 北京化工大学 | The mechanical fault diagnosis construction of knowledge base method of knowledge based collection of illustrative plates |
| CN113723632A (en) * | 2021-08-27 | 2021-11-30 | 北京邮电大学 | Industrial equipment fault diagnosis method based on knowledge graph |
| CN113870046A (en) * | 2021-09-07 | 2021-12-31 | 国网河北省电力有限公司电力科学研究院 | Power equipment fault diagnosis method and equipment |
| CN114579875A (en) * | 2022-03-03 | 2022-06-03 | 中国科学院大学 | Equipment fault diagnosis and maintenance knowledge recommendation system based on knowledge graph |
| CN115310630A (en) * | 2022-08-03 | 2022-11-08 | 彩虹无线（北京）新技术有限公司 | Vehicle fault maintenance method based on knowledge graph |
Non-Patent Citations (1)
| Title |
|---|
| 肖萌: "基于混合遗传算法的电力设备检测柔性调度研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829577A (en) * | 2024-03-06 | 2024-04-05 | 江苏益捷思信息科技有限公司 | Operation management method and system of business operation and maintenance platform |
| CN117829577B (en) * | 2024-03-06 | 2024-05-14 | 江苏益捷思信息科技有限公司 | Operation management method and system of business operation and maintenance platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Masmoudi et al. | Project scheduling under uncertainty using fuzzy modelling and solving techniques | |
| CN111144710B (en) | Construction and dynamic scheduling method of sustainable hybrid flow shop | |
| Karthikeyan et al. | A hybrid discrete firefly algorithm for multi-objective flexible job shop scheduling problem with limited resource constraints | |
| Rohaninejad et al. | A hybrid multi-objective genetic algorithm based on the ELECTRE method for a capacitated flexible job shop scheduling problem | |
| CN110070235B (en) | Flexible scheduling method for multiple mobile robots | |
| Wang et al. | A novel multi-objective optimization algorithm for the integrated scheduling of flexible job shops considering preventive maintenance activities and transportation processes | |
| Xu et al. | Optimization approaches for solving production scheduling problem: A brief overview and a case study for hybrid flow shop using genetic algorithms | |
| CN108182518B (en) | In-process state personnel scheduling method based on improved genetic algorithm | |
| CN112561194A (en) | Production and logistics integrated scheduling method and system for hybrid flow shop | |
| CN115860435A (en) | Power equipment preventive maintenance dynamic flexible scheduling method and system with AGV | |
| CN111353646B (en) | Steelmaking flexible scheduling optimization method, system, medium and equipment with switching time | |
| CN111966049B (en) | Scheduling control method for production equipment of mixed flow shop | |
| Sivaramkumar et al. | Demonstrating the importance of using total time balance instead of route balance on a multi-objective vehicle routing problem with time windows | |
| CN107230023A (en) | Based on the production and transportation coordinated dispatching method and system for improving harmony search | |
| Hani et al. | Simulation based optimization of a train maintenance facility | |
| Hosseinian et al. | An energy-efficient mathematical model for the resource-constrained project scheduling problem: an evolutionary algorithm | |
| Nie et al. | A game-theory approach based on genetic algorithm for flexible job shop scheduling problem | |
| Liu et al. | Dynamic parallel machine scheduling with deep Q-network | |
| CN117892958A (en) | Fuzzy carrier-based aircraft hangar maintenance scheduling method considering influence of personnel fatigue uncertainty | |
| CN117391564A (en) | New energy logistics vehicle energy supplementing and scheduling data model and scheduling optimization method | |
| CN117314055A (en) | Intelligent manufacturing workshop production-transportation joint scheduling method based on reinforcement learning | |
| CN116822217A (en) | Man-machine double-resource constraint multi-target production scheduling method considering man-hour uncertainty | |
| CN116027741A (en) | Edge cloud collaborative artificial intelligence framework for complex manufacturing scene | |
| Liang et al. | Balancing stochastic mixed-model two-sided disassembly line using multiobjective genetic flatworm algorithm | |
| Xu et al. | Mixed model assembly line balancing problem with fuzzy operation times and drifting operations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230328 |