CN110554673B - Intelligent RGV processing system scheduling method and device - Google Patents

Intelligent RGV processing system scheduling method and device Download PDF

Info

Publication number
CN110554673B
CN110554673B CN201910877811.2A CN201910877811A CN110554673B CN 110554673 B CN110554673 B CN 110554673B CN 201910877811 A CN201910877811 A CN 201910877811A CN 110554673 B CN110554673 B CN 110554673B
Authority
CN
China
Prior art keywords
cnc
rgv
workpieces
clinker
workpiece
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.)
Active
Application number
CN201910877811.2A
Other languages
Chinese (zh)
Other versions
CN110554673A (en
Inventor
胡华清
赵瑞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN201910877811.2A priority Critical patent/CN110554673B/en
Publication of CN110554673A publication Critical patent/CN110554673A/en
Application granted granted Critical
Publication of CN110554673B publication Critical patent/CN110554673B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32252—Scheduling production, machining, job shop
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Quality & Reliability (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Automation & Control Theory (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Manufacturing & Machinery (AREA)
  • General Factory Administration (AREA)

Abstract

The invention discloses a scheduling method and a scheduling device for an intelligent RGV (graphics processing volume) processing system, and belongs to the field of workpiece scheduling. The method adopts a mode of nesting an ant colony algorithm by a genetic algorithm, the genetic algorithm optimizes the workpiece feeding sequence, and the ant colony algorithm optimizes the information transfer process between nodes. Each ant is provided with a chromosome recorded with a workpiece feeding sequence and a taboo table, and the selectable CNC is obtained through the taboo table and the scheduling logic. And determining the next node of the ant through the transition probability, transferring to the CNC and performing the operation through the scheduling logic. When the workpieces are completely processed, the ant optimizing process is finished; and when all ants finish optimizing, recording the optimal solution and carrying out cross variation on the chromosome, and finishing the iteration. And obtaining the optimal scheduling route through multiple iterations. The invention makes up the characteristic of weak global optimization capability of the ant colony algorithm through the genetic algorithm, has stronger parallel search capability of the ant colony algorithm, realizes complementation and improves the optimization capability of the algorithm.

Description

Intelligent RGV processing system scheduling method and device
Technical Field
The invention relates to the field of workpiece scheduling, in particular to a scheduling method and a scheduling device for an intelligent RGV (graphics processing volume) processing system.
Background
Fig. 1 shows a schematic diagram of an intelligent RGV (Rail Guided Vehicle) processing system, which includes 1 Rail-type RGV for carrying and loading/unloading workpieces, 1 RGV linear Rail, several CNC (computer Numerical Control) machines for processing workpieces, 1 loading conveyor, 1 unloading conveyor, and other accessories. The RGV can automatically control the moving direction and distance according to the feeding and discharging instructions and reach the designated CNC position. The RGV is provided with a mechanical arm, two mechanical claws (each mechanical claw can only carry one workpiece at most) for grabbing/placing/carrying, and a material cleaning tank for cleaning the final material before leaving the processing system, so that the operation tasks of loading and unloading, material cleaning and the like can be completed.
In the prior art, an ant colony algorithm is generally used for optimizing a scheduling path of an RGV processing system, but the global optimization capability of the traditional ant colony algorithm is not strong.
Disclosure of Invention
In order to solve the technical problems, the invention provides a scheduling method and a scheduling device for an intelligent RGV processing system.
The technical scheme provided by the invention is as follows:
in a first aspect, the present invention provides a method for scheduling an intelligent RGV processing system, the method comprising:
step S1: initializing an ant population, an pheromone concentration of each process, an equipment dimension tabu, and a remaining process time for each CNC, the ant population including NmEach ant carries a chromosome representing the feeding sequence, initial values are given to the chromosomes, and the chromosomes of all the ants form a feeding matrix; (ii) a Wherein the device dimension tabu table records the current processing conditions of each CNC for determining which CNC the RGV can go to next
Step S2: obtaining all the CNC to which the current ants can go from the current CNC in the next step as optional CNC according to the equipment dimension taboo table and preset scheduling logic;
step S3: judging whether loading operation is needed according to the scheduling logic, and if the loading operation is needed, selecting workpiece numbers at corresponding positions in the loading matrix to wait for loading;
step S4: calculating the transition probability of the current ant moving from the current CNC to each selectable CNC according to the pheromone concentration of each processing process and the remaining processing time of each CNC, and selecting the next CNC going from the current CNC position from the selectable CNC according to the transition probability;
step S5: moving the current ants from the current CNC to the to-be-sent CNC, and carrying out operation according to scheduling logic, wherein if loading operation is required, workpieces selected from a loading matrix are loaded to the corresponding CNC;
step S6: repeating the step S2 to the step S5 until all N workpieces are processed, and obtaining a scheduling route;
step S7: repeating the steps S2-S6 for the next ant until all N ants are traversedmAnts alone to obtain NmA scheduling route;
step S8: will NmRecording the shortest scheduling route used in each scheduling route;
step S9: will NmPerforming cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix;
step S10: repeating the step S2 to the step S9 until the set total iteration times G are reached, and obtaining G recorded dispatching routes;
step S11: and selecting the scheduling route with the shortest time from the G recorded scheduling routes, namely the final scheduling route.
Further, the transition probability p of the current ant moving from the current m-th CNC to the m' th selectable CNC is calculated through the following formulam′：
Figure GDA0002492734460000021
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m 'are CNC serial numbers, m is a serial number of the current CNC, m' is a serial number of the selectable CNC, n and n 'are serial numbers of the workpiece, (m) m is a heuristic factor, m' is a heuristicn,m′n′) The processing process that the current ant feeds the nth workpiece to the mth CNC, the nth 'workpiece on the mth CNC is fed, and the nth' workpiece is moved to the mth CNC for processing is shown, and tau (m) (m is the workpiece for processing)n,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC;
η(m’)＝min{max(MTmm’,LTm’)}，MTmm’time taken for current ant to move from m CNC to m' CNC, LTm’And the residual machining time is the time which is required by the machined workpiece on the CNC (m') until the machining is finished.
Further, according to the transition probability, selecting the CNC to which the current CNC is sent from the selectable CNC further comprises: selecting the CNC to which the current CNC goes from the selectable CNC through a roulette algorithm according to the transition probability;
the step S6 further includes: and updating the residual processing time of each CNC and updating an equipment dimension tabu table.
Further, the step S9 further includes: updating the pheromone concentration of each processing process through the following formula;
Figure GDA0002492734460000031
tau is the pheromone concentration before updating, tau' is the pheromone concentration after updating, I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing process, CiRepresents the completion time of each ant;
Figure GDA0002492734460000032
rho is a pheromone evaporation operator, and NC is the current iteration number.
Further, the intelligent RGV processing system comprises two processing procedures, wherein one part of CNC is used for processing a first procedure, and the other part of CNC is used for processing a second procedure; the unprocessed workpieces are referred to as raw material workpieces, the workpieces processed in the first process are referred to as semi-clinker workpieces, and the workpieces processed in the second process are referred to as clinker workpieces, and the scheduling logic comprises:
when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV does not have semi-clinker workpieces, feeding the semi-clinker workpieces on the CNC to the RGV;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no operation task exists;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, blanking the clinker workpieces on the CNC to the RGV, and feeding the semi-clinker workpieces on the RGV to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC has a clinker workpiece and the RGV does not have a semi-clinker workpiece, blanking the clinker workpiece on the CNC to the RGV;
when the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, feeding the semi-clinker workpiece on the RGV to the CNC;
when the RGV reaches the CNC processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, no operation task is performed.
In a second aspect, the present invention provides an intelligent RGV processing system scheduling apparatus, the apparatus comprising:
an initialization module for initializing ant population, pheromone concentration of each processing process, equipment dimension taboo table and remaining processing time of each CNC, wherein the ant population comprises NmEach ant carries a chromosome representing the feeding sequence, initial values are given to the chromosomes, and the chromosomes of all the ants form a feeding matrix; the device dimension tabu table records the current processing condition of each CNC and is used for determining which CNC the RGV can go to next step;
the optional CNC determining module is used for obtaining all CNC to which the current ants can go from the current CNC in the next step according to the equipment dimension taboo table and preset scheduling logic and taking the obtained CNC as the optional CNC;
the loading judging module is used for judging whether loading operation is needed according to the scheduling logic, and if the loading operation is needed, selecting the workpiece number of the corresponding position in the loading matrix to wait for loading;
the transfer probability calculation module is used for calculating the transfer probability of the current ants moving from the current CNC to each selectable CNC according to the pheromone concentration of each processing process and the residual processing time of each CNC, and selecting the next CNC going from the current CNC position from the selectable CNC according to the transfer probability;
the simulation operation module is used for moving the current ants from the current CNC to the to-be-sent CNC and performing operation according to the scheduling logic, wherein if the feeding operation is needed, the workpieces selected from the feeding matrix are fed to the corresponding CNC;
the first circulation module is used for repeating the selectable CNC determining module, the feeding judging module, the transition probability calculating module and the simulation operation module until all the N workpieces are processed to obtain a scheduling route;
a second circulation module for repeating the optional CNC determination module, the feeding judgment module, the transition probability calculation module, the simulation operation module and the first circulation module for the next ant until all N ants are traversedmAnts alone to obtain NmA scheduling route;
a recording module for converting NmRecording the shortest scheduling route used in each scheduling route;
genetic module for converting NmPerforming cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix;
the third circulation module is used for repeatedly selecting the CNC determining module, the feeding judging module, the transition probability calculating module, the simulation operation module, the first circulation module, the second circulation module, the recording module and the genetic module until the set total iteration times G are reached, and obtaining G recorded scheduling routes;
and the scheduling route determining module is used for selecting the scheduling route with the shortest time from the G recorded scheduling routes, namely the final scheduling route.
Further, the transition probability p of the current ant moving from the current m-th CNC to the m' th selectable CNC is calculated through the following formulam′：
Figure GDA0002492734460000061
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m 'are CNC serial numbers, m is a serial number of the current CNC, m' is a serial number of the selectable CNC, n and n 'are serial numbers of the workpiece, (m) m is a heuristic factor, m' is a heuristicn,m′n′) The processing process that the current ant feeds the nth workpiece to the mth CNC, the nth 'workpiece on the mth CNC is fed, and the nth' workpiece is moved to the mth CNC for processing is shown, and tau (m) (m is the workpiece for processing)n,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC;
η(m’)＝min{max(MTmm’,LTm’)}，MTmm’time taken for current ant to move from m CNC to m' CNC, LTm’And the residual machining time is the time which is required by the machined workpiece on the CNC (m') until the machining is finished.
Further, in the transition probability calculation module, according to the transition probability, selecting the CNC to which the current CNC is going from the selectable CNC further comprises: selecting the CNC to which the current CNC goes from the selectable CNC through a roulette algorithm according to the transition probability;
the step first loop module further comprises: and updating the residual processing time of each CNC and updating an equipment dimension tabu table.
Further, the second cycle module further comprises: updating the pheromone concentration of each processing process through the following formula;
Figure GDA0002492734460000062
tau is the pheromone concentration before updating, tau' is the pheromone concentration after updating, I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing process, CiRepresents the completion time of each ant;
Figure GDA0002492734460000071
rho is a pheromone evaporation operator, and NC is the current iteration number.
Further, the intelligent RGV processing system comprises two processing procedures, wherein one part of CNC is used for processing a first procedure, and the other part of CNC is used for processing a second procedure; the unprocessed workpieces are referred to as raw material workpieces, the workpieces processed in the first process are referred to as semi-clinker workpieces, and the workpieces processed in the second process are referred to as clinker workpieces, and the scheduling logic comprises:
when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV does not have semi-clinker workpieces, feeding the semi-clinker workpieces on the CNC to the RGV;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no operation task exists;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, blanking the clinker workpieces on the CNC to the RGV, and feeding the semi-clinker workpieces on the RGV to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC has a clinker workpiece and the RGV does not have a semi-clinker workpiece, blanking the clinker workpiece on the CNC to the RGV;
when the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, feeding the semi-clinker workpiece on the RGV to the CNC;
when the RGV reaches the CNC processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, no operation task is performed.
The invention has the following beneficial effects:
according to the invention, the node sequence is optimized through the ant colony algorithm, the feeding sequence is optimized through the genetic algorithm, the genetic algorithm makes up for the characteristic of weak global optimization capability of the ant colony algorithm, the parallel search capability of the ant colony algorithm is strong, the complementation is realized, and the optimization capability of the algorithm is improved.
Drawings
FIG. 1 is a schematic view of one embodiment of a smart RGV processing system of the present invention;
FIG. 2 is a schematic diagram of the intelligent RGV processing system scheduling method of the present invention;
FIG. 3 is a diagram of an example of chromosomal codes of the present invention;
FIG. 4 is a diagram showing an example of a chromosome crossing process of the present invention;
FIG. 5 is a diagram illustrating an exemplary process of chromosomal variation according to the present invention;
FIG. 6 is a flowchart of chromosomal variation according to the present invention;
FIG. 7 is a diagram of a pheromone evaporation operator Rho according to the present invention;
FIG. 8 is an exemplary diagram of scheduling logic in accordance with the present invention;
FIG. 9 is a flow chart of a method of scheduling an intelligent RGV processing system of the present invention;
fig. 10 is a schematic diagram of the intelligent RGV processing system scheduling apparatus of the present invention.
Detailed Description
In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
In one aspect, an embodiment of the present invention provides a method for scheduling an intelligent RGV processing system, where a processing mode of the intelligent RGV processing system is: the machining time of each machining process of each workpiece is different, and each process is machined by a specific group of CNC parallel machines. The goal of scheduling is to obtain a machining path (scheduling route) such that the total machining time (Makespan) is minimized.
As shown in fig. 2, the method comprises the steps of:
step S1: initializing ant population, pheromone concentration of each processing process, equipment dimension tabu and residual processing time of each CNC, wherein the ant population comprises NmAnd each ant carries a chromosome representing the feeding sequence, the chromosomes are assigned with initial values, and the chromosomes of all the ants form a feeding matrix.
The invention optimizes the transfer process of the RGV among the CNC by using the ant colony algorithm and optimizes the feeding sequence of the workpiece by using the genetic algorithm. When the algorithm starts, the related parameters of the ant colony algorithm and the genetic algorithm need to be initialized.
The basic idea of applying the ant colony algorithm to solve the optimization problem is as follows: the feasible solution of the problem to be optimized is represented by the walking paths of the ants, and all paths of the whole ant colony form a solution space of the problem to be optimized. The shorter ants release more pheromone (because the shorter the path, the more the ants go back and forth in the same time, the more pheromones), and as the time advances, the concentration of the pheromone accumulated on the shorter path gradually increases, and the number of ants selecting the path also increases. Finally, the whole ant can be concentrated on the optimal path under the action of positive feedback, and the corresponding optimal solution of the problem to be optimized is obtained.
Initialization of the ant colony includes the number of ants NmDetermining the total iteration number G of the current iteration number NC (initialized to 0), initializing an information heuristic factor α and an expected heuristic factor β, assigning an initial value 1 to the current iteration number NC, initializing a coefficient factor β and the like, wherein in the ant colony algorithm, the walking path of an ant is simulated to be the dispatching path of the RGV.
The pheromone concentration of each process also needs to be initialized, and a process refers to a minimum no longer-separable section of the complete processing path, and is a process of transferring the RGV from one CNC to the next CNC and performing operations (such as feeding, blanking and the like) on the two CNC. For example: the RGV feeds the machined No. 2 workpiece on the No. 1 CNC to the RGV, and the RGV feeds the No. 1 workpiece on the No. 1 CNC, and then the RGV is transferred to the No. 8 CNC, and the No. 2 workpiece is further machined up to the No. 8 CNC, which is one machining process.
Each processing process is assigned with an initial pheromone concentration, the ant colony algorithm aims to update the pheromone concentration of each processing process through a plurality of iterative processes, the higher the pheromone concentration is, the faster the processing process is, the easier the processing process is to select, and a plurality of processing processes form a complete processing path (scheduling route).
And the equipment dimension tabu table records the current processing condition of each CNC and judges whether the current RGV needs to go to the CNC for operation or not by combining the workpiece condition on the current RGV, and the purpose of the equipment dimension tabu table is to determine which CNC the RGV can go to next step. An example of a device dimension tabu table is shown in table 1.
Table 1: device dimension taboo table
Figure GDA0002492734460000101
In 1, each ant corresponds to one row, each CNC corresponds to one column, and the variable 0-1 represents the current processing condition of each CNC. 0 represents the workpiece which is not machined by the CNC, 1 represents the workpiece which is machined or machined but not blanked on the CNC, and the initial value of the equipment dimension tabu table is 0.
The remaining machining time of each CNC means the time required for the workpiece being machined on the CNC to be completed, and the initial value is 0.
The initialization of the genetic algorithm mainly comprises the initialization of chromosome coding and related parameters (cross probability Pc and variation probability Pm). The invention combines the genetic algorithm and the ant colony algorithm, each ant carries a chromosome, the chromosome code is the serial number of the workpiece, and the chromosome sequence represents the feeding sequence of the workpiece. Examples are as follows: assuming that N is 7, i.e. 7 workpieces are processed, fig. 3 shows an embodiment of a chromosome, which represents that the corresponding ants are processed sequentially with the raw material numbered 3657142.
Furthermore, each ant corresponds to a chromosome representing the workpiece feeding sequence, so that the chromosomes of all ants form a workpiece feeding matrix, and the chromosomes are randomly assigned initial values to obtain an initial feeding matrix. One embodiment of the loading matrix may be represented by table 2 below. Each row of table 2 corresponds to a chromosome of each ant, and each column represents a sequential loading sequence of workpieces.
Table 2: workpiece feeding matrix
Figure GDA0002492734460000102
Figure GDA0002492734460000111
Step S2: and obtaining all the CNC to which the current ant (the serial number of the ant is marked as i) can go from the current CNC next step according to the equipment dimension taboo table and preset scheduling logic as the optional CNC.
As previously mentioned, the role of the device dimension tabu table is to determine which CNC the RGV can go to next and which CNC it cannot go to. And the scheduling logic refers to operations (such as loading, unloading, no operation, etc.) that can be made after the RGV (ant) goes from the current CNC to the next CNC. The scheduling logic is related to a machining process (whether it is the first process, the second process, or the … th process) of the next CNC to be addressed, a state of the RGV (ant) (whether a workpiece is carried on the RGV, etc.), presence/absence of a workpiece on the next CNC to be addressed, presence/absence of a workpiece (raw material workpiece) which has not started to be machined, and the like.
The invention judges which CNC the RGV can go to next step through the device dimension taboo table, and removes the CNC which the RGV cannot go to. The pre-established scheduling logic determines what operations can be done after the next CNC is reached, and removes the CNC which cannot be operated (or obviously invalid operation) after the next CNC is reached, and all the CNC which can be go to next step is the optional CNC.
Step S3: and judging whether the loading operation is needed according to the scheduling logic, and if the loading operation is needed, selecting the workpiece number at the corresponding position in the loading matrix to wait for loading.
The loading operation is to load the workpieces (raw material workpieces) which are not started to be machined onto the CNC of the first process, and if ants with the current serial number i have already loaded p raw material workpieces and judge that the next node to go needs to be loaded, the workpieces corresponding to the position of [ i, p +1] are selected from the workpiece loading matrix.
Step S4: and calculating the transition probability of the current ants moving from the current CNC to each optional CNC according to the pheromone concentration of each machining process and the residual machining time of each CNC, and selecting the next CNC going from the current CNC position from the optional CNC according to the transition probability.
The pheromone concentration of course of working and CNC's remaining process time can both influence the speed of processing, consequently synthesizes and considers both, obtains the transition probability to next CNC, can determine the CNC that goes according to the transition probability, also promptly the shortest CNC consuming time.
Step S5: and moving the current ants from the current CNC to the destined CNC, and carrying out operation according to scheduling logic, wherein if the loading operation is required, the workpieces selected from the loading matrix are loaded to the corresponding CNC.
And after the next CNC to be visited is determined, simulation operation can be carried out according to the scheduling logic, and if loading is needed in the scheduling logic, the workpiece corresponding to the position of [ i, p +1] selected from the loading matrix can be loaded.
Step S6: and repeating the step S2 to the step S5 until all the N workpieces are processed, and obtaining a scheduling route.
The steps S2 to S5 are to optimize the path from the current CNC to the next CNC, which is only one section of the complete dispatching route. At this time, the ants (RGV) arrive at the next CNC, and the former next CNC becomes the current CNC, then the steps S2 to S5 are repeated, and the next route is continuously optimized until all the N workpieces are processed, so as to obtain a dispatching route.
Step S7: repeating the steps S2-S6 for the next ant until all N ants are traversedmAnts alone to obtain NmAnd scheduling the route.
The scheduling route is obtained by optimizing an ant in an iteration process, wherein the ant group has NmOnly ants are needed, and the steps are repeated for each ant, so that N corresponding to the iteration can be obtainedmAnd scheduling the route.
Step S8: will NmThe shortest time-consuming scheduling route is recorded in the scheduling routes.
N corresponding to the iterationmThe shortest route, which may be the route that we need, is recorded.
Step S9: will NmAnd performing cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix.
The foregoing steps S2 to S8 are an iterative process of the ant colony algorithm optimizing the process of transferring the RGV between CNC. In the iteration process, the invention also uses a genetic algorithm to optimize the workpiece feeding sequence (step S9).
And (3) a crossing process: in the present invention, the result of chromosome crossing is to ensure that both chromosomes contain genes from 1 to N, and is not redundant. If there are gene duplications on the same chromosome after crossover, the following conflict-exchange approach should be used, as shown in FIG. 4. When two chromosomes A and B are crossed, if the gene sequence numbers in the chromosome A are repeated (assumed to be a and a '), finding out a repeated gene a and a gene B of the gene at the corresponding position of the chromosome B in the uncrossed segment of the chromosome A, and interchanging the gene B and the gene a' in the crossed segment of the chromosome A, thereby completing one-time conflict interchange. By analogy, the genes of two chromosomes contain 1 to N genes, so that the method is not heavy.
And (3) mutation process: generation of [0, 1] by a program]Random number R within rangeaWhen R isaLess than a mutation probability set value PmThen, chromosomal mutation is performed. The mutation process is first programmed to generate [1, N]Integer random number in range RbR is to bebAnd N-RbThe gene individuals corresponding to the positions are exchanged, so that the variation is generated, as shown in FIG. 5.
In particular, when the number of processed workpieces N is even and RbN-R is exactly N/2bThe position is RbPosition, unable to be exchanged, then a new random number R is regeneratedbThe general process of chromosomal variation is shown in FIG. 6.
Step S10: and repeating the steps S2 to S9 until the set total iteration number G is reached, and obtaining G recorded scheduling routes.
The foregoing steps S2 to S9 are an iterative process, and the present invention records a plurality of scheduling routes through a plurality of iterations.
Step S11: and selecting the scheduling route with the shortest time from the G recorded scheduling routes, namely the final scheduling route.
The method adopts a mode of nesting the ant colony algorithm by the genetic algorithm, the genetic algorithm optimizes the workpiece feeding sequence, and the ant colony algorithm optimizes the RGV transfer process at the node. Each ant is provided with a chromosome recorded with a workpiece feeding sequence and a taboo table recorded with CNC (computerized numerical control) processing information, nodes which cannot be visited by the current ant are excluded through the taboo table, and the selectable CNC is obtained by combining scheduling logic. And determining the next state transition node of the ant through the calculated transition probability, transferring to the CNC and determining the specific job content through the scheduling logic. When the workpieces are completely processed, the ant optimizing process is finished; and when all ants finish optimizing, recording the optimal solution and carrying out cross variation on the chromosome, and finishing the iteration. And when the iteration times reach a set value, finishing the program to obtain a final optimal scheduling route.
According to the invention, the node sequence is optimized through the ant colony algorithm, the feeding sequence is optimized through the genetic algorithm, the genetic algorithm makes up for the characteristic of weak global optimization capability of the ant colony algorithm, the parallel search capability of the ant colony algorithm is strong, the complementation is realized, and the optimization capability of the algorithm is improved.
In the invention, the pheromone concentration of each processing process is preferably represented by a pheromone table, the pheromone table is designed into two layers, the coordinates 1-K of the outer layer correspond to the serial number of a CNC (computer numerical control), the coordinates 1-N of the inner layer represent the serial number of a workpiece for feeding/discharging, the data in each small grid is the pheromone concentration tau of one processing process, and the concentrations of the initialized pheromone table at the beginning of iteration are tau0Specific design is shown in table 3, which is 0.1.
Table 3: pheromone table
Figure GDA0002492734460000141
Since the same workpiece cannot be machined twice on one CNC, the pheromone accumulation is 0. In table 3, each box represents the pheromone concentration for one process. For example, by "", it is meant that the RGV feeds the machined workpiece No. 2 on CNC No. 1 to the RGV, and the RGV feeds the workpiece No. 1 on CNC No. 1, and then the RGV is transferred to CNC No. 8, and the workpiece No. 2 is further machined up to CNC No. 8. Therefore, the pheromone table can convert the scheduling scheme of the original three-dimensional problem (workpiece dimension, process dimension and CNC equipment dimension) into a two-dimensional table record, and is more intuitive.
Based on the record form of the pheromone table, the invention can calculate that the current ant moves from the current m-th CNC to the m' -th selectable numberTransition probability p of CNCm′：
Figure GDA0002492734460000151
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m ' are CNC serial numbers, m is a serial number of the current CNC, m ' is a serial number of the selectable CNC, n and n ' are serial numbers of the workpiece, m is an information heuristic factor, m is a sequence number of the current CNC, m is a sequence number of the selectable CNC, n is a sequence number of the workpiece, n is a sequence number of thenWherein m represents a CNC number (outer layer of the pheromone table), n represents a workpiece number (inner layer of the pheromone table), (m) represents a CNC number (outer layer of the pheromone table), andn,m′n′) The machining process (refer to the pheromone table) of 'loading the nth workpiece to the mth CNC by the current ants, unloading the nth' workpiece on the mth CNC and moving the nth 'workpiece to the mth CNC for machining' is shown, and tau (m) (mn,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC, N being the total number of workpieces.
η (m ') is heuristic information, η (m') is min { Max (MT)mm’,LTm’)}，。
When receiving a plurality of CNC loading and unloading instructions, the RGV needs to judge which CNC is going to carry out loading and unloading operation, so that heuristic information of each CNC such as a formula is formulated, and ants can select the going CNC according to the heuristic information of the CNC.
m denotes the current RGV CNC position, m' denotes the next possible CNC number to go to, MTmm’Time taken for current ant to move from m CNC to m' CNC, LTm’The remaining machining time of the m ' th CNC is the time required by the workpiece being machined on the m ' th CNC to finish machining, and the meaning is the fastest machining starting time of the m ' number CNC. In particular, when the m' numbered CNC does not have a workpiece being machined, then its LTm’The CNC representing the m' number can be put into a machining state at any time when it is 0.
After the transition probability is determined, theoretically, the CNC with the highest transition probability is the CNC to which the transition probability is most likely to go, i.e., the CNC to go to. However, this may be trapped in the local optimum, and therefore, the present invention selects the CNC to which the current CNC is destined from the selectable CNC by the roulette algorithm according to the transition probability, thereby avoiding the trapping in the local optimum.
Step S6 further includes: and updating the residual processing time of each CNC and updating an equipment dimension tabu table. In the invention, in the optimization process of one ant, the remaining processing time and the equipment dimension taboo table of each CNC are required to be updated.
In the invention, after each iteration is finished, the pheromone concentration (pheromone table) is required to be updated according to the optimizing track of each ant. Step S9 further includes: updating the pheromone concentration of each processing process through the following formula;
Figure GDA0002492734460000161
take the pheromone update at "+" as an example in table 3, where Rho is pheromone evaporation operator, τ is pheromone concentration before update at "+", τ' is pheromone concentration after update at "+", I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing procedure (i.e., operation at "+"), CiThe completion time (Makespan) of each ant is shown.
Rho has the formula:
Figure GDA0002492734460000162
the ant colony algorithm inevitably has the following defects besides the advantages of a feedback mechanism, parallel computation, distributed computation, difficulty in falling into a global optimal solution and the like: 1) the number of pheromones in the early stage of iteration is small, the feedback mechanism is weak, and the searching capability is not strong; 2) all ants tend to the same path easily appearing in the middle and later stages of the iteration, resulting in earlier convergence of the algorithm.
In order to solve the problems, the invention provides a self-adaptive pheromone evaporation operator Rho, wherein G is set total iteration times; NC is the current iteration number when the program runs. FIG. 7 shows Rho as a function of NC for G taken 50, 100, 500, 1000 times, respectively.
As can be seen from fig. 7, the pheromone evaporation operator satisfies the following forms and advantages regardless of the change of the preset iteration number G:
in the initial stage of iteration, the pheromone evaporation operator is between [1 and 3], the function of strengthening the initial pheromone is achieved, the pheromone in the pheromone table can be rapidly accumulated in the early stage of iteration, and the defect that the pheromone of the ant colony algorithm is difficult to accumulate in the initial stage of iteration is well overcome.
In the early stage of iteration, the pheromone evaporation operator is rapidly reduced to below 1 (but is guaranteed to be above 0.5), so that pheromones are prevented from being accumulated early, a solution set is disturbed, and a local optimal point is avoided.
The pheromone evaporation operator slowly increases (slowly volatilizes) in the later stages of the iteration in order to gradually accumulate good pheromones, so that the algorithm converges orderly.
At the later stage of iteration, the pheromone infinitely tends to 1, so that the optimization result of the ant colony finally converges on a solution.
The pheromone evaporation operator provided by the invention solves the problems of the ant colony algorithm in the prior art.
The scheduling logic of the invention is related to the system composition of the intelligent RGV processing system, when the system composition is determined, all the conditions occurring in the scheduling are fully considered, induction summary is carried out, the obvious meaningless conditions are abandoned, and the scheduling logic can be obtained. One specific example is given below: as shown in fig. 1, the intelligent RGV processing system of this example includes 1 rail-type RGV for carrying and loading and unloading workpieces, 1 RGV linear rail, K (preferably 8) CNC for processing workpieces, 1 loading conveyor, 1 unloading conveyor, and other accessories. The RGV is provided with a mechanical arm, two mechanical claws for grabbing/placing materials are arranged on the mechanical arm, and a material cleaning tank for cleaning the final material before leaving the processing system is arranged, so that the operation tasks of loading, unloading, material cleaning and the like can be completed.
This example includes two machining passes for machining N different workpieces, one part CNC for machining the first pass and another part CNC for machining the second pass, preferably. 4 CNC (CNC1, CNC3, CNC5 and CNC7) on one side are processed in a first process; and 4 CNC (CNC2, CNC4, CNC6 and CNC8) on the other side are processed in a second process. The unprocessed work piece is called a raw material work piece, the work piece processed in the first step is called a semi-clinker work piece, and the work piece processed in the second step is called a clinker work piece.
The state of the RGV in the dispatching process has two possibilities of carrying the semi-clinker workpiece and not carrying the semi-clinker workpiece. RGVs may go to CNC that have both a first pass and a second pass of machining, and that have both a workpiece on and a workpiece off. For the system, there is the possibility that both the raw meal pieces have been completely charged and the raw meal has not been completely charged.
For the above cases, the total number of the theoretical values is 2416 branches. However, according to the actual situation, the branches without discussion meaning are deleted, and the similar situations are merged, and the scheduling logic of the system is as follows:
1. when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock; this case includes the following two branches:
1) when the RGV reaches the CNC processing the first process and the raw material is not completely loaded at this time, if there are workpieces (semi-clinker workpieces) being processed on the CNC and a manipulator of the RGV carries the semi-clinker workpieces, the RGV needs to wait for the workpieces processed on the CNC to be removed to load the CNC at this time, and the CNC needs to wait for the semi-clinker workpieces on the RGV to be removed to load the CNC, so that the CNC enters into a deadlock.
2) When the RGV reaches the CNC of the first process and all raw materials are loaded at the time, if the CNC is provided with workpieces being processed and a manipulator of the RGV is loaded with semi-clinker, the RGV needs to wait for the workpieces processed on the CNC to be removed to load the CNC, and the CNC also waits for the semi-clinker workpieces on the RGV to be removed to carry out blanking, so that deadlock is involved.
2. When the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the RGV is not fed again because the raw material is fed completely.
3. When the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV does not have the semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV.
4. When the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no matter whether the RGV is loaded with the semi-clinker or not, the RGV can not be fed to the CNC or can not be fed from the CNC at the moment because the raw material is fed completely, and therefore no operation task exists.
5. When the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC. This case includes the following two branches:
1) when the RGV reaches the CNC of the first process and the raw material is not completely loaded at this time, if there is no workpiece being processed on the CNC and one manipulator of the RGV carries the semi-clinker, the RGV can only be used by the other manipulator to pick up the raw material and put it on the CNC for processing.
2) When the RGV reaches the CNC of the first process and the raw meal is not completely loaded at this time, if there is no work piece being processed on the CNC and the RGV is not loaded with semi-clinker, the RGV grabs the raw meal with one of the manipulators and puts it on the CNC for processing.
6. When the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, the clinker workpieces on the CNC are fed to a cleaning groove of the RGV, a mechanical arm is rotated to feed the semi-clinker workpieces on the RGV to the CNC, and finally the semi-clinker in the cleaning groove is placed on a feeding worktable to be transported away.
7. When the RGV reaches the CNC for processing the second procedure, if the CNC has clinker workpieces and the RGV does not have semi-clinker workpieces, the clinker workpieces on the CNC are fed to the RGV and put into a cleaning groove, and then the semi-clinker in the cleaning groove is put on a feeding worktable and conveyed away.
8. When the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, the semi-clinker workpiece on the RGV is fed to the CNC.
9. When the RGV reaches the CNC for processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, the RGV can not be used for CNC feeding and blanking from the CNC, and no operation task is required.
The scheduling logic for the 9 cases, 11 branches, is shown in fig. 8.
Finally, the specific scheduling process of the present invention is illustrated in an overall flowchart, see fig. 9. Wherein NC represents the current iteration times, G represents the set iteration total times, i is the serial number of ants, p is the number of the fed raw materials, q is the number of the discharged clinker, N represents the total number of the processed workpiecesmRepresenting the total number of ants in the ant colony.
In another aspect, an embodiment of the present invention provides an intelligent RGV processing system scheduling apparatus, as shown in fig. 10, the apparatus includes:
an initialization module 1 for initializing ant population, pheromone concentration of each processing procedure, equipment dimension taboo table and remaining processing time of each CNC, the ant population including NmEach ant carries a chromosome representing the feeding sequence, initial values are given to the chromosomes, and the chromosomes of all the ants form a feeding matrix; the device dimension tabu table records the current machining condition of each CNC and is used for determining which CNC the RGV can go to next.
The module also initializes the current iteration number NC (initialized to 0), the cross probability Pc, the variation probability Pm, the information heuristic factor α and the expectation heuristic factor β.
And the optional CNC determining module 2 is used for obtaining all the CNC to which the current ants can go from the current CNC in the next step according to the equipment dimension taboo table and preset scheduling logic and taking the obtained CNC as the optional CNC.
And the loading judging module 3 is used for judging whether loading operation is required according to the scheduling logic, and if the loading operation is required, selecting the workpiece number at the corresponding position in the loading matrix to wait for loading.
And the transition probability calculation module 4 is used for calculating the transition probability of the current ants moving from the current CNC to each optional CNC according to the pheromone concentration of each machining process and the residual machining time of each CNC, and selecting the next CNC going from the current CNC position from the optional CNC according to the transition probability.
And the simulation operation module 5 is used for moving the current ants from the current CNC to the destined CNC and performing operation according to the scheduling logic, wherein if the loading operation is required, the workpieces selected from the loading matrix are loaded to the corresponding CNC.
And the first circulation module 6 is used for repeating the selectable CNC determining module, the feeding judging module, the transition probability calculating module and the simulation operation module until all the N workpieces are processed, so as to obtain a scheduling route.
A second circulation module 7 for repeating the optional CNC determination module, the feeding judgment module, the transition probability calculation module, the simulation operation module and the first circulation module for the next ant until all N ants are traversedmAnts alone to obtain NmAnd scheduling the route.
A recording module 8 for converting NmThe shortest time-consuming scheduling route is recorded in the scheduling routes.
Genetic module 9 for converting NmAnd performing cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix.
And the third circulation module 10 is used for repeatedly selecting the CNC determining module, the feeding judging module, the transition probability calculating module, the simulation operation module, the first circulation module, the second circulation module, the recording module and the genetic module until the set total iteration number G is reached, and obtaining G recorded scheduling routes.
And the scheduling route determining module 11 is configured to select a scheduling route that is shortest in time from the G recorded scheduling routes, that is, a final scheduling route.
According to the invention, the node sequence is optimized through the ant colony algorithm, the feeding sequence is optimized through the genetic algorithm, the genetic algorithm makes up for the characteristic of weak global optimization capability of the ant colony algorithm, the parallel search capability of the ant colony algorithm is strong, the complementation is realized, and the optimization capability of the algorithm is improved.
The invention calculates the transfer probability p of the current ant moving from the current m-th CNC to the m' -th selectable CNC through the following formulam′：
Figure GDA0002492734460000211
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m 'are CNC serial numbers, m is a serial number of the current CNC, m' is a serial number of the selectable CNC, n and n 'are serial numbers of the workpiece, (m) m is a heuristic factor, m' is a heuristicn,m′n′) The processing process that the current ant feeds the nth workpiece to the mth CNC, the nth 'workpiece on the mth CNC is fed, and the nth' workpiece is moved to the mth CNC for processing is shown, and tau (m) (m is the workpiece for processing)n,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC.
η(m’)＝min{max(MTmm’,LTm’)}，MTmm’Time taken for current ant to move from m CNC to m' CNC, LTm’The remaining machining time is the time required by the workpiece being machined on the m' th CNC until the machining is completed.
In the foregoing transition probability calculating module, according to the transition probability, the CNC to which the current CNC is selected from the selectable CNC is further: and selecting the CNC to which the current CNC goes from the selectable CNC through a roulette algorithm according to the transition probability.
The step first loop module may further include: and updating the residual processing time of each CNC and updating an equipment dimension tabu table.
In the present invention, the second circulation module may further include: updating the pheromone concentration of each processing process through the following formula;
Figure GDA0002492734460000221
tau is the pheromone concentration before updating, tau' is the pheromone concentration after updating, I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing process, CiIndicating the completion time of each ant.
Figure GDA0002492734460000222
Rho is a pheromone evaporation operator, and NC is the current iteration number.
In one embodiment of the smart RGV processing system of the present invention, two processing steps are included, one part of CNC is used for processing the first step and the other part of CNC is used for processing the second step; the scheduling logic includes:
when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV does not have semi-clinker workpieces, feeding the semi-clinker workpieces on the CNC to the RGV;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no operation task exists;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, blanking the clinker workpieces on the CNC to the RGV, and feeding the semi-clinker workpieces on the RGV to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC has a clinker workpiece and the RGV does not have a semi-clinker workpiece, blanking the clinker workpiece on the CNC to the RGV;
when the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, feeding the semi-clinker workpiece on the RGV to the CNC;
when the RGV reaches the CNC processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, no operation task is performed.
The device provided by the embodiment of the present invention has the same implementation principle and technical effect as the method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the method embodiments without reference to the device embodiments. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the apparatus and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (8)

1. A method of scheduling an intelligent RGV processing system, the method comprising:
step S1: initializing an ant population, an pheromone concentration of each process, an equipment dimension tabu, and a remaining process time for each CNC, the ant population including NmEach ant carries a chromosome representing the feeding sequence, initial values are given to the chromosomes, and the chromosomes of all the ants form a feeding matrix; the device dimension tabu table records the current processing condition of each CNC and is used for determining which CNC the RGV can go to next step;
step S2: obtaining all the CNC to which the current ants can go from the current CNC in the next step as optional CNC according to the equipment dimension taboo table and preset scheduling logic;
step S3: judging whether loading operation is needed according to the scheduling logic, and if the loading operation is needed, selecting workpiece numbers at corresponding positions in the loading matrix to wait for loading;
step S4: calculating the transition probability of the current ant moving from the current CNC to each selectable CNC according to the pheromone concentration of each processing process and the remaining processing time of each CNC, and selecting the next CNC going from the current CNC position from the selectable CNC according to the transition probability;
step S5: moving the current ants from the current CNC to the to-be-sent CNC, and carrying out operation according to scheduling logic, wherein if loading operation is required, workpieces selected from a loading matrix are loaded to the corresponding CNC;
step S6: repeating the step S2 to the step S5 until all N workpieces are processed, and obtaining a scheduling route;
step S7: repeating the steps S2-S6 for the next ant until all N ants are traversedmAnts alone to obtain NmA scheduling route;
step S8: will NmRecording the shortest scheduling route used in each scheduling route;
step S9: will NmPerforming cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix;
step S10: repeating the step S2 to the step S9 until the set total iteration times G are reached, and obtaining G recorded dispatching routes;
step S11: selecting the scheduling route with the shortest time consumption from the G recorded scheduling routes, namely the final scheduling route;
the intelligent RGV processing system comprises two processing procedures, wherein one part of CNC is used for processing a first procedure, and the other part of CNC is used for processing a second procedure; the unprocessed workpieces are referred to as raw material workpieces, the workpieces processed in the first process are referred to as semi-clinker workpieces, and the workpieces processed in the second process are referred to as clinker workpieces, and the scheduling logic comprises:
when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV does not have semi-clinker workpieces, feeding the semi-clinker workpieces on the CNC to the RGV;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no operation task exists;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, blanking the clinker workpieces on the CNC to the RGV, and feeding the semi-clinker workpieces on the RGV to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC has a clinker workpiece and the RGV does not have a semi-clinker workpiece, blanking the clinker workpiece on the CNC to the RGV;
when the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, feeding the semi-clinker workpiece on the RGV to the CNC;
when the RGV reaches the CNC processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, no operation task is performed.
2. The scheduling method of an intelligent RGV processing system as claimed in claim 1, wherein the transition probability p for the current ant to move from the current m CNC to the m' CNC is calculated by the following formulam′：
Figure FDA0002492734450000031
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m 'are CNC serial numbers, m is a serial number of the current CNC, m' is a serial number of the selectable CNC, n and n 'are serial numbers of the workpiece, (m) m is a heuristic factor, m' is a heuristicn,m′n′) The processing process that the current ant feeds the nth workpiece to the mth CNC, the nth 'workpiece on the mth CNC is fed, and the nth' workpiece is moved to the mth CNC for processing is shown, and tau (m) (m is the workpiece for processing)n,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC;
η(m’)＝min{max(MTmm’,LTm’)}，MTmm’time taken for current ant to move from m CNC to m' CNC, LTm’And the residual machining time is the time which is required by the machined workpiece on the CNC (m') until the machining is finished.
3. The intelligent RGV processing system scheduling method of claim 2, wherein the selecting the CNC to which the current CNC goes from the selectable CNC according to the transition probability further comprises: selecting the CNC to which the current CNC goes from the selectable CNC through a roulette algorithm according to the transition probability;
the step S6 further includes: and updating the residual processing time of each CNC and updating an equipment dimension tabu table.
4. The intelligent RGV processing system scheduling method of claim 3, wherein the step S9 further includes: updating the pheromone concentration of each processing process through the following formula;
Figure FDA0002492734450000032
tau is the pheromone concentration before updating, tau' is the pheromone concentration after updating, I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing process, CiRepresents the completion time of each ant;
Figure FDA0002492734450000041
rho is a pheromone evaporation operator, and NC is the current iteration number.
5. An intelligent RGV processing system scheduling apparatus, the apparatus comprising:
an initialization module for initializing ant population, pheromone concentration of each processing process, equipment dimension taboo table and remaining processing time of each CNC, wherein the ant population comprises NmEach ant carries a chromosome representing the feeding sequence, initial values are given to the chromosomes, and the chromosomes of all the ants form a feeding matrix; the device dimension tabu table records the current processing condition of each CNC and is used for determining which CNC the RGV can go to next step;
the optional CNC determining module is used for obtaining all CNC to which the current ants can go from the current CNC in the next step according to the equipment dimension taboo table and preset scheduling logic and taking the obtained CNC as the optional CNC;
the loading judging module is used for judging whether loading operation is needed according to the scheduling logic, and if the loading operation is needed, selecting the workpiece number of the corresponding position in the loading matrix to wait for loading;
the transfer probability calculation module is used for calculating the transfer probability of the current ants moving from the current CNC to each selectable CNC according to the pheromone concentration of each processing process and the residual processing time of each CNC, and selecting the next CNC going from the current CNC position from the selectable CNC according to the transfer probability;
the simulation operation module is used for moving the current ants from the current CNC to the to-be-sent CNC and performing operation according to the scheduling logic, wherein if the feeding operation is needed, the workpieces selected from the feeding matrix are fed to the corresponding CNC;
the first circulation module is used for repeating the selectable CNC determining module, the feeding judging module, the transition probability calculating module and the simulation operation module until all the N workpieces are processed to obtain a scheduling route;
a second circulation module for repeating the optional CNC determination module, the feeding judgment module, the transition probability calculation module, the simulation operation module and the first circulation module for the next ant until all N ants are traversedmAnts alone to obtain NmA scheduling route;
a recording module for converting NmRecording the shortest scheduling route used in each scheduling route;
genetic module for converting NmPerforming cross operation on the chromosomes of the ants, performing mutation operation, and updating the feeding matrix;
the third circulation module is used for repeatedly selecting the CNC determining module, the feeding judging module, the transition probability calculating module, the simulation operation module, the first circulation module, the second circulation module, the recording module and the genetic module until the set total iteration times G are reached, and obtaining G recorded scheduling routes;
the scheduling route determining module is used for selecting the scheduling route with the shortest time from the G recorded scheduling routes, namely the final scheduling route;
the intelligent RGV processing system comprises two processing procedures, wherein one part of CNC is used for processing a first procedure, and the other part of CNC is used for processing a second procedure; the unprocessed workpieces are referred to as raw material workpieces, the workpieces processed in the first process are referred to as semi-clinker workpieces, and the workpieces processed in the second process are referred to as clinker workpieces, and the scheduling logic comprises:
when the RGV reaches the CNC of the first procedure, if the CNC has a semi-clinker workpiece and the RGV carries the semi-clinker workpiece, then the RGV is trapped into deadlock;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if the CNC has semi-clinker workpieces and the RGV does not have semi-clinker workpieces, feeding the semi-clinker workpieces on the CNC to the RGV;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if the CNC has the semi-clinker workpieces and the RGV has no semi-clinker workpieces, the semi-clinker workpieces on the CNC are fed to the RGV, and the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC of the first procedure and all raw material workpieces are fed, if no semi-clinker workpiece exists on the CNC, no operation task exists;
when the RGV reaches the CNC of the first procedure and the raw material workpieces are not all loaded, if no semi-clinker workpieces exist on the CNC, the raw material workpieces are loaded to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC is provided with clinker workpieces and the RGV is loaded with semi-clinker workpieces, blanking the clinker workpieces on the CNC to the RGV, and feeding the semi-clinker workpieces on the RGV to the CNC;
when the RGV reaches the CNC for processing the second procedure, if the CNC has a clinker workpiece and the RGV does not have a semi-clinker workpiece, blanking the clinker workpiece on the CNC to the RGV;
when the RGV reaches the CNC for processing the second procedure, if the CNC does not have a clinker workpiece and the RGV carries a semi-clinker workpiece, feeding the semi-clinker workpiece on the RGV to the CNC;
when the RGV reaches the CNC processing the second procedure, if the CNC has no clinker workpiece and the RGV has no semi-clinker workpiece, no operation task is performed.
6. The scheduling apparatus of an intelligent RGV processing system as claimed in claim 5, wherein the transition probability p for the current ant to move from the current m CNC to the m' CNC is calculated by the following formulam′：
Figure FDA0002492734450000061
Wherein α is an information heuristic factor, β is an expected heuristic factor, m and m 'are CNC serial numbers, m is a serial number of the current CNC, m' is a serial number of the selectable CNC, n and n 'are serial numbers of the workpiece, (m) m is a heuristic factor, m' is a heuristicn,m′n′) The processing process that the current ant feeds the nth workpiece to the mth CNC, the nth 'workpiece on the mth CNC is fed, and the nth' workpiece is moved to the mth CNC for processing is shown, and tau (m) (m is the workpiece for processing)n,m′n′) Is represented by (m)n,m′n′) Pheromone concentration during processing, JkIs the set of all CNC and the matched loading workpieces, in which set m, m' is 1 … K; n, N ═ 1 … N, K being the total number of CNC;
η(m’)＝min{max(MTmm’,LTm’)}，MTmm’time taken for current ant to move from m CNC to m' CNC, LTm’And the residual machining time is the time which is required by the machined workpiece on the CNC (m') until the machining is finished.
7. The intelligent RGV processing system scheduling apparatus of claim 6 wherein the transition probability calculation module, based on the transition probability, further selects the CNC to which the current CNC is going from the selectable CNC as: selecting the CNC to which the current CNC goes from the selectable CNC through a roulette algorithm according to the transition probability;
the step first loop module further comprises: and updating the residual processing time of each CNC and updating an equipment dimension tabu table.
8. The smart RGV processing system scheduler of claim 7, wherein the second cycle module further comprises: updating the pheromone concentration of each processing process through the following formula;
Figure FDA0002492734450000071
tau is the pheromone concentration before updating, tau' is the pheromone concentration after updating, I represents the serial number of each ant, I represents the serial number of the ant corresponding to the current updated processing process, CiRepresents the completion time of each ant;
Figure FDA0002492734450000072
rho is a pheromone evaporation operator, and NC is the current iteration number.
CN201910877811.2A 2019-09-17 2019-09-17 Intelligent RGV processing system scheduling method and device Active CN110554673B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910877811.2A CN110554673B (en) 2019-09-17 2019-09-17 Intelligent RGV processing system scheduling method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910877811.2A CN110554673B (en) 2019-09-17 2019-09-17 Intelligent RGV processing system scheduling method and device

Publications (2)

Publication Number Publication Date
CN110554673A CN110554673A (en) 2019-12-10
CN110554673B true CN110554673B (en) 2020-07-03

Family

ID=68740574

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910877811.2A Active CN110554673B (en) 2019-09-17 2019-09-17 Intelligent RGV processing system scheduling method and device

Country Status (1)

Country Link
CN (1) CN110554673B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111136495B (en) * 2019-12-25 2021-02-26 广东宏远新科自动化技术开发有限公司 Intelligent unmanned CNC (computer numerical control) machining method and system for multiple varieties of small-batch machined parts
CN111882144B (en) * 2020-06-10 2022-03-11 中国人民解放军海军航空大学 Airplane guarantee procedure parallel scheduling method considering resource transfer
CN112987669B (en) * 2021-02-22 2022-04-08 广东鑫光智能***有限公司 RGV scheduling method based on serial scheduling algorithm
CN114063584B (en) * 2021-11-19 2024-04-26 江苏科技大学 Scheduling control method, device and system for integrated processing of ship weight closing parts

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103112680A (en) * 2013-01-31 2013-05-22 西安科技大学 Stereo logistics system access cargo path optimization control system and method
CN103345657A (en) * 2013-04-02 2013-10-09 江苏大学 Task scheduling method based on heredity and ant colony in cloud computing environment
CN103439885A (en) * 2013-07-26 2013-12-11 同济大学 Semiconductor production line optimized dispatching device
CN103489042A (en) * 2013-09-17 2014-01-01 中国科学院深圳先进技术研究院 Method for optimizing disaster emergency decision system path
KR20160104909A (en) * 2015-02-27 2016-09-06 고려대학교 산학협력단 Scheduling method and system using ant algorithm for physical distribution delivery
CN106228265A (en) * 2016-07-18 2016-12-14 中山大学 Based on Modified particle swarm optimization always drag phase transport project dispatching algorithm
CN106503789A (en) * 2016-11-08 2017-03-15 西安电子科技大学宁波信息技术研究院 Loop-free shortest path searching method based on Di Jiesitela and minimax ant colony
CN107330561A (en) * 2017-07-05 2017-11-07 青岛大学附属医院 A kind of multiple target bank bridge berth scheduling optimization method based on ant group algorithm
CN108596455A (en) * 2018-04-11 2018-09-28 哈尔滨理工大学 A kind of automobile chassis blanking Job-Shop method based on material allocation
CN110345960A (en) * 2019-06-13 2019-10-18 福建工程学院 Route planning intelligent optimization method for avoiding traffic obstacles
CN110610304A (en) * 2019-08-28 2019-12-24 广西电网有限责任公司电力科学研究院 Cross-region scheduling method of inspection robot based on hybrid genetic-ant colony algorithm

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2457927B (en) * 2008-02-28 2013-02-13 Ultra Global Ltd Method and system for resolving deadlocks
US8296712B2 (en) * 2010-09-30 2012-10-23 Synopsys, Inc. Method and apparatus for improving the interconnection and multiplexing cost of circuit design from high level synthesis using ant colony optimization
CN105302153B (en) * 2015-10-19 2018-04-17 南京航空航天大学 The planing method for the task of beating is examined in the collaboration of isomery multiple no-manned plane
CN109974711A (en) * 2019-04-12 2019-07-05 重庆渝博创智能装备研究院有限公司 A kind of AGV multiple target point autonomous navigation method towards wisdom factory

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103112680A (en) * 2013-01-31 2013-05-22 西安科技大学 Stereo logistics system access cargo path optimization control system and method
CN103345657A (en) * 2013-04-02 2013-10-09 江苏大学 Task scheduling method based on heredity and ant colony in cloud computing environment
CN103439885A (en) * 2013-07-26 2013-12-11 同济大学 Semiconductor production line optimized dispatching device
CN103489042A (en) * 2013-09-17 2014-01-01 中国科学院深圳先进技术研究院 Method for optimizing disaster emergency decision system path
KR20160104909A (en) * 2015-02-27 2016-09-06 고려대학교 산학협력단 Scheduling method and system using ant algorithm for physical distribution delivery
CN106228265A (en) * 2016-07-18 2016-12-14 中山大学 Based on Modified particle swarm optimization always drag phase transport project dispatching algorithm
CN106503789A (en) * 2016-11-08 2017-03-15 西安电子科技大学宁波信息技术研究院 Loop-free shortest path searching method based on Di Jiesitela and minimax ant colony
CN107330561A (en) * 2017-07-05 2017-11-07 青岛大学附属医院 A kind of multiple target bank bridge berth scheduling optimization method based on ant group algorithm
CN108596455A (en) * 2018-04-11 2018-09-28 哈尔滨理工大学 A kind of automobile chassis blanking Job-Shop method based on material allocation
CN110345960A (en) * 2019-06-13 2019-10-18 福建工程学院 Route planning intelligent optimization method for avoiding traffic obstacles
CN110610304A (en) * 2019-08-28 2019-12-24 广西电网有限责任公司电力科学研究院 Cross-region scheduling method of inspection robot based on hybrid genetic-ant colony algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
求解Job-shop 调度问题的遗传蚁群算法;吴宇明 等;《计算机应用研究》;20100930;第27卷(第9期);第3247-3249页，第3282页 *

Also Published As

Publication number Publication date
CN110554673A (en) 2019-12-10

Similar Documents

Publication Publication Date Title
CN110554673B (en) Intelligent RGV processing system scheduling method and device
CN109784566B (en) Order sorting optimization method and device
CN107036618A (en) A kind of AGV paths planning methods based on shortest path depth optimization algorithm
CN111674795A (en) Task scheduling method of cross-layer and cross-roadway shuttle storage system
CN114022028B (en) Automatic mixed pipeline scheduling layout integrated optimization method
CN112561194A (en) Production and logistics integrated scheduling method and system for hybrid flow shop
CN114707294A (en) Multi-target scheduling method for job shop with limited transportation capacity constraint
CN108229853B (en) Intelligent factory dynamic cooperative scheduling method based on static scheduling prediction
CN116523165B (en) Collaborative optimization method for AMR path planning and production scheduling of flexible job shop
CN111126770A (en) Task scheduling method of cross-layer shuttle storage system
Dang et al. A heuristic approach to schedule mobile robots in flexible manufacturing environments
Chaudhry et al. Integrated scheduling of machines and automated guided vehicles (AGVs) in flexible job shop environment using genetic algorithms
CN111923066B (en) Process scheduling method and device in wafer cleaning equipment
CN114610015A (en) Dynamic scheduling method and device for flexible job shop with carrying robot
CN110750079A (en) Hybrid flow shop scheduling optimization method allowing process jump
CN113361813A (en) Optimized scheduling method for scheduling system of wafer equipment
CN115034143A (en) Multi-resource cooperative intelligent workshop equipment configuration optimization method
JPH07237095A (en) Continuous production method for various kinds and device therefor
Badakhshian et al. Performance optimization of simultaneous machine and automated guided vehicle scheduling using fuzzy logic controller based genetic algorithm
CN115983423A (en) Feeding and discharging scene scheduling optimization method considering double resource constraints
Kumar et al. Integrated scheduling of part, tool and automated guided vehicles in a flexible manufacturing system using modified genetic algorithm
Reddy et al. Simultaneous scheduling of machines and tools considering tool transfer times in multimachine FMS using CSA
CN110187673B (en) Nonlinear programming RGV dynamic scheduling method
Mareddy et al. Simultaneous Scheduling of Machines, Tool Transporter and Tools in a Multi Machine Flexible Manufacturing System Without Tool Delay Using Crow Search Algorithm.
Narapureddy et al. Optimal scheduling methodology for machines, tool transporter and tools in a multi-machine flexible manufacturing system without tool delay using flower pollination algorithm

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
GR01 Patent grant
GR01 Patent grant