CN105629873A - Hybrid task scheduling method suitable for numerical control system - Google Patents

Hybrid task scheduling method suitable for numerical control system Download PDF

Info

Publication number
CN105629873A
CN105629873A CN201410625941.4A CN201410625941A CN105629873A CN 105629873 A CN105629873 A CN 105629873A CN 201410625941 A CN201410625941 A CN 201410625941A CN 105629873 A CN105629873 A CN 105629873A
Authority
CN
China
Prior art keywords
time
real
tasks
scheduling
periodic tasks
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.)
Granted
Application number
CN201410625941.4A
Other languages
Chinese (zh)
Other versions
CN105629873B (en
Inventor
郭锐锋
刘娴
王鸿亮
邓昌义
张忆文
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.)
Shenyang Institute of Computing Technology of CAS
Original Assignee
Shenyang Institute of Computing Technology of CAS
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 Shenyang Institute of Computing Technology of CAS filed Critical Shenyang Institute of Computing Technology of CAS
Priority to CN201410625941.4A priority Critical patent/CN105629873B/en
Publication of CN105629873A publication Critical patent/CN105629873A/en
Application granted granted Critical
Publication of CN105629873B publication Critical patent/CN105629873B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Abstract

The present invention relates to a hybrid task scheduling method suitable for a numerical control system. The method comprises the steps of adopting a predistribution subalgorithm to reserve the processor time for the real-time periodic tasks in the numerical control system before the task scheduling starts, and by deferring the execution of the real-time periodic tasks, advancing the idle time in the periodic task scheduling as much as possible. During the task scheduling process, the reservation time of the periodic tasks is adjusted dynamically according to the execution situation of the periodic tasks, thereby providing the maximum available idle time to the aperiodic tasks in the numerical control system. Compared with a conventional hybrid task scheduling algorithm, and according to the method of the present invention, the shortest aperiodic task response time can be obtained with a smaller calculation and storage cost while guaranteeing the cut-off time limit requirements of the real-time periodic tasks.

Description

A kind of hybrid tasks scheduling method suitable in digital control system
Technical field
The present invention relates to Real Time CNC System field, specifically a kind of hybrid tasks scheduling method suitable in digital control system.
Background technology
Digital control system is responsible for controlling in real time the correct operation of Digit Control Machine Tool, and handled task is by the periodicity in triggered time and whether has requirement of real-time, can be divided into real-time periodic tasks, accidental task and aperiodic task. In order to ensure the crudy of digital control system, its hybrid task need to be carried out rational management, while ensureing real-time periodic tasks and the requirement of real-time of accidental task, shorten the response time of aperiodic task as far as possible.
Traditional backstage performs method does not have when ready real-time periodic tasks or accidental task to dispatch execution aperiodic task in systems, the method is easily achieved, but when system has higher periodic duty load, aperiodic task can the processor time very few, its response time is unpredictable.
" Schedulingaperiodictasksindynamicprioritysystems " (publication " JournalofReal-TimeSystems " 1996 that Spuri et al. delivers, 10 (2): 179-210) EDLServer algorithm is proposed, by accurately calculating distribution and the size of the free time in real-time period task scheduling, and free time is shifted to an earlier date as far as possible, make aperiodic task meet with a response as early as possible. EDLServer algorithm can provide the shortest aperiodic task response time, but it is required for when each aperiodic task arrives calculating current free time, under worst case, the computation complexity of algorithm is O (N n), (wherein, the number of real-time periodic tasks in n expression system, N represents total operation number of this n periodic duty), its computing cost is excessive and is unsuitable for practical application.
It is coated with firm et al. " the soft aperiodic task dispatching algorithm of optimum based on dynamic priority strategy " (publication " Journal of Computer Research and Development 2004 delivered, 41 (11): 2026-2034. " ISA algorithm) is proposed, determine the relevant parameters such as the execution of real-time periodic tasks is interval and the free time is interval off-line, and in task scheduling process synchronized update relevant parameter, when aperiodic task arrives, according to current parameter value calculate aperiodic task can maximum idle time. This algorithm needs storage and safeguards quantity of parameters, and it calculates and storage overhead is bigger.
Summary of the invention
For hybrid tasks scheduling algorithm in prior art in the weak point existed in improving algorithmic dispatching performance and reducing algorithmic dispatching expense, the technical problem to be solved in the present invention is to provide a kind of hybrid tasks scheduling method suitable in digital control system.
For solving above-mentioned technical problem, the technical solution used in the present invention is: a kind of hybrid tasks scheduling method suitable in digital control system, comprises the following steps:
Step 1) set up computation model for the real-time periodic tasks in digital control system and aperiodic task;
Step 2) before task scheduling starts, it is the real-time periodic tasks reservation process device time by the predistribution algorithm of off-line, it is determined that the notice timetable of real-time periodic tasks;
Step 3) after task scheduling starts, according to the notice timetable of real-time periodic tasks, the real-time periodic tasks in digital control system and aperiodic task are carried out mixed scheduling.
The computation model of described real-time periodic tasks is: real-time periodic tasks set ��p={ ��1,��2,...,��nRepresent, each real-time periodic tasks ��iWith four-tuple < ��i,Ti,Di,Ci> represents, wherein, 1��i��n, n is the quantity of real-time periodic tasks, ��iFor ��iPhase place, TiFor ��iCycle, DiFor ��iRelatively end the time limit, CiFor ��iThe execution time; I-th real-time periodic tasks ��iJth subjob be expressed as ��ij��
The computation model of described aperiodic task is: the aperiodic task arrived in digital control system constitutes set ��ap={ ��1,��2... }, each aperiodic task ��kWith two tuple < rk,ek> represents, wherein, k is natural number, rkRepresent ��kThe time of advent, ekRepresent ��kThe execution time.
The predistribution algorithm of described off-line adopts the initial time of the reservation process device time of every subjob of the reverse EDF algorithm each real-time periodic tasks of calculating.
The notice timetable �� of described real-time periodic tasks is gathered �� by real-time periodic taskspIn reservation process device time of each real-time period operation arrange sequentially and form.
After described task scheduling starts, according to the notice timetable of real-time periodic tasks, the real-time periodic tasks in digital control system and aperiodic task are carried out mixed scheduling, comprise the following steps:
To notify that timetable �� is for input, starts real-time periodic tasks �� from the moment 0pCarry out EDF scheduling;
At scheduling instance t, if without pending aperiodic task, then scheduling performs the job sequence of real-time periodic tasks, the notice timetable �� according to the implementation status synchronized update real-time periodic tasks of real-time period operation;
If there being pending aperiodic task to arrive, this aperiodic task is inserted into aperiodic task queue, as moment t=��-head, if �� is-head=amn, then operation ��mnSeize other periodic duties and aperiodic task and perform until operation ��mnComplete, update the notice timetable �� of real-time periodic tasks; As moment t < ��-head, being then engaged in from the first to be appointed to an office of aperiodic task queue, order performs the aperiodic task in aperiodic task queue, until aperiodic task queue is empty or arrival notice timetable header element time ��-head; If both without pending aperiodic task, also endlessly ready real-time period operation, then processor is idle, and arranging moment t is next scheduling instance;
Repeat previous step until the superperiod terminates; The described superperiod is the least common multiple in the cycle of all real-time periodic tasks.
The notice timetable of described renewal real-time periodic tasks comprises the following steps: as real-time period operation ��ijAt moment t1Perform x unit of time, then move operation �� afterwardsijThe initial time a of reservation process device timeij, make ��ijTime decreased x unit of time of reservation process device, adjust all the other affected nodes simultaneously; Work as ��ijWhen completing, from notice timetable ��, delete aijNode.
There is pending aperiodic task to arrive if described, this aperiodic task is inserted into aperiodic task queue by first come first served basis.
The present invention while the cut-off time limit ensureing real-time periodic tasks requires, can obtain the shortest aperiodic task response time with less calculating and storage overhead.
Accompanying drawing explanation
Fig. 1 is task-set �� in embodimentp1={ ��1,��2}={ < 0,5,2 >, < 0,9,4 > } predistribution result under backwards-EDF algorithm;
Fig. 2 is task-set �� in embodimentp1={ ��1,��2}={ < 0,5,2 >, < 0,9,4 > } and ��ap1={ ��1,��2}={ < 11,3 >, < 27,4 > } mixed scheduling result under the inventive method;
Fig. 3 is that the performance of mixed scheduling algorithm is with UpThe situation of change;
Wherein, Fig. 3 (a) for operation aperiodic average response time with UpChange situation, Fig. 3 (b) for operation aperiodic completion rate with UpThe situation of change.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is described in further detail.
Consider the mixed scheduling of the scheduling real-time periodic task in digital control system and aperiodicity task. By all real-time periodic tasks set �� in systemp={ ��1,��2,...,��nRepresent, n �� N+Number for real-time periodic tasks. Each real-time periodic tasks ��i(1��i��n) uses four-tuple < ��i,Ti,Di,Ci> represents, wherein ��iFor ��iPhase place, TiFor ��iCycle, DiFor ��iRelatively end the time limit, CiFor ��iThe execution time. The execution each time of scheduling real-time periodic task is called its operation, so not being same concept. I is used for identifying task, and j is used for identifying which time of this task to be performed, namely which operation. ��iJth time perform to be designated as operation ��ij, ��ijThe r time of adventij=(j-1) Ti, end time limit dij=(j-1) Ti+Di, ��i=ri1����pIn the least common multiple of all duty cycles be designated asWillIt is called ��pSuperperiod. This model assumption real-time periodic tasks ��iMeet ��i=0, Di=Ti, and separate between each task. In system, the processor overall utilization of all real-time periodic tasks is designated asBy the aperiodic task �� in systemk(k is natural number) is with two tuple < rk,ek> represents, wherein rkRepresent ��kThe time of advent, ekRepresent ��kThe execution time, the attribute of aperiodic task is just known when task arrives. The aperiodic task arrived in system constitutes set ��ap={ ��1,��2... }, the Mean Time Between Replacement that aperiodic task arrives is designated as Tap, then aperiodic task load
The present invention is a kind of based on preallocated idle algorithm (Preallocation-basedSlack-StealingAlgorithm, PSSA) of diverting, abbreviation PSSA algorithm. Real-time periodic tasks in digital control system and aperiodic task, based on the computation model of above-mentioned foundation, are carried out mixed scheduling by PSSA algorithm, and its processing procedure comprises the hybrid tasks scheduling two parts during predistribution of processor time and the operation of off-line.
In order to ensure that the execution of aperiodic task does not affect the schedulability of real-time periodic tasks, before task scheduling starts, adopt reverse EDF (backwards-EDF) algorithm (having corresponding discussion in " Schedulingaperiodictasksindynamicprioritysystems " that Spuri that this algorithm is mentioned in the background et al. delivers),Interior from the momentStart to the moment 0, to real-time periodic tasks collection ��pCarry out back scheduling. The processor time that each real-time period operation in back scheduling process is taken as its reserved execution time in task-set forward scheduling process.
Define 1. operation ��ijNotice time aij: at ��pBackwards-EDF scheduling result in, operation ��ijDeadline be called ��ijThe notice time, be designated as aij. In task-set forward scheduling process, at aijMoment, if ��ijNot yet complete, then ��ijSeize other tasks carryings.
Define 2. ��pNotice timetable ��: by ��pThe notice time dispatching each real-time period operation obtained at backwards-EDF arranges vector sequentially, and the header element constituting notice timetable ��, �� is designated as ��-head.
In one embodiment, it is considered to two real-time periodic tasks �� in digital control systemp1={ ��1,��2}={ < 0,5,2 >, < 0,9,4 > },To �� in first superperiod [0,45]p1Carrying out backwards-EDF scheduling, acquired results is as shown in Figure 1. Periodic duty ��1Notice time series be 2,8,13,18,23,28,33,37,43}, periodic duty ��2Notice time series be 4,12,21,30,39}, then ��pNotice timetable ��={ 2,4,8,12,13,18,21,23,28,30,33,37,39,43}.
After task scheduling starts, the notice timetable �� that PSSA determines with backwards-EDF off-line is input, rightInterior real-time periodic tasks and aperiodic task carry out mixed scheduling. In real-time periodic tasks execution process, the time that notifies accordingly in synchronized update ��, to retain the available free time of aperiodic task. PSSA existsInterior mixed scheduling process is as follows:
Step 1. is to notify that timetable �� is for input, starts �� from the moment 0pCarry out EDF scheduling;
Step 2. is at scheduling instance t, if without pending aperiodic task, then scheduling performs the job sequence of real-time periodic tasks, the implementation status according to real-time period operation, synchronized update notice timetable ��. If real-time period operation ��ijAt moment t1Perform x unit of time, then move a afterwardsijMake ��ijTime decreased x unit of time of reservation process device, adjust all the other affected nodes simultaneously. Work as ��ijWhen completing, from ��, delete aijNode. If there being pending aperiodic task, then proceed to step 3;
Step 3. sets aperiodic task ��jArrive at moment t, by ��jAperiodic task queue is inserted by first come first served basis. If moment t=��-head, if �� is-head=amn, then ��mnSeize other periodic duties and aperiodic task performs until it completes, update ��; If t is < ��-head, then being engaged in from the first to be appointed to an office of aperiodic task queue, order performs, till aperiodic task queue is empty or arrival notice time ��-head.
Step 4. is at scheduling instance t, if both without pending aperiodic task, also endlessly ready real-time period operation, then processor is idle, and arranging t is next scheduling instance.
Step 5. repeated execution of steps 2,3,4, terminated to the superperiod.
At ��p1First superperiod [0,45] in have two aperiodic task ��ap1={ ��1,��2}={ < 11,3 >, < 27,4 > } arrive, the backwards-EDF predistribution result according to Fig. 1, adopt PSSA to ��p1With ��ap1Carrying out mixed scheduling, scheduling result is as shown in Figure 2.
It is respectively adopted PSSA, EDLServer algorithm and backstage performs algorithm and with aperiodic task, the real-time periodic tasks in digital control system is simulated scheduling. In blue sky numerical control G301 system, every 6ms performs a locus interpolation, the result that each interpolation calculates to deliver to the command register of servosystem at twice to control the displacement of each coordinate axes, and namely each interpolation cycle domestic demand completes 2 these position controls, and the cycle of position control is 3ms; Every 15ms performs a PLC and processes task, and the PLC control program of input is compiled, and completes switching value logical operations, generates digital output modul information. Considering that the execution time needed for the different implementations of real-time periodic tasks is different, experiment sets real-time periodic tasks load UpValue in [0.4,0.9]. Considering the uncertainty of aperiodic task and infrequently property feature in digital control system, in order to the scheduling result of algorithms of different is contrasted, emulation experiment is according to exponential E (1/Tap) determine the interval time of advent of aperiodic task at random, the Mean Time Between Replacement that aperiodic task arrives is TapTake 75ms, simultaneously the execution time of stochastic generation aperiodic task, and meet Uap=20%. The completion rate of the aperiodic task that experiment statistics system has arrived in the mixed scheduling in 1s duration and the average response time of completed aperiodic taskAs the performance indications weighing hybrid tasks scheduling algorithm. Experimental result is as shown in Figure 3.
Being can be seen that by Fig. 3 (a), backstage performs the aperiodic task average response time under algorithmic dispatching with periodic duty load UpIncrease and accelerate gradually to increase; And PSSA and EDLServer algorithm is at UpMeet Up+UapDuring < 1, it is possible to willMaintain a lower value, work as Up+UapWhen >=1,Increase gradually, but its growth rate is slower than backstage and performs algorithm. This is because backstage performs algorithm just performs aperiodic task after ready real-time periodic tasks completes, work as UpDuring increase, the waiting time of aperiodic task extends, and its response time increases. And PSSA and EDLServer algorithm is as far as possible by the free time in real-time period task scheduling in advance, for the aperiodic task arrived, to reduce the waiting time of aperiodic task. Work as UpDuring increase, free time total amount in one superperiod reduces, press first come first served basis due to aperiodic task to perform, the aperiodic task relatively early arrived can use existing free time, obtain shorter response time, the aperiodic task of follow-up arrival can only wait the free time in the next superperiod, thus the situation of change occurred in Fig. 3 (b), works as Up+UapWhen >=1, the aperiodic task completion rate in the superperiod constantly declines. By 3 (a) it can also be seen that work as UpTime less, the scheduling performance of PSSA is close with EDLSever algorithm, along with UpIncreasing, PSSA is obvious gradually compared with the advantage of EDLSever algorithm, this is because relatively EDLServer algorithm, the extra scheduling overhead that PSSA introduces is relatively low, and the impact of hybrid tasks scheduling is inconspicuous.

Claims (8)

1. the hybrid tasks scheduling method being applicable to digital control system, it is characterised in that comprise the following steps:
Step 1) set up computation model for the real-time periodic tasks in digital control system and aperiodic task;
Step 2) before task scheduling starts, it is the real-time periodic tasks reservation process device time by the predistribution algorithm of off-line, it is determined that the notice timetable of real-time periodic tasks;
Step 3) after task scheduling starts, according to the notice timetable of real-time periodic tasks, the real-time periodic tasks in digital control system and aperiodic task are carried out mixed scheduling.
2. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 1, it is characterised in that the computation model of described real-time periodic tasks is: real-time periodic tasks set ��p={ ��1,��2,...,��nRepresent, each real-time periodic tasks ��iWith four-tuple < ��i,Ti,Di,Ci> represents, wherein, 1��i��n, n is the quantity of real-time periodic tasks, ��iFor ��iPhase place, TiFor ��iCycle, DiFor ��iRelatively end the time limit, CiFor ��iThe execution time; I-th real-time periodic tasks ��iJth subjob be expressed as ��ij��
3. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 1, it is characterised in that the computation model of described aperiodic task is: the aperiodic task arrived in digital control system constitutes set ��ap={ ��1,��2... }, each aperiodic task ��kWith two tuple < rk,ek> represents, wherein, k is natural number, rkRepresent ��kThe time of advent, ekRepresent ��kThe execution time.
4. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 1, it is characterized in that, the predistribution algorithm of described off-line adopts the initial time of the reservation process device time of every subjob of the reverse EDF algorithm each real-time periodic tasks of calculating.
5. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 1, it is characterised in that the notice timetable �� of described real-time periodic tasks is gathered �� by real-time periodic taskspIn reservation process device time of each real-time period operation arrange sequentially and form.
6. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 1, it is characterized in that, after described task scheduling starts, real-time periodic tasks in digital control system and aperiodic task are carried out mixed scheduling by the notice timetable according to real-time periodic tasks, comprise the following steps:
To notify that timetable �� is for input, starts real-time periodic tasks �� from the moment 0pCarry out EDF scheduling;
At scheduling instance t, if without pending aperiodic task, then scheduling performs the job sequence of real-time periodic tasks, the notice timetable �� according to the implementation status synchronized update real-time periodic tasks of real-time period operation;
If there being pending aperiodic task to arrive, this aperiodic task is inserted into aperiodic task queue, as moment t=��-head, if �� is-head=amn, then operation ��mnSeize other periodic duties and aperiodic task and perform until operation ��mnComplete, update the notice timetable �� of real-time periodic tasks; As moment t < ��-head, being then engaged in from the first to be appointed to an office of aperiodic task queue, order performs the aperiodic task in aperiodic task queue, until aperiodic task queue is empty or arrival notice timetable header element time ��-head; If both without pending aperiodic task, also endlessly ready real-time period operation, then processor is idle, and arranging moment t is next scheduling instance;
Repeat previous step until the superperiod terminates; The described superperiod is the least common multiple in the cycle of all real-time periodic tasks.
7. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 6, it is characterised in that the notice timetable of described renewal real-time periodic tasks comprises the following steps: as real-time period operation ��ijAt moment t1Perform x unit of time, then move operation �� afterwardsijThe initial time a of reservation process device timeij, make ��ijTime decreased x unit of time of reservation process device, adjust all the other affected nodes simultaneously; Work as ��ijWhen completing, from notice timetable ��, delete aijNode.
8. a kind of hybrid tasks scheduling method suitable in digital control system according to claim 6, it is characterised in that have pending aperiodic task to arrive if described, this aperiodic task is inserted into aperiodic task queue by first come first served basis.
CN201410625941.4A 2014-11-07 2014-11-07 A kind of hybrid tasks scheduling method suitable for digital control system Active CN105629873B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410625941.4A CN105629873B (en) 2014-11-07 2014-11-07 A kind of hybrid tasks scheduling method suitable for digital control system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410625941.4A CN105629873B (en) 2014-11-07 2014-11-07 A kind of hybrid tasks scheduling method suitable for digital control system

Publications (2)

Publication Number Publication Date
CN105629873A true CN105629873A (en) 2016-06-01
CN105629873B CN105629873B (en) 2018-08-24

Family

ID=56044937

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410625941.4A Active CN105629873B (en) 2014-11-07 2014-11-07 A kind of hybrid tasks scheduling method suitable for digital control system

Country Status (1)

Country Link
CN (1) CN105629873B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107800646A (en) * 2017-11-06 2018-03-13 国网山东省电力公司电力科学研究院 A kind of communication task adaptive optimization dispatching method for supporting dynamic flow to control
CN109298917A (en) * 2017-07-25 2019-02-01 沈阳高精数控智能技术股份有限公司 A kind of self-adapting dispatching method suitable for real-time system hybrid task

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006268754A (en) * 2005-03-25 2006-10-05 Toshiba Corp Scheduling possibility determination method, real-time system and program
CN103163870A (en) * 2013-04-07 2013-06-19 沈阳化工大学 Method for scheduling wireless networked control system with maximal tolerant time delay
CN103365710A (en) * 2012-03-31 2013-10-23 索尼公司 Real-time task scheduling device and method and computer system
CN103810043A (en) * 2012-11-09 2014-05-21 中国科学院沈阳计算技术研究所有限公司 Energy-saving scheduling method suitable for numerical control system periodic tasks
CN103810026A (en) * 2012-11-09 2014-05-21 中国科学院沈阳计算技术研究所有限公司 Mixing scheduling method suitable for real-time system periodic tasks

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006268754A (en) * 2005-03-25 2006-10-05 Toshiba Corp Scheduling possibility determination method, real-time system and program
CN103365710A (en) * 2012-03-31 2013-10-23 索尼公司 Real-time task scheduling device and method and computer system
CN103810043A (en) * 2012-11-09 2014-05-21 中国科学院沈阳计算技术研究所有限公司 Energy-saving scheduling method suitable for numerical control system periodic tasks
CN103810026A (en) * 2012-11-09 2014-05-21 中国科学院沈阳计算技术研究所有限公司 Mixing scheduling method suitable for real-time system periodic tasks
CN103163870A (en) * 2013-04-07 2013-06-19 沈阳化工大学 Method for scheduling wireless networked control system with maximal tolerant time delay

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
赵公怡: "《单处理器环境下实时调度算法研究》", 《中国优秀硕士学位论文全文数据库(电子期刊)》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109298917A (en) * 2017-07-25 2019-02-01 沈阳高精数控智能技术股份有限公司 A kind of self-adapting dispatching method suitable for real-time system hybrid task
CN107800646A (en) * 2017-11-06 2018-03-13 国网山东省电力公司电力科学研究院 A kind of communication task adaptive optimization dispatching method for supporting dynamic flow to control
CN107800646B (en) * 2017-11-06 2019-10-15 国网山东省电力公司电力科学研究院 A kind of communication task adaptive optimization dispatching method for supporting dynamic flow to control

Also Published As

Publication number Publication date
CN105629873B (en) 2018-08-24

Similar Documents

Publication Publication Date Title
CN107831745B (en) A kind of slotting single action state method for optimizing scheduling of flexible job shop
CN104536835B (en) Self-adapting task scheduling method in a kind of Hot Spare real-time control system
CN101339521B (en) Tasks priority dynamic dispatching algorithm
CN103433926B (en) A kind of Multi-robot coordination control device and method thereof
CN104731657B (en) A kind of resource regulating method and system
CN101923487A (en) Comprehensive embedded type real-time period task scheduling method
CN105629873A (en) Hybrid task scheduling method suitable for numerical control system
WO2017071230A1 (en) Method for short-term optimal scheduling of multi-agent hydropower station group
CN106445070B (en) A kind of resource-constrained accidental task energy optimization dispatching method of hard real-time system
CN106802825B (en) A kind of dynamic task scheduling method and system based on real-time system
CN103455375A (en) Load-monitoring-based hybrid scheduling method under Hadoop cloud platform
CN105516024B (en) A kind of task flux monitoring method and system based on queue
CN105187327A (en) Distributed message queue middleware
CN107329817A (en) A kind of stand-by system mixing divides reliability and perceives energy consumption optimization method
CN105974891B (en) A kind of mold production process self-adaptation control method based on dynamic billboard
Ma et al. A task-grouped approach for the multi-robot task allocation of warehouse system
CN110161983A (en) Control device, control system, control method and storage medium
CN105335130A (en) Processor and method for processing task
CN104270418A (en) Cloud agent appointment allocation method orientated to user demanded Deadlines
CN102023895A (en) Singlechip programming method with priority task switching
CN105302640A (en) Time slice and management and control method applied to multiple time slices
CN106843121A (en) The method of the independent control period allocated of each axle in PLC alignment systems
CN106445659B (en) A kind of spacecraft periodically blendes together Random Task dispatching method
CN106681692A (en) Control device, integrated circuit and management method of task stack
CN105159752A (en) Real-time task and resource scheduling method with function of machine startup time perception in virtualized clouds

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
GR01 Patent grant