• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Real-time process control method, program and apparatus for: maintaining (210) constraints for resource allocation in performing requested operations for failure recovery and service needs; maintaining (220), for a constrained resource set comprising a plurality of resources, for each of the plurality resources, resource capability data; generating (230) with a genetic algorithm an optimised selection of individual solutions for the use of each one of the plurality of resources in each of different resource requirement state combinations constrained by the resource capability data, further including automatically and repeatedly: receiving (240) a new operation request; defining (250) current resource requirement state based on pending operation requests; and selecting (260) next one of the pending operation requests and a resource to perform the requested operation from the resource set according to the optimised selection of individual solutions.
    公开号:FI20205929A1
    申请号:FI20205929
    申请日:2020-09-25
    公开日:2021-12-15
    发明作者:Ville Ruutu；Jussi Ruutu
    申请人:Elisa Oyj；
    IPC主号:
    专利说明:

    REAL-TIME PROCESS CONTROL WITH CONSTRAINED RESOURCES
    TECHNICAL FIELD The present disclosure generally relates to real-time process control with constrained resources. The disclosure relates particularly, though not exclusively, to process control of parallel failure recovery and service action with given resources.
    BACKGROUND This section illustrates useful background information without admission of any technique described herein representative of the state of the art.
    Various industries manufacture articles and produce different services using equipment in which different failure recovery and service needs may occur at a number of places and overlapping times. Such needs often compete of shared resources, such as pieces of equipment and service robots or provisioning agents as the case may be, for example, in mobile networks, for example.
    There are numerous methods for optimising various processes based on available resources and given constraints. However, such optimisation methods may become overly complex and difficult to understand or require unfeasible amount of time and resources while there is an urgency to decide on best action in real time.
    SUMMARY o 20 It is an object of the present invention to improve real-time process control with
    QA a constrained resources such that adverse effects to guality and production could be 3 avoided or reduced. It is another object of the present invention to mitigate O drawbacks present in prior art. It is yet another object of the present invention to at E least provide a new technical alternative to existing technology.
    | 25 The appended claims define the scope of protection. Any examples and technical S descriptions of apparatuses, products and/or methods in the description and/or drawings not covered by the claims are presented not as embodiments of the invention but as background art or examples useful for understanding the invention.
    According to a first example aspect there is provided a method for real-time process control, the method comprising: maintaining constraints for resource allocation in performing requested operations for failure recovery and service needs; maintaining, for a constrained resource set comprising a plurality of resources, for each of the plurality resources, resource capability data; generating with a genetic algorithm an optimised selection of individual solutions for the use of each one of the plurality of resources in each of different resource requirement state combinations constrained by the resource capability data; and the method further comprising automatically and repeatedly: receiving a new operation request; defining current resource requirement state based on pending operation requests; and selecting next one of the pending operation requests and a resource to perform the requested operation from the resource set according to the optimised selection of individual solutions. The method may comprise computing with an alternative optimising method competing selecting of the next one of the pending operation requests and the resource to perform the requested operation; comparing a given performance derivative value; and performing corrective action on meeting a given threshold. The corrective action may comprise repeating the generating of the optimised selection of the individual solutions using new training data. The new training data N may comprise recent operation reguests.
    N 2 25 The threshold may be that the performance derivative value of the alternative 0 method in comparison to that of the genetic algorithm based optimised selection of z individual solutions exceeds a given value. The given value may be greater than 50 e %; 80 %; 100 %, 120 %; and 150 %. 3 The recovery and service needs may comprise any one or more of: a need for S 30 updating software or firmware of a device; a need for replacing a power supply of a device; a need for replacing a port of a device; a need for replacing a display of a device; a need for upgrading software or firmware of a device; a need for upgrading a power supply of a device; a need for upgrading a port of a device; a need for upgrading a display of a device; a need for removing a blockage of a production line; a need for supplementing consumables of a device; a need for replacing a broken part; a need for adjusting radio resource distribution between different cells or beams; a need for steering hand-overs of cellular subscribers in a cellular network; and a need for rebooting a device.
    The plurality of resources may comprise one or more service agents such as, for example, autonomous robots; remote diagnosing agents; remote programming agents; remote adjusting agents; autonomous vehicles; diagnostic stations; computer servers; virtual computers; artificial intelligence functions; or data buses.
    The real-time process control may be performed in a production line.
    The real-time process control may be performed by a mobile operator.
    The real-time process control may be performed in a telecommunication network.
    The capability data may indicate a capacity of a resource to provide required operations of a particular failure recovery and service need.
    A cost function may be defined for balancing different recovery and service needs.
    The genetic algorithm may be developed in a process comprising any one or more of the following: a) defining a group of actions for different recovery and service resources; b) defining possible states of different recovery and service needs for each of a plurality of different groups; S c) defining a set of algorithm individuals for different recovery and service & resources and assigning each individual with given genetic encoding, LO 25 wherein the different recovery and service resources may have different 2 capabilities; and initialising genetic encoding of at least some of the * programmed individuals; wherein the initialising may be performed by 3 randomising; N d) defining an evaluation measure such as a cost function of prevailing states N 30 of each of the different groups; e) generating a test status in which the states are simulated or formed using historic data and/or randomised data for each of the groups and for a queue of pending recovery and service resource requests; f) deciding by each individual a new task from the queue based on the genetic encoding of the individual in question; the prevailing states and the queue; computing a new evaluation measure after each decision; and storing the computed new evaluation measure for each individual; g) updating the test status for each of the individuals; wherein the updating comprises updating the queue based on historic data or simulation and taking into account the decisions made by each individual, h) for a number of times and repeating steps, repeating steps f) and g) i) evaluating performance of each individual based on the stored evaluation measure and selecting a suitable subset of the individuals; j) creating new individuals at least partly based on the genetic encoding of the subset of the individuals and optionally included mutations; k) repeating evaluation for a formed new generation of individuals by repeating steps e) to j) unless or until one or more termination criteria are achieved, such as the evaluation measure meeting a given threshold and/or reaching a given number of generations.
    Advantageously, a population of programmed individuals may be used in the assigning such that set pre-determined and verifiable factors define the assigning and allow verification of the assigning. According to a second example aspect there is provided a computer program comprising computer executable program code which when executed by at least N one processor causes an apparatus at least to perform:
    N 2 25 According to a third example aspect there is provided a computer program product 0 comprising a non-transitory computer readable medium having the computer E program of the third example aspect stored thereon. | According to a fourth example aspect there is provided an apparatus comprising S means for performing the method of the first example aspect. The means for performing the first example aspect may comprise the computer program of the second example aspect. The means for performing the first example aspect may comprise a memory for storing the computer program and at least one processor for executing the program. Any foregoing memory medium may comprise a digital data storage such as a data disc or diskette, optical storage, magnetic storage, holographic storage, opto- 5 magnetic storage, phase-change memory, resistive random access memory, magnetic random access memory, solid-electrolyte memory, ferroelectric random access memory, organic memory or polymer memory. The memory medium may be formed into a device without other substantial functions than storing memory or it may be formed as part of a device with other functions, including but not limited to a memory of a computer, a chip set, and a sub assembly of an electronic device. Different non-binding example aspects and embodiments have been illustrated in the foregoing. The embodiments in the foregoing are used merely to explain selected aspects or steps that may be utilised in different implementations. Some embodiments may be presented only with reference to certain example aspects. It should be appreciated that corresponding embodiments may apply to other example aspects as well.
    BRIEF DESCRIPTION OF THE FIGURES Some example embodiments will be described with reference to the accompanying figures, in which: Fig. 1 schematically shows a system of an example embodiment; Fig. 2 shows a block diagram of an apparatus of an example embodiment; and Fig. 3 shows a flow chart of an example embodiment.N
    N DETAILED DESCRIPTION In the following description, like reference signs denote like elements or steps.
    N = 25 Fig. 1 schematically shows a system 100 of an example embodiment. The system > 100 comprises three production lines 110, 120, 130; a plurality of resources & represented by a first resource R1 and a second resource R2; and an optimiser 140. ä The plurality of resources each have given capabilities for performing different operations for failure recovery and service needs. The resources can attend to one task at the time at each of the different production lines.
    In Fig. 1, different events are drawn representative of arising needs for failure recovery and service at the different production lines. The production lines of Fig. 1 generally represent different groups of actions combined here by their functional and positional context. Different groups of actions are represented by the different events e11 to e14 in a first group of the first production line 110; events e21 and e22 represent actions of a second group of the second production line 120; and respectively a third group of actions is illustrated at third production line 130 with events e31 and e32. The different events may differ by resource requirements such that only some of the resources may be capable of rendering required action to correct or service as required, or generally speaking address requests. In Fig. 1, this is illustrated by using event markers of two types, triangles and hexagons. Out of the drawn resources, some, here first resource R1, is capable of addressing all requests whereas others, here the second resource, only can address particular requests according to their capabilities. This is indicated by drawn dotted lines indicating which events could be addressed by which resources. In Fig. 1, the second resource R2 is assumed to be not capable of addressing requests corresponding to events e13 and e21 drawn as hexagons. Fig. 1 further shows an optimiser 140. The optimiser is configured to optimise the plurality of resources to independently select next tasks after completing previous ones so as to steer the user of the resources in an optimised manner. For example, different production lines may be exposed to different risks in result of downtime and the subseguent needs for produce of different production lines may also vary such N that some groups of actions should be weighed more than others. In case of N 25 subsequent needs and different states of the production lines, the weighs may be 7 different, though. Hence, a genetic algorithm is employed in this case for optimising - the resource consumption by programming the different resources to offer their E capacity as optimally as possible within the constraints of algorithms used and | random occurrence of unexpected events, for example.
    LO N 30 Fig. 2 shows a flow chart of a process of a first example aspect for real-time process N control, such as performing requested operations for failure recovery and service needs. The process comprises:
    210. maintaining constraints for resource allocation in performing requested operations for failure recovery and service needs;
    220. maintaining, for a constrained resource set comprising a plurality of resources, for each of the plurality resources, resource capability data; and
    230. generating with a genetic algorithm an optimised selection of individual solutions for the use of each one of the plurality of resources in each of different resource requirement state combinations constrained by the resource capability data. The process further comprises automatically and repeatedly performing the steps of:
    240. receiving a new operation request;
    250. defining current resource requirement state based on pending operation requests; and
    260. selecting next one of the pending operation requests and a resource to perform the requested operation from the resource set according to the optimised selection of individual solutions. In an embodiment, the method further comprises 270. computing with an alternative optimising method competing selecting of the next one of the pending operation requests and the resource to perform the requested operation; comparing a given performance derivative value; and performing corrective action on meeting a given threshold. In an embodiment, the corrective action comprises repeating the generating of the optimised selection of the individual solutions using new training data. The new training data may comprise recent operation requests.
    O O In an embodiment, the threshold is that in the comparison, the performance 2 25 derivative value of the alternative method exceeding a given proportion of that when 0 using the genetic algorithm based optimised selection of individual solutions. The E given proportion may be greater than 50 %; 80 %; 100 %, 120 %; and 150 %. | The recovery and service needs may greatly vary in different implementations so S that they may comprise, for example, any one or more of: a need for updating N 30 software or firmware of a device; a need for replacing a power supply of a device; a need for replacing a port of a device; a need for replacing a display of a device; a need for upgrading software or firmware of a device; a need for upgrading a power supply of a device; a need for upgrading a port of a device; a need for upgrading a display of a device; a need for removing a blockage of a production line; a need for supplementing consumables of a device; a need for replacing a broken part; a need for adjusting radio resource distribution between different cells or beams; a need for steering hand-overs of cellular subscribers in a cellular network; and a need for rebooting a device.
    In an embodiment, the plurality of resources comprise one or more service agents such as, for example, autonomous robots; remote diagnosing agents; remote programming agents; remote adjusting agents; autonomous vehicles; diagnostic stations; computer servers; virtual computers; artificial intelligence functions; or data buses.
    In an embodiment, the capability data indicates a capacity of a resource to provide required operations of a particular failure recovery and service need.
    In an embodiment, a cost function is defined for balancing different recovery and service needs.
    Fig. 3 shows a flow chart of a process of developing a genetic algorithm by the optimiser 140, for example, comprising.
    300. defining a group of actions for different recovery and service resources;
    305. defining possible states of different recovery and service needs for each of a plurality of different groups, for example, as whether being behind, at, or ahead of a target, which in case of three groups could infer 33 = 27 different states; Q 310. defining a set of algorithm individuals for different recovery and service N resources and assigning each individual with given genetic encoding such as genes, S 25 wherein the different recovery and service resources may have different capabilities; N and initialising genetic encoding of at least some of the programmed individuals; E: wherein the initialising may be performed by randomising; 2 315. defining an evaluation measure such as a cost function of prevailing states of 2 each of the different groups, such as a cost function f = a (S1-S1)+az2 (S2'- N 30 — S2)+a3:(S3'-S3), wherein a4, az, and as are relative weighs and S1, Sz, and Ss indicate present states respective groups 1 to 3, and S1', Sy’, and Sz’ indicate target states of these groups 1 to 3;
    320. generating a test status in which the states are simulated or formed using historic data and/or randomised data for each of the groups and for a queue of pending recovery and service resource requests;
    325. deciding by each individual a new task from the queue based on the genetic encoding of the individual in question; the prevailing states and the queue; computing a new evaluation measure after each decision; and storing the computed new evaluation measure for each individual;
    330. updating the test status for each of the individuals; wherein the updating comprises updating the queue based on historic data or simulation and taking into account the decisions made by each individual,
    335. for a number of times and repeating steps, repeating the deciding and updating steps;
    340. evaluating performance of each individual based on the stored evaluation measure and selecting a suitable subset of the individuals, for example, by computing a mean or average of the evaluation measures of each individual or by using current values of the evaluation measures of each individual, the subset being, for example, wherein the subset comprises two or more individuals;
    345. creating new individuals at least partly based on the genetic encoding of the subset of the individuals and optionally based on mutations included into the creating of new individuals;
    350. repeating evaluation for a formed new generation of individuals by repeating steps 320 to 345 unless or until one or more termination criteria are achieved, such S as the evaluation measure meeting a given threshold and/or reaching a given O 25 number of generations. 3 A production line example was described with reference to Fig. 1. It is appreciated N that the processes of Figs. 2 and 3 are also usable in other contexts, such as E maintenance of telecommunication networks. | Fig. 4 shows a block diagram of an apparatus 400 of an example embodiment. The S 30 apparatus 400 comprises a communication interface 410; a processor 420; a user S interface 430; and a memory 440. The apparatus 400 may be usable for at least partially implementing the optimiser 140; and/or the first or second resource R1, R2.
    The communication interface 410 comprises in an embodiment a wired and/or wireless communication circuitry, such as Ethernet; Wireless LAN; Bluetooth; GSM; CDMA; WCDMA; LTE; and/or 5G circuitry. The communication interface can be integrated in the apparatus 400 or provided as a part of an adapter, card or the like, that is attachable to the apparatus 400. The communication interface 410 may support one or more different communication technologies. The apparatus 400 may also or alternatively comprise more than one of the communication interfaces 410. The processor 420 may be a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a graphics processing unit, an application specific integrated circuit (ASIC), a field programmable gate array, a microcontroller or a combination of such elements. The user interface may comprise a circuitry for receiving input from a user of the apparatus 400, e.g., via a keyboard, graphical user interface shown on the display of the apparatus 400, speech recognition circuitry, or an accessory device, such as a headset, and for providing output to the user via, e.g., a graphical user interface or a loudspeaker. The memory 440 comprises a work memory 442 and a persistent memory 444 configured to store computer program code 446 and data 448. The memory 440 may comprise any one or more of: a read-only memory (ROM); a programmable read-only memory (PROM); an erasable programmable read-only memory (EPROM); arandom-access memory (RAM); a flash memory; a data disk; an optical storage; a magnetic storage; a smart card; a solid state drive (SSD); or the like. The apparatus 400 may comprise a plurality of the memories 440. The memory 440 may N be constructed as a part of the apparatus 400 or as an attachment to be inserted 3 25 into a slot, port, or the like of the apparatus 400 by a user or by another person or LO by a robot. The memory 440 may serve the sole purpose of storing data or be - constructed as a part of an apparatus 400 serving other purposes, such as E processing data.
    Q 3 A skilled person appreciates that in addition to the elements shown in Figure 4, the ä 30 apparatus 400 may comprise other elements, such as microphones, displays, as well as additional circuitry such as input/output (I/O) circuitry, memory chips, application-specific integrated circuits (ASIC), processing circuitry for specific purposes such as source coding/decoding circuitry, channel coding/decoding circuitry, ciphering/deciphering circuitry, and the like. Additionally, the apparatus 400 may comprise a disposable or rechargeable battery (not shown) for powering the apparatus 400 when external power if external power supply is not available.
    Atechnical effect of at least one embodiment is that decisions on resource use is distributed to different resources using data derived by genetic algorithm. Hence, the resources may operate without continuous connectivity. Another technical effect is that resource use decisions can be made with low computational cost without need to evaluate different scenarios at the time when a new decision should be made. Yet another technical effect is that resource use decisions can be made in real-time with low computational complexity while taking into account an extensive evaluation of different alternative scenarios in the presence of numerous resources and tasks.
    Various embodiments have been presented. It should be appreciated that in this document, words comprise, include and contain are each used as open-ended expressions with no intended exclusivity.
    The foregoing description has provided by way of non-limiting examples of particular implementations and embodiments a full and informative description of the best mode presently contemplated by the inventors for carrying out the invention. It is however clear to a person skilled in the art that the invention is not restricted to details of the embodiments presented in the foregoing, but that it can be implemented in other embodiments using equivalent means or in different o combinations of embodiments without deviating from the characteristics of the O invention.
    3 25 Furthermore, some of the features of the afore-disclosed example embodiments a may be used to advantage without the corresponding use of other features. As such, E the foregoing description shall be considered as merely illustrative of the principles R of the present invention, and not in limitation thereof. Hence, the scope of the D invention is only restricted by the appended patent claims.
    S 30
    权利要求:
    Claims (15)
    [1] 1. A method for real-time process control, the method comprising: maintaining constraints for resource allocation in performing requested operations for failure recovery and service needs; maintaining, for a constrained resource set comprising a plurality of resources, for each of the plurality resources, resource capability data; generating with a genetic algorithm an optimised selection of individual solutions for the use of each one of the plurality of resources in each of different resource requirement state combinations constrained by the resource capability data; and the method further comprising automatically and repeatedly: receiving a new operation request; defining current resource requirement state based on pending operation requests; and selecting next one of the pending operation requests and a resource to perform the requested operation from the resource set according to the optimised selection of individual solutions.
    [2] 2. The method of claim 1, further comprising repeating the generating of the optimised selection of the individual solutions using new training data.
    [3] 3 The method of claim 2, wherein the new training data comprises recent operation requests.
    [4] 4. The method of any one of preceding claims, wherein the method is performed N by a mobile operator. >
    [5] 5. The method of any one of preceding claims, further comprising computing with 2 25 an alternative optimising method competing selecting of the next one of the pending I operation requests and the resource to perform the requested operation; comparing * a given performance derivative value; and performing corrective action on meeting 3 a given threshold. ä
    [6] 6. The method of claim 5, wherein the threshold is that the performance derivative value of the alternative method in comparison to that of the genetic algorithm based optimised selection of individual solutions exceeds a given value.
    [7] 7. The method of any one of preceding claims, wherein the recovery and service needs comprise any one or more of. a need for updating software or firmware of a device; a need for replacing a power supply of a device; a need for replacing a port of a device; a need for replacing a display of a device; a need for upgrading software or firmware of a device; a need for upgrading a power supply of a device; a need for upgrading a port of a device; a need for upgrading a display of a device; a need for removing a blockage of a production line; a need for supplementing consumables of a device; a need for replacing a broken part; a need for adjusting radio resource distribution between different cells or beams; a need for steering hand-overs of cellular subscribers in a cellular network; and a need for rebooting a device; a need for changing network configuration; a need for adjusting network capacity; a need for changing one or more network parameters.
    [8] 8. The method of any one of preceding claims, wherein the plurality of resources comprise one or more service agents.
    [9] 9 The method of claim 8, wherein the service agents comprise at least one of the following: an autonomous robot; a remote diagnosing agent; a remote programming agent; a remote adjusting agent; an autonomous vehicle; a diagnostic station; a computer server; a virtual computer; an artificial intelligence function; or a data bus.
    [10] 10. The method of any one of preceding claims, wherein the resource capability data indicates a capacity of a resource to provide required operations of a particular failure recovery and service need.
    N
    [11] 11. The method of any one of preceding claims, further comprising: N defining a cost function for balancing different recovery and service needs; and using the cost function in the generating of the optimised selection. 2
    [12] 12. The method of any one of preceding claims, wherein the genetic algorithm is * developed in a process comprising: 3 a) defining a group of actions for different recovery and service resources; N b) defining possible states of different recovery and service needs for each of N 30 a plurality of different groups; c) defining a set of algorithm individuals for different recovery and service resources and assigning each individual with given genetic encoding, wherein the different recovery and service resources may have different capabilities; and initialising genetic encoding of at least some of the programmed individuals; wherein the initialising may be performed by randomising; d) defining an evaluation measure such as a cost function of prevailing states of each of the different groups; e) generating a test status in which the states are simulated or formed using historic data and/or randomised data for each of the groups and for a queue of pending recovery and service resource requests; f) deciding by each individual a new task from the queue based on the genetic encoding of the individual in question; the prevailing states and the queue; computing a new evaluation measure after each decision; and storing the computed new evaluation measure for each individual; g) updating the test status for each of the individuals; wherein the updating comprises updating the queue based on historic data or simulation and taking into account the decisions made by each individual; h) for a number of times and repeating steps, repeating steps f) and g) i) evaluating performance of each individual based on the stored evaluation measure and selecting a suitable subset of the individuals; j) creating new individuals at least partly based on the genetic encoding of the subset of the individuals and optionally included mutations; k) repeating evaluation for a formed new generation of individuals by S repeating steps e) to j) unless or until one or more termination criteria are O 25 achieved, such as the evaluation measure meeting a given threshold 2 and/or reaching a given number of generations. N
    [13] 13. A computer program comprising computer executable program code which E when executed by at least one processor causes an apparatus at least to perform ER the method of any one of preceding claims. o S 30
    [14] 14. A computer program product comprising a non-transitory computer readable N medium having the computer program of claim 13 stored thereon.
    [15] 15. An apparatus comprising means for performing the method of any one of claims 1 to 12.
    O Ql
    O
    N o Q@
    LO
    N
    I [am o
    O
    N o
    LO
    O Ql
    O
    N
    类似技术:
    公开号 | 公开日 | 专利标题
    CN107133645A|2017-09-05|Passenger is estimated to cancel an order method, equipment and the storage medium of behavior
    EP3690767A1|2020-08-05|Method and apparatus for determining risk management decision-making critical values
    CN107336681A|2017-11-10|The control method and device of vehicle
    CN109886699A|2019-06-14|Activity recognition method and device, electronic equipment, storage medium
    WO2021068513A1|2021-04-15|Abnormal object recognition method and apparatus, medium, and electronic device
    CN108574933B|2020-11-27|User track recovery method and device
    CN111984400A|2020-11-24|Memory allocation method and device of neural network
    FI129343B|2021-12-15|Real-time process control with constrained resources
    CN110231990B|2021-06-11|Block chain resource optimal allocation method and device based on secondary auction
    CN112488831A|2021-03-12|Block chain network transaction method and device, storage medium and electronic equipment
    CN111209930A|2020-05-29|Method and device for generating credit granting strategy and electronic equipment
    CN112734005A|2021-04-30|Method and device for determining prediction model, electronic equipment and storage medium
    CN109976901B|2021-04-16|Resource scheduling method, device, server and readable storage medium
    CN111221827A|2020-06-02|Database table connection method and device based on graphic processor, computer equipment and storage medium
    CN109101329B|2021-02-12|Fine-grained task allocation method and system for acquiring data through multiple mobile terminals
    CN108694672A|2018-10-23|Order processing method, apparatus, computer equipment and storage medium
    CN110021166B|2021-09-07|Method and device for processing user travel data and computing equipment
    CN106557787B|2019-07-02|It is leapfroged the user's clustering method and device of algorithm based on quantum
    JP2020030738A|2020-02-27|Method for analyzing learning data and computing system
    CN109788489B|2022-03-18|Base station planning method and device
    TW202145031A|2021-12-01|A management method, a device for gene sequencing progress, a computer device and a storage media
    CN111330269B|2021-03-26|Application difficulty adjustment and strategy determination method, device, system, equipment and medium
    CN112887371A|2021-06-01|Edge calculation method and device, computer equipment and storage medium
    CN113850390A|2021-12-28|Method, device, equipment and medium for sharing data in federal learning system
    CN112598470A|2021-04-02|Service data processing method, device and storage medium
    同族专利:
    公开号 | 公开日
    FI129343B|2021-12-15|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    法律状态:
    2021-12-15| FG| Patent granted|Ref document number: 129343 Country of ref document: FI Kind code of ref document: B |
    优先权:
    申请号 | 申请日 | 专利标题
    FI20205929A|FI129343B|2020-09-25|2020-09-25|Real-time process control with constrained resources|FI20205929A| FI129343B|2020-09-25|2020-09-25|Real-time process control with constrained resources|
    [返回顶部]