EP4338019A1 - Procédé et système de configuration pour configurer un dispositif de commande de machine - Google Patents

Procédé et système de configuration pour configurer un dispositif de commande de machine

Info

Publication number
EP4338019A1
EP4338019A1 EP22747991.2A EP22747991A EP4338019A1 EP 4338019 A1 EP4338019 A1 EP 4338019A1 EP 22747991 A EP22747991 A EP 22747991A EP 4338019 A1 EP4338019 A1 EP 4338019A1
Authority
EP
European Patent Office
Prior art keywords
action execution
tree
machine
action
trees
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
EP22747991.2A
Other languages
German (de)
English (en)
Inventor
Dieter Bogdoll
Markus Michael Geipel
Daniel Hein
Johannes Kehrer
Carlos Andres Palacios Valdes
Axel Reitinger
Ferdinand Strixner
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.)
Siemens AG
Original Assignee
Siemens AG
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 Siemens AG filed Critical Siemens AG
Publication of EP4338019A1 publication Critical patent/EP4338019A1/fr
Pending legal-status Critical Current

Links

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
    • G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/409—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by using manual data input [MDI] or by using control panel, e.g. controlling functions with the panel; characterised by control panel details or by setting parameters
    • 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
    • G05B2219/36—Nc in input of data, input key till input tape
    • G05B2219/36121—Tree oriented menu, go to root, scroll up down, select mode

Definitions

  • Complex machines such as robots, motors, machine tools, turbines, 3D printers, production plants or motor vehicles generally require a complex configured controller for productive operation in order to specifically optimize the performance of the controlled machine.
  • the performance to be optimized can relate, for example, to power, yield, resource requirements, efficiency, precision, pollutant emissions, stability, wear and/or other target parameters of the machine.
  • Machine controls are often configured by experts during design or commissioning of the machine to be controlled. However, such configuration by experts is often relatively time-consuming. In many cases, additional effort is required to optimize the configuration.
  • Machine learning Data-driven methods of machine learning are increasingly being used to automatically configure machine controls.
  • a machine controller can be trained to derive the specific control actions from the machine's current operating signals that specifically bring about a desired or otherwise optimal behavior of the machine.
  • a large number of known machine learning methods, in particular methods of reinforcement learning, are available for these purposes.
  • Predetermined action patterns are read in for configuring a machine controller by means of an action execution tree specifying an execution of actions by a machine. Furthermore, a large number of action execution trees are generated for the machine. The action execution trees can in particular be so-called behavior trees. For a respectively generated action execution tree, a performance for controlling the machine is determined using the respective action execution tree. Furthermore, the specified action patterns are searched for in the respective action execution tree. An action pattern found in the respective action execution tree is then at least partially replaced by a reference to the specified action pattern. A tree size of the action execution tree modified in this way is also determined. Using the generated action execution trees, a numerical optimization method is then used to determine an action execution tree that is optimized with regard to greater performance and a smaller tree size and is output for configuring the machine controller.
  • a configuration system, a computer program product and a computer-readable, preferably non-volatile, storage medium are provided for carrying out the method according to the invention.
  • the method according to the invention and the configuration system according to the invention can be carried out, for example, by means of one or more computers, processors, application-specific integrated circuits (ASIC), digital signal processors (DSP) and/or so-called "Field Programmable Gate Arrays”.
  • an action execution tree can be reduced in size in many cases. In this way, the complexity of the action execution tree can usually be reduced. Insofar as the tree size of action execution trees is used as an optimization criterion in addition to performance, the optimization can be driven in the direction of action execution trees that have predetermined action patterns. In this way, an action execution tree or a configuration can be determined that is usually both high-performance and low in complexity. Configurations with low complexity are generally easier to interpret by experts and are therefore easier to validate and/or further develop.
  • the multiplicity of action execution trees can be generated at least partially on the basis of the action pattern.
  • action execution trees can be assembled at least partially from the action patterns. In this way, action execution trees can be generated in a targeted manner, which have known and/or easily interpretable action patterns.
  • a predefined initial action execution tree can be read.
  • the large number of action execution trees can then be at least partially based on the initial action on execution tree are generated.
  • the initial action execution tree can in particular be an action execution tree created, tested and/or validated by an expert. In this way, expert specifications can be taken into account when determining an optimized action execution tree.
  • the initial action execution tree may be executed by an interpreter, determining a reference sequence of machine actions specified by the initial action execution tree.
  • the generated action execution trees can then each be executed by the interpreter, with a respective sequence of machine actions specified by the respective generated action execution tree being determined.
  • a generated action execution tree whose respective sequence does not match the reference sequence can thus be discarded.
  • optimization can be restricted to action execution trees that reproduce the reference sequence. If the initial action execution tree is advantageously chosen in such a way that the machine actions specified thereby or a predetermined selection of these machine actions only include permissible or required target machine actions, it can be ensured in this way that the optimization remains limited to action execution trees, meet the specified development goals.
  • the numerical optimization experience can be a genetic optimization method, which also generates the multiplicity of action execution trees.
  • the genetic optimization method can be initialized in particular by the initial action execution tree.
  • a fitness function can be provided for the genetic optimization method, which assigns greater fitness to a higher performance and/or a smaller tree size than to a lower performance and/or a larger tree size. In this way, the optimization can preferably be driven in the direction of high-performance and/or less complex configurations. A large number of efficient standard routines are available for carrying out such genetic optimization methods.
  • a Pareto front can be determined for the large number of action execution trees, with an increase in performance and a reduction in the tree size being used as Pareto target criteria.
  • the optimized action execution tree can then be derived from Pareto front action execution trees.
  • the optimized action execution tree can be selected or interpolated from action execution trees of the Pareto front. While a selection is particularly easy to carry out, an interpolation can also combine several advantageous properties of action execution trees.
  • an optimized action execution tree can be determined, which is usually both performant and has a small tree size and thus low complexity.
  • a Pareto front is also understood to mean a set of action execution trees whose distance from a mathematically exact Pareto optimum is less than a given threshold value, for example.
  • a restriction to action execution trees of the Pareto front usually opens up a space of possible action execution trees. considerably restricted, with non-optimal action execution trees in particular being eliminated. Determining the optimized action execution tree and, if necessary, further optimization can be simplified considerably in this way.
  • action execution trees of the Pareto front can be fed into the genetic optimization experience.
  • action execution trees not included in the Pareto Front may be discarded.
  • the genetic optimization process can be enriched with advantageous action execution trees.
  • the optimization process can be terminated after a Pareto front has been determined.
  • a simulation model of the machine a data-driven model of the machine, the machine itself and/or a machine similar to it can be controlled using the respective action execution tree and a performance of the machine resulting therefrom can be determined.
  • a surrogate mode11 of the machine which requires fewer computing resources than a complete physical simulation, can preferably be used as the simulation model.
  • an edge number, a node number and/or a tree depth of the modified action execution tree or a weighted combination of the edge number, the node number and/or the tree depth can be determined.
  • the above criteria allow a particularly simple evaluation of the tree size or the complexity of a respective action execution tree.
  • Figure 1 shows a machine with a machine controller for controlling the machine
  • Figure 2 an action execution tree
  • FIG. 3 shows a configuration system according to the invention when configuring a machine control
  • Figure 4 shows a modification of an action execution tree.
  • FIG. 1 shows a schematic representation of a machine M, which is coupled to a machine controller CTL for controlling the machine M.
  • the machine M can be a robot, a motor, a machine tool, a turbine, a manufacturing plant, a motor vehicle, a 3D printer, a computer, a mechanical system, an electrical system or another device or another be plant.
  • the machine M has a sensor system S for continuously measuring and/or acquiring operating signals BS from the machine M and, if applicable, data from an area surrounding the machine M.
  • the machine control CTL can be configured with the aid of a computer and can be implemented externally to the machine M as part of the machine M or in whole or in part.
  • the machine control CTL is to be configured by means of the invention in such a way that the machine M is controlled in an optimized manner. Under a control is also a rule, as well as an output and use of control-relevant, i. H. understood for the targeted influencing of the machine M contributing data or control signals.
  • the machine control CTL is to be in particular a genetic optimization method, can be configured in such a way that operation of the machine M is optimized as a function of the detected operating signals BS.
  • the term optimization is generally also understood to mean an approximation to an optimum. In this case, a performance of the machine M should preferably be increased.
  • a control behavior of the machine control CTL is defined by its configuration.
  • the configuration includes in particular an action execution tree, preferably in the form of a behavior tree known to those skilled in the art, through which actions to be executed by the machine M and in particular sequences of actions are specified.
  • An action execution tree is represented by a tree-linked directed graph, which is preferably implemented as a tree-linked data structure.
  • FIG. 2 illustrates a basic structure of such an action execution tree BT, which can be configured as a behavior tree, for example.
  • the action execution tree BT comprises a directed, tree-like, branched graph with a plurality of nodes A0, A1, A2, A3, A4, . . . which are connected by directed edges.
  • Certain nodes of the graph represent machine actions that are to be executed by the machine M depending on the given conditions.
  • the directed edges of the graph can be used to specify relationships between conditions and machine actions to be carried out conditionally or a sequence of machine actions to be carried out.
  • a process flow determined by a sequence of machine actions to be carried out branches depending on the specified conditions along the tree structure of the graph.
  • An action execution tree BT configured as a behavior tree can in particular have so-called fallback nodes, sequence nodes, control flow nodes and Decorators such as “always”, “guard”, “not”, “retry”, “time limit”, “wait for” and/or “while” include.
  • an optimized action execution tree BTO preferably in the form of an optimized behavior tree, is transmitted to the machine control CTL for the optimized configuration of the machine control CTL.
  • a control behavior of the machine control CTL set in this way is often also referred to as a policy.
  • the optimized action execution tree BTO is determined by a method according to the invention.
  • the continuously detected operating signals BS of the machine M are transmitted to the machine control system CTL by the sensor system S.
  • the operating signals BS can provide information about the operating states of the machine M, about positions or movements of components, about switching states, about control states, about control actions, about physical, chemical or electrical measured variables and/or about other parameters relevant to the operation of the machine M include parameters.
  • control signals CS are transmitted from the machine control CTL to the machine M for optimized control of the latter.
  • the control signals CS are generated depending on the optimized action execution tree BTO in such a way that a sequence of machine actions specified by the optimized action execution tree BTO is executed.
  • FIG. 3 shows a schematic representation of a configuration system KS according to the invention when configuring the machine control CTL.
  • the machine control CTL can form part of the configuration system KS or be arranged completely or partially externally to the configuration system KS.
  • the configuration system KS and / or the machine control CTL have one or several processors for carrying out the method according to the invention and one or more memories for storing data to be processed.
  • the machine control CTL should be configured by an optimized action execution tree BTO in such a way that the machine M is controlled in an optimized manner.
  • the optimized action execution tree BTO is determined by the configuration system KS using a genetic optimization method.
  • action patterns AP specified by the configuration system KS and an initial action execution tree BTI are read in, preferably in the form of a behavior tree.
  • the action patterns AP can in particular specify sequences of machine actions and/or conditions for their execution.
  • the action patterns AP are preferably represented by subgraphs or subtrees of action execution trees or behavior trees.
  • idiomatic, easily interpretable, reliable, validated and/or permissible action patterns that have proven themselves for solving control subtasks can be specified as action patterns AP.
  • the action patterns AP are read in from a database DB or from another widely available action pattern library.
  • the initial action execution tree BTI is preferably created by an expert USR according to technical specifications, requirements or other technical constraints for controlling the machine M and fed into the configuration system KS.
  • the initial action execution tree BTI specifies an execution of permitted and/or required machine actions and thus forms a reference for the control of the machine M or for a target configuration of the machine control CTL. This is how the initial encodes Action execution tree BTI to a certain extent expert knowledge about the control of the machine M.
  • selection information SI is fed into the configuration system KS by the expert USR.
  • an optimized action execution tree BTO a genetic optimization method is used, among other things, in the present exemplary embodiment.
  • Advantageous action execution trees are searched for and further optimized in a space of possible action execution trees.
  • a generator GEN of the configuration system KS generates—at least in part randomly—a multiplicity of action execution trees BT, each of which specifies a possible execution of machine actions by the machine M.
  • a so-called fitness of the machine actions specified thereby is evaluated for the generated action execution trees BT.
  • Action execution trees BT with higher fitness are more likely to be used to generate further action execution trees BT in a subsequent iteration of the genetic optimization process.
  • Action execution trees BT with lower fitness are correspondingly sorted out with a higher probability and/or replaced by newly generated action execution trees. In this way, action execution trees BT with higher fitness are increasingly being generated by the generator GEN.
  • a performance for controlling the machine M is determined using the action execution tree BT to be evaluated, a tree size of this action execution tree BT and/or a weighted combination of the performance and the tree size to determine a respective fitness. In this way, the generation of action execution trees is driven towards higher performance and smaller tree size.
  • a large number of standard procedure available. Alternatively or additionally, other methods of machine learning can also be used for optimization.
  • the initial action execution tree BTI and the action pattern AP are fed into the generator GEN.
  • the action execution trees BT are then generated by the generator GEN starting from the initial action execution tree BTI and using at least some of the action patterns AP.
  • the generated action execution trees BT are transmitted from the generator GEN to an interpreter IP of the configuration system KS.
  • the initial action execution tree BTI and the selection information SI are also fed into the interpreter IP.
  • the interpreter IP is used to execute action execution trees, with a sequence of machine actions specified by the relevant action execution tree being determined in each case. Using the selection information SI, the individual machine actions that must be carried out according to the technical specifications are selected. In this way, for the initial action execution tree BTI, the interpreter IP determines a reference sequence RSQ of machine actions that must be carried out. Furthermore, a specified sequence SQ of machine actions is determined by the interpreter IP for each generated action execution tree BT. The reference sequence RSQ, the sequences SQ and the associated action execution trees BT are transmitted to a filter F of the configuration system KS.
  • the filter F serves in particular to filter the generated action execution trees BT.
  • the sequence SQ determined for a respective action execution tree BT is compared with the reference sequence RSQ. If the sequence SQ does not match the reference sequence RSQ this sequence SQ and the relevant action execution tree BT are discarded and not forwarded through the filter F. If, on the other hand, the sequence SQ agrees with the reference sequence RSQ, this sequence SQ and the relevant action execution tree BT are forwarded through the filter F.
  • the action execution trees BT forwarded by the filter F and the action patterns AP are fed into an action pattern search module APS of the configuration system KS.
  • the action pattern search module APS searches for the action pattern AP in each of the supplied action execution trees BT. If no action pattern AP is found in a respective action execution tree BT, this action execution tree is forwarded unchanged by the action pattern search module APS. If, on the other hand, an action pattern AP is found in an action execution tree BT, this action execution tree BT is modified by the action pattern search module APS.
  • FIG. Such a modification of an action execution tree BT is illustrated by FIG.
  • a subtree comprising the nodes A2, A3 and A4 of the action execution tree BT corresponds to a predefined action pattern API.
  • this subtree or the action pattern API in the action execution tree BT is replaced by a reference L to the action pattern API, whereby a modified action execution tree BT' is formed.
  • the reference L has several nodes, here A2,
  • modified action execution trees are referred to below with the same reference sign BT as unchanged action execution trees.
  • the unchanged or modified action execution trees BT are transmitted by the action pattern search module APS to a complexity assessor EVC of the configuration system KS.
  • the purpose of the complexity evaluator EVC is to quantify a tree size BG for a respective action execution tree BT.
  • the complexity evaluator EVC counts nodes and/or edges of a respective action execution tree BT and/or its tree depth is determined and a tree size BG is derived therefrom.
  • a sum of the number of nodes, number of edges and/or tree depth weighted by predetermined weighting factors can preferably be determined and output as tree size BG.
  • the determined tree size BG is transmitted from the complexity evaluator EVC to a Pareto optimizer PO of the configuration system KS.
  • action execution trees BT forwarded by the filter F are fed into a performance evaluator EVP of the configuration system KS and into the Pareto optimizer PO.
  • sequences SQ forwarded by the filter F are also fed into the performance evaluator EVP.
  • the performance evaluator EVP serves the purpose of quantifying a performance of the machine M controlled by means of this action execution tree BT for a respective action execution tree BT. In this way, a control performance of a machine control CTL configured by this action execution tree BT is evaluated to a certain extent.
  • the performance to be determined can in particular be an output, a yield, a target value compliance, a cycle rate, a product quality, a speed, a time requirement, a running time, a precision, an error rate, a consumption of resources, an effectiveness, an efficiency, a pollutant emission, one stability, one wear, one Service life, physical behavior, mechanical behavior, chemical behavior, electrical behavior, a constraint to be complied with or other target parameters of the machine M to be controlled that are to be optimized.
  • the performance evaluator EVP determines a respective performance value PV for the respective action execution tree BT and in particular for the associated sequence SQ of machine actions, which quantifies the performance of the machine M executing the sequence SQ.
  • the performance evaluator EVP has a simulation model SIM of the machine M .
  • SIM a simulation model
  • a behavior of the machine M induced by the respective action execution tree BT or the respective sequence SQ of machine actions is simulated for a large number of magazines.
  • a cumulative reward or a cumulative yield of the simulated behavior is measured.
  • a so-called surrogate model for example a data-driven model of the machine M, is preferably used as the simulation model, which generally requires fewer computing resources than a detailed physical simulation.
  • a cumulative reward determined in this way can be output as the resulting performance value PV.
  • the performance value PV determined for the respective action execution tree BT is transmitted from the performance evaluator EVP to the Pareto optimizer PO.
  • the Pareto optimizer PO is used to carry out a Pareto optimization and in particular to determine a Pareto front PF for the action execution trees BT evaluated by the complexity evaluator EVC and the performance evaluator EVP.
  • a Pareto optimization is a multi-criteria optimization in which several different target criteria, so-called Pareto target criteria, are independently taken into account.
  • Pareto front PF determined.
  • Such a Pareto front is often also referred to as a Pareto set.
  • a Pareto front, here PF are those solutions of a multi-criteria optimization problem where one goal criterion cannot be improved without worsening another goal criterion.
  • a Pareto front forms a set of optimal compromises, so to speak. In particular, solutions that are not contained in the Pareto front can still be improved with regard to at least one target criterion.
  • a Pareto front PF for the set of action execution trees BT is determined by the Pareto optimizer PO, with an increase in performance and a reduction in the tree size being used as Pareto target criteria.
  • a Pareto front PF within the generated action execution trees BT is determined by the Pareto optimizer PO depending on the transmitted tree sizes BG and performance values PV of the generated action execution trees BT.
  • An optimized action execution tree BTO is then selected or interpolated from the resulting Pa reto front PF. If necessary, predetermined selection criteria, in particular Special one or more other optimization criteria to be applied.
  • action execution trees BT(PF) of the Pareto front PF can be fed into the genetic optimization method in order to enrich the genetic optimization method with advantageous action execution trees.
  • a corresponding feedback of action execution trees BT(PF) of the Pareto front PF from the Pareto optimizer PO to the generator GEN is indicated in FIG. 3 by a dotted arrow.
  • the optimized action execution tree BTO is intended to be output for configuring the machine control CRL and/or transmitted directly to the machine control CTL in order to configure it for optimized control of the machine M.
  • the resulting configuration of the machine control CTL usually leads to a control behavior that is the same shows both high performance and low complexity at an early stage.
  • the last-named property leads to configurations that can usually be interpreted much better by experts and are therefore easier to validate and/or develop further.

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Manufacturing & Machinery (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

Pour configurer un dispositif de commande de machine (CTL) au moyen d'un arbre d'exécution d'action (BT), des motifs d'action prédéfinis (AP) sont lus. Une multiplicité d'arbres d'exécution d'action (BT) pour une machine (M) à commander est également générée. Pour un arbre d'exécution d'action (BT) respectivement généré, une performance (PV) pour commander la machine (M) sur la base de l'arbre d'exécution d'action (BT) respectif est déterminée. Les motifs d'action prédéfinis (AP) sont également recherchés dans l'arbre d'exécution d'action (BT) respectif. Un motif d'action (AP) trouvé dans l'arbre d'exécution d'action (BT) respectif est ensuite remplacé au moins en partie par une référence (L) au motif d'action (AP) prédéfini. Une taille d'arbre (BG) de l'arbre d'exécution d'action ainsi modifié (BTM) est en outre déterminée. Sur la base des arbres d'exécution d'action (BT) générés, un procédé d'optimisation numérique est ensuite utilisé pour déterminer un arbre d'exécution d'action (BTO) qui est optimisé en ce qui concerne une meilleure performance et une taille d'arbre plus petite, et ceci est délivré afin de configurer le dispositif de commande de machine (CTL).
EP22747991.2A 2021-07-21 2022-07-07 Procédé et système de configuration pour configurer un dispositif de commande de machine Pending EP4338019A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP21186997.9A EP4123399A1 (fr) 2021-07-21 2021-07-21 Procédé et système de configuration permettant de configurer une commande de machine
PCT/EP2022/068817 WO2023001561A1 (fr) 2021-07-21 2022-07-07 Procédé et système de configuration pour configurer un dispositif de commande de machine

Publications (1)

Publication Number Publication Date
EP4338019A1 true EP4338019A1 (fr) 2024-03-20

Family

ID=77021169

Family Applications (2)

Application Number Title Priority Date Filing Date
EP21186997.9A Withdrawn EP4123399A1 (fr) 2021-07-21 2021-07-21 Procédé et système de configuration permettant de configurer une commande de machine
EP22747991.2A Pending EP4338019A1 (fr) 2021-07-21 2022-07-07 Procédé et système de configuration pour configurer un dispositif de commande de machine

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP21186997.9A Withdrawn EP4123399A1 (fr) 2021-07-21 2021-07-21 Procédé et système de configuration permettant de configurer une commande de machine

Country Status (3)

Country Link
EP (2) EP4123399A1 (fr)
CN (1) CN117693719A (fr)
WO (1) WO2023001561A1 (fr)

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3651082A1 (fr) * 2018-11-12 2020-05-13 Siemens Aktiengesellschaft Procédé et système d'aide à la génération des données d'entraînement destinées à la configuration à base d'apprentissage d'un système technique
EP3835894A1 (fr) * 2019-12-09 2021-06-16 Siemens Aktiengesellschaft Procédé et appareil de configuration d'une commande de machine
CN112882449A (zh) * 2021-01-13 2021-06-01 沈阳工业大学 一种多品种小批量多目标柔性作业车间能耗优化调度方法

Also Published As

Publication number Publication date
CN117693719A (zh) 2024-03-12
WO2023001561A1 (fr) 2023-01-26
EP4123399A1 (fr) 2023-01-25

Similar Documents

Publication Publication Date Title
EP2999998B1 (fr) Méthode de détermination d'un modèle d'une grandeur de sortie d'un système technique
DE102010005648A1 (de) Zielbezogenes Rapid Control Prototyping
WO2021058223A1 (fr) Procédé d'application de fonctions de conduite automatisée de manière efficace et simulée
DE102018116911A1 (de) Verfahren zur Erzeugung von Quellcode
EP3629151A1 (fr) Procédé de changement de modèles pour la génération du code source
EP2363809B1 (fr) Méthode pour optimer un programme de contrôle pour des actuateurs
DE112008003511T5 (de) Integrierter technischer Analyseprozess
WO2023001561A1 (fr) Procédé et système de configuration pour configurer un dispositif de commande de machine
DE102011007434A1 (de) Simulationsmodell für eine Windenergieanlage sowie Erstellung und Verwendung
DE102019212912A1 (de) Komprimieren eines tiefen neuronalen Netzes
EP3650964B1 (fr) Procédé de commande ou de régulation d'un système technique
EP3944158A1 (fr) Procédé mis en uvre par ordinateur destiné à l'apprentissage d'une machine permettant de faire fonctionner un composant de véhicule et procédé de fonctionnement d'un composant de véhicule
EP3989012A1 (fr) Procédé et système de configuration destinés à la configuration d'un dispositif de commande pour un système technique
WO2022253541A1 (fr) Procédé et système de configuration pour configurer un dispositif de commande de machine
EP3637354A1 (fr) Procédé de commande d'un appareil électronique au moyen d'un code de programme optimal
EP3712720B1 (fr) Procédé et dispositif de commande d'une installation technique avec une architecture simplex adaptative
EP3489773A1 (fr) Procédé de commande assisté par ordinateur d'un système technique, en particulier d'une installation de production d'énergie
WO2023144014A1 (fr) Système technique à utilisation de services en nuage
DE102007007601A1 (de) Vorrichtung zur Herstellung eines Steuerprogramms aus Funktionsmodulen, mehrachsige Maschine, Verfahren und graphische Benutzerschnittstelle
EP4068011A1 (fr) Commande de machine, ainsi que procédé de configuration de la commande de machine
EP4075210A1 (fr) Procédé d'optimisation pour une unité de commande, commande, installation automatisée et produit programme informatique
EP4131053A1 (fr) Procédé et système de commande d'un système de production
EP3961517A1 (fr) Procédé et système de génération d'une solution industrielle à l'aide d'une intelligence artificielle
DE102022212901A1 (de) Automatisches Ermitteln einer optimalen Architektur für ein neuronales Netzwerk
DE102020213366A1 (de) Verfahren zum Bereitstellen eines Algorithmus maschinellen Lernens, Verfahren und System zur Klassifikation und/oder Regression von Prozessdaten und/oder Prozess-Metadaten

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20231215

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR