CN115048826A - Variable prediction method, device and equipment of system dynamics model of industrial equipment - Google Patents

Variable prediction method, device and equipment of system dynamics model of industrial equipment Download PDF

Info

Publication number
CN115048826A
CN115048826A CN202210978259.8A CN202210978259A CN115048826A CN 115048826 A CN115048826 A CN 115048826A CN 202210978259 A CN202210978259 A CN 202210978259A CN 115048826 A CN115048826 A CN 115048826A
Authority
CN
China
Prior art keywords
node
nodes
measurable
target node
dynamics model
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
CN202210978259.8A
Other languages
Chinese (zh)
Other versions
CN115048826B (en
Inventor
田春华
张�浩
张硕
徐地
孟越
袁文飞
胡坤
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.)
Kunlun Intellectual Exchange Data Technology Beijing Co ltd
Original Assignee
Kunlun Intellectual Exchange Data Technology Beijing Co ltd
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 Kunlun Intellectual Exchange Data Technology Beijing Co ltd filed Critical Kunlun Intellectual Exchange Data Technology Beijing Co ltd
Priority to CN202210978259.8A priority Critical patent/CN115048826B/en
Publication of CN115048826A publication Critical patent/CN115048826A/en
Application granted granted Critical
Publication of CN115048826B publication Critical patent/CN115048826B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a variable prediction method, a device and equipment of a system dynamics model of industrial equipment, wherein the method comprises the following steps: acquiring a target node in a system dynamics model of industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node; constructing a deduction tree of a target node according to a topological relation between a plurality of measurable nodes and non-measurable nodes and a preset prediction autovariable set; and determining whether the target node is predictable according to the deduction tree, wherein the preset prediction independent variable set comprises the following steps: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model. The scheme of the invention can accurately identify the predictability of the variables in the model and improve the accuracy of data analysis.

Description

Variable prediction method, device and equipment of system dynamics model of industrial equipment
Technical Field
The invention relates to the technical field of data information processing of industrial equipment, in particular to a variable prediction method, a variable prediction device and variable prediction equipment of a system dynamics model of industrial equipment.
Background
The system dynamics model of the industrial equipment expresses key elements and driving relations of a field operation mechanism, and the complexity of the model can be determined according to research problems, so that the cross-field communication efficiency and the information correctness are improved;
in the existing system dynamics model, complex driving relations exist among different index variables, when relevant data analysis of the system dynamics model of the industrial equipment is carried out, the driving relations among the variables are concerned more in the data analysis, and for a machine learning subject, whether the variables can be measured or not limits the relation structure among the variables, so that the variable combination space is reduced; meanwhile, due to the complex model structure, the predictability of the variable nodes in the system dynamics model and the decision variable nodes of the variable nodes cannot be determined, and the related index variables of the industrial equipment cannot be accurately and efficiently analyzed.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a variable prediction method, a variable prediction device and variable prediction equipment of a system dynamics model of industrial equipment, so as to accurately identify the predictability of index variables in the model and the driving relation among corresponding index variables, and improve the accuracy and efficiency of data analysis.
To solve the above technical problem, an embodiment of the present invention provides a variable prediction method for a system dynamics model of an industrial device, including:
acquiring a target node in a system dynamics model of industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
constructing a deduction tree of the target node according to the topological relation between the plurality of measurable nodes and the non-measurable nodes and a preset prediction autovariate set;
determining whether the target node is predictable according to the deduction tree, wherein the preset prediction independent variable set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
Optionally, constructing a deduction tree of the target node according to the topological relations between the measurable nodes and the non-measurable nodes and a preset prediction autovariate set, where the deduction tree includes:
in the system dynamics model, according to the topological relations between the plurality of nodes and the unmeasured nodes, the target node and all the associated nodes of the target node are sequentially decomposed to obtain a decomposition amount node set
Figure 226769DEST_PATH_IMAGE001
And set of topological relationships between the resolution nodes
Figure 339082DEST_PATH_IMAGE002
Wherein, in the step (A),
Figure 15526DEST_PATH_IMAGE001
and
Figure 213290DEST_PATH_IMAGE002
composition diagram
Figure 26525DEST_PATH_IMAGE003
；
According to the diagram
Figure 117978DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 624045DEST_PATH_IMAGE005
And constructing a deduction tree of the target node.
Optionally, according to said figure
Figure 309105DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 535818DEST_PATH_IMAGE005
And constructing a deduction tree of the target node, including:
in the figure
Figure 622722DEST_PATH_IMAGE004
In the method, the graph is traversed from the layer where the target node is located in sequence
Figure 158746DEST_PATH_IMAGE004
Each layer of decomposition volume node;
if the current decomposition value node is in the preset prediction autovariate set
Figure 65522DEST_PATH_IMAGE005
Adding the current decomposition amount node into the actually used prediction variable set when the medium or current decomposition amount node is an immeasurable anchor quantitative node
Figure 220560DEST_PATH_IMAGE006
(ii) a The anchor quantity node is a node corresponding to an index variable which is not influenced by other index variables in the system dynamics model;
according to the prediction variable set
Figure 37337DEST_PATH_IMAGE006
And topological relation set
Figure 619628DEST_PATH_IMAGE002
And obtaining a current graph, and determining the current graph as a deduction tree of the target node.
Optionally, the variable prediction method of the system dynamics model of the industrial equipment further includes:
according to the diagram
Figure 403914DEST_PATH_IMAGE004
Obtaining the downstream node independently influenced by the current decomposition quantity node, wherein the downstream node is in the preset prediction autovariate set
Figure 97063DEST_PATH_IMAGE005
Adding the downstream node into the prediction variable set actually used
Figure 892981DEST_PATH_IMAGE006
；
According to the diagram
Figure 521539DEST_PATH_IMAGE004
Obtaining the upstream node of the current decomposition quantity node, if the upstream node is not in the graph
Figure 668487DEST_PATH_IMAGE004
Adding the upstream node to the graph
Figure 289961DEST_PATH_IMAGE004
。
Optionally, determining whether the target node is predictable according to the deduction tree includes:
traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence;
and if all lower-layer nodes of the target node are predictable nodes, determining the target node to be predictable, otherwise, determining the target node to be unpredictable.
Optionally, the variable prediction method for the system dynamics model of the industrial equipment further includes:
if all the driving nodes of the current lower layer node of the traversed target node are not all in the preset prediction autovariate set
Figure 205965DEST_PATH_IMAGE005
And determining the current lower node as a result node.
Optionally, the variable prediction method of the system dynamics model of the industrial equipment further includes:
if the set of predicted variables is
Figure 864479DEST_PATH_IMAGE006
Including a set of predetermined predictive arguments
Figure 99721DEST_PATH_IMAGE005
An additional unmeasurable anchor quantity node, where the resulting node is predictable, determining that the target node is partially predictable.
An embodiment of the present invention further provides a variable prediction apparatus for a system dynamics model of an industrial device, including:
the acquisition module is used for acquiring a target node in a system dynamics model of the industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
the processing module is used for constructing a deduction tree of the target node according to the topological relation between the measurable nodes and the non-measurable nodes and a preset prediction autovariate set; determining whether the target node is predictable according to the deduction tree, wherein the preset prediction autovariate set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
Embodiments of the present invention also provide a computing device, comprising: a processor, a memory storing a computer program which, when executed by the processor, performs the method as described above.
Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method as described above.
The scheme of the invention at least comprises the following beneficial effects:
obtaining a target node in a system dynamics model of industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the plurality of nodes and the non-measurable intermediate quantity nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node; constructing a deduction tree of the target node according to the topological relation between the plurality of measurable nodes and the non-measurable nodes and a preset prediction autovariate set; determining whether the target node is predictable according to the deduction tree, wherein the preset prediction variable set comprises: a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model; the predictability of the index variables and the driving relation among the index variables are accurately identified, and the accuracy and the efficiency of data analysis are improved.
Drawings
FIG. 1 is a flow chart of a method for predicting variables of a system dynamics model according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the operation of a coal pulverizer provided in accordance with an alternative embodiment of the present invention;
FIG. 3 is a schematic diagram of a system dynamics model architecture provided in accordance with an alternative embodiment of the present invention;
FIG. 4 is a schematic diagram of the system dynamics model of FIG. 3 with the index variables replaced with nodes;
FIG. 5 is a block diagram of a target variable association node decomposition provided in an alternative embodiment of the present invention
Figure 400252DEST_PATH_IMAGE004
A schematic diagram of (a);
FIG. 6 is a block diagram of another target variable association node decomposition provided in an alternative embodiment of the present invention
Figure 295396DEST_PATH_IMAGE004
A schematic diagram of (a);
FIG. 7 is a schematic flow chart illustrating a variable prediction process of a system dynamics model according to an alternative embodiment of the present invention;
fig. 8 is a block diagram of a variable prediction apparatus of a system dynamics model according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As shown in fig. 1, an embodiment of the present invention provides a variable prediction method of a system dynamics model of an industrial device, including:
step 11, acquiring a target node in a system dynamics model of the industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
step 12, constructing a deduction tree of the target node according to the topological relation between the plurality of measurable nodes and the plurality of non-measurable nodes and a preset prediction autovariate set;
step 13, determining whether the target node is predictable according to the deduction tree, wherein the preset prediction independent variable set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
In this embodiment, the nodes corresponding to the index variables in the system dynamics model include: measurable nodes corresponding to the observable index variables and non-measurable nodes corresponding to the non-observable index variables; one observable index variable corresponds to one measurable node in measurable nodes corresponding to the observable index variables; among the non-observable nodes corresponding to the non-observable index variables, one non-observable index variable corresponds to one non-observable node, or among all the non-observable index variables, part of the non-observable index variables respectively correspond to one non-observable node, and the rest of the non-observable index variables correspond to one non-observable intermediate node after being simplified to obtain non-observable intermediate index variables; wherein the observable indicator variables represent variables that can be measured by associated equipment or instruments, and the non-observable indicator variables represent variables that cannot be directly measured and need to be obtained according to other observable indicator variables or causal relationships between non-observable indicator variables
The topological relations in the system dynamics model comprise: the nodes in the topological relation represent the directional driving relation among index variables of the industrial equipment in the production application process and the positions of the index variables in the system dynamic model;
taking a coal mill control system as an example, the coal mill control system comprises coal feeder rotation speed control, coal mill primary air volume control and coal mill outlet temperature control, a primary hot/cold air pipeline is commonly used by a plurality of coal mills, as shown in fig. 2, a working principle diagram of the coal mill control system is shown, a corresponding system dynamics model is shown in fig. 3, an observable index variable is shown in a solid line frame, an unobservable index variable is shown in a dotted line frame, and U1, U2 and U3 can respectively represent three unobservable index variables or three unobservable intermediate index variables obtained by simplifying a plurality of unobservable index variables; the directed connecting line represents the logic driving relation among different index variables;
for the sake of clarity, only the topological relationships among different index variables are considered here, the names of the index variables are ignored, and the different index variables are represented by numbered nodes in the system dynamics model, as shown in fig. 4, wherein nodes corresponding to observable index variables are represented by solid circles, nodes corresponding to unobservable index variables are represented by dashed circles, unobservable index variables or unobservable intermediate index variables U1, U2, U3 correspond to nodes 15, 20, and 25, respectively, and a connecting line indicates the logical driving relationships among the different index variables; observable index variables such as 'coal mill rotation speed', 'loading pressure', 'coal mill current', 'coal feeding amount' and the like respectively correspond to the node 16, the node 17, the node 18 and the node 4; unobservable index variables such as 'raw coal ash content' and 'coal powder demand' respectively correspond to the node 12 and the node 2;
the preset prediction independent variable set is a set of measurable nodes corresponding to a plurality of observable index variables in the system dynamic model which can be determined according to actual needs;
according to the preset prediction independent variable set and the topological relation in the system dynamics model, a deduction tree of the target node can be constructed; the deduction tree is used for representing the logic driving relation between the target node and part of nodes in the system dynamics model which have causal relation with the target node; the deduction tree further comprises at least one measurable node and/or at least one non-measurable node besides the target node, a logic driving relation exists between the at least one measurable node and/or at least one non-measurable node and the target node, and the at least one measurable node in the deduction tree is an intersection of the preset prediction independent variable set and a node set corresponding to the observable index variable in the system dynamics model;
according to the logic driving relationship between the target node and other nodes in the deduction tree and the predictability of other nodes, the predictability of the target node is researched and judged, the accuracy of the research and judgment of the predictability of the target node is guaranteed, then the accurate analysis of relevant index variables of industrial equipment is realized, and the efficiency and the accuracy of data analysis are further improved.
In an optional embodiment of the present invention, the step 12 may include:
step 121, in the system dynamics model, according to the topological relations between the plurality of nodes and the unmeasured nodes, sequentially decomposing the target node and all the associated nodes of the target node to obtain a decomposition quantity node set
Figure 390391DEST_PATH_IMAGE001
And set of topological relationships between the resolution nodes
Figure 246351DEST_PATH_IMAGE002
Wherein, in the step (A),
Figure 225940DEST_PATH_IMAGE001
and
Figure 585377DEST_PATH_IMAGE002
composition diagram
Figure 975907DEST_PATH_IMAGE003
；
Step 122, according to the graph
Figure 53584DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 961497DEST_PATH_IMAGE005
And constructing a deduction tree of the target node.
In this embodiment, all the associated nodes of the target node are obtained according to the topological relation in the system dynamical model, and each associated node is decomposed, that is, whether an upstream node or a downstream node exists in each associated node is determined according to the topological relation in the system dynamical model, and then the upstream node or the downstream node is further decomposed until the corresponding node cannot be decomposed any more, and a decomposition amount node set is obtained
Figure 316386DEST_PATH_IMAGE001
The decomposition quantity node set comprises the target node, a direct associated node of the target node and an upstream or downstream node of the associated node; forming a topological relation set according to the directed driving relation among the decomposition quantity nodes
Figure 18763DEST_PATH_IMAGE002
Further according to said scoreSolution node set
Figure 708371DEST_PATH_IMAGE001
And the set of topological relations
Figure 888816DEST_PATH_IMAGE002
Composition diagram
Figure 488425DEST_PATH_IMAGE003
The figure
Figure 971490DEST_PATH_IMAGE004
Representing a logical driving relationship among the target node, a direct associated node of the target node, and an upstream or downstream node of the associated node;
it should be noted that in the figures
Figure 289339DEST_PATH_IMAGE004
In the method, on the premise of not considering the driving relationship among the nodes, the target node may serve as an initial node, and a directly associated node of the target node may also serve as a first-layer decomposition amount node obtained by decomposing the target node, and each decomposition amount node in the first layer is decomposed in sequence to obtain a second-layer decomposition amount node; by analogy, until the obtained decomposition quantity nodes can not be decomposed, a deduction tree with multi-level decomposition quantity nodes is formed at the same time;
taking the node 6 and the node 26 in the system dynamics model diagram of fig. 4 as target nodes as an example; decomposing the directly related nodes 7, 15, 20 of the target node 6, as shown in fig. 5, obtaining the graph of the target node 6
Figure 132530DEST_PATH_IMAGE004
(ii) a Decomposing the direct associated node 25 of the target node 26, as shown in fig. 6, to obtain a graph of the target node 26
Figure 586645DEST_PATH_IMAGE004
(ii) a In this embodiment, only the nodes 6, 26 measurable in FIG. 4 are usedTaking the target node as an explanation, and decomposing the target node to obtain a graph
Figure 630824DEST_PATH_IMAGE004
It should be appreciated that in practical applications, the predictability of the target node is unknown;
further, according to the figure
Figure 42826DEST_PATH_IMAGE004
And presetting a prediction autovariable set
Figure 830654DEST_PATH_IMAGE005
Constructing a deduction tree of said target nodes, i.e. from said graph
Figure 263909DEST_PATH_IMAGE004
Logic driving relation between nodes, preset prediction autovariate set for input
Figure 478990DEST_PATH_IMAGE005
The measurable nodes corresponding to the plurality of observable index variables in the graph are screened, and the nodes obtained according to the screening and the graph are obtained
Figure 240273DEST_PATH_IMAGE004
And constructing a deduction tree of the target node to ensure the accuracy of subsequent predictability study and judgment of the target node and further improve the accuracy of data analysis.
In an optional embodiment of the present invention, the step 122 may include:
step 1221, in said figure
Figure 707157DEST_PATH_IMAGE004
In the method, the graph is traversed from the layer where the target node is located in sequence
Figure 870285DEST_PATH_IMAGE004
Each layer of decomposition volume node;
step 1222, if the current resolution node is in the pre-resolution stateSetting a set of predicted autovariables
Figure 256267DEST_PATH_IMAGE005
Adding the current decomposition amount node into the actually used prediction variable set
Figure 895059DEST_PATH_IMAGE006
(ii) a The anchor quantity node is a node corresponding to an index variable which is not influenced by other index variables in the system dynamics model;
1223, based on the set of predicted variables
Figure 24689DEST_PATH_IMAGE006
And topological relation set
Figure 917690DEST_PATH_IMAGE002
And obtaining a current graph, and determining the current graph as a deduction tree of the target node.
In this embodiment, in the figure
Figure 740152DEST_PATH_IMAGE004
In the method, besides the target node, the decomposition quantity nodes of each layer are sequentially traversed, and whether the traversed current decomposition quantity node is in the preset prediction autovariate set or not is determined
Figure 476027DEST_PATH_IMAGE005
That is, determining the intersection of the decomposition quantity node set and the preset prediction autovariate set, and using the node set in the intersection as the prediction variable set for actually predicting the testability of the target node
Figure 533982DEST_PATH_IMAGE006
The preset prediction autovariate set is screened, so that the accuracy of a final prediction result is ensured, and the efficiency of prediction analysis is improved;
meanwhile, when the decomposition quantity node is traversed, when the traversed current decomposition quantity node is an immeasurable anchor quantity node, the node is determined byThe immeasurable anchor quantitative node is not influenced by other nodes, but can also play a decision role on the target node when certain preset conditions are met, so that the graph is used
Figure 406123DEST_PATH_IMAGE004
Adding the intermediate undetectable anchor quantitative node into the prediction variable set
Figure 9274DEST_PATH_IMAGE006
Further, according to the prediction variable set
Figure 498024DEST_PATH_IMAGE006
And sets of topological relationships
Figure 703877DEST_PATH_IMAGE002
Obtaining a deduction tree of the target node;
it should be noted that the deduction tree should be the graph
Figure 86317DEST_PATH_IMAGE004
When said figure is a subset of
Figure 985003DEST_PATH_IMAGE004
When all measurable nodes in (a) are in the preset set of predicted independent variables, and the graph
Figure 695470DEST_PATH_IMAGE004
When the unmeasured nodes in (1) meet certain preset conditions, the graph
Figure 583310DEST_PATH_IMAGE004
I.e. the deduction tree of the target node.
In an optional embodiment of the present invention, the step 1222 may include:
step 12221, according to the graph
Figure 695623DEST_PATH_IMAGE004
Obtaining the downstream node independently influenced by the current decomposition amount node, andthe downstream node in the preset prediction autovariate set
Figure 889844DEST_PATH_IMAGE005
Adding the downstream node into the prediction variable set actually used
Figure 87607DEST_PATH_IMAGE006
Step 12222, according to the graph
Figure 900842DEST_PATH_IMAGE004
Obtaining the upstream node of the current decomposition quantity node, if the upstream node is not in the graph
Figure 743027DEST_PATH_IMAGE004
Adding the upstream node to the graph
Figure 983516DEST_PATH_IMAGE004
。
In this embodiment, when traversing the current resolution node, if the current resolution node is not an anchor quantity node, the upstream of the current resolution node and the downstream node independently affected by the current resolution node may be obtained;
if the downstream node of the current decomposition quantity node is in the preset prediction autovariate set
Figure 793209DEST_PATH_IMAGE005
Adding the downstream node to the set of predicted variables
Figure 144556DEST_PATH_IMAGE006
When the downstream node is in the preset prediction autovariate set
Figure 231460DEST_PATH_IMAGE005
In the above method, it should be determined that the current decomposition quantity node should also be in the preset prediction autovariate set
Figure 518216DEST_PATH_IMAGE005
Performing the following steps; otherwise, the downstream node can be directly deleted;
if the upstream node independently influenced by the current decomposition quantity node is not in the graph
Figure 424992DEST_PATH_IMAGE004
Should the upstream node be added to the graph
Figure 704664DEST_PATH_IMAGE004
In (1), form a new figure
Figure 646075DEST_PATH_IMAGE004
And determining the upstream node as a new graph
Figure 838153DEST_PATH_IMAGE004
In the new node layer, in the new graph
Figure 497805DEST_PATH_IMAGE004
In the method, the new graph is traversed from the layer where the target node is located in sequence
Figure 190954DEST_PATH_IMAGE004
Until no new node layer is generated, the final prediction variable set is obtained
Figure 111506DEST_PATH_IMAGE006
And the accuracy of the final target node studying and judging analysis result is ensured.
In an optional embodiment of the present invention, the step 13 may include:
step 131, traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence;
step 132a, if the lower nodes of the target node are all predictable nodes, determining that the target node is predictable; otherwise, determining the target node to be unpredictable.
In this embodiment, in the deduction tree, the target node is an initial layer of the deduction tree, the resolution node that cannot be decomposed any more is a bottommost layer of the deduction tree, and the resolution nodes of the respective layers in the deduction tree are sequentially traversed upward from the bottommost layer of the deduction tree until the initial layer where the target node is located is traversed; when the lower-layer decomposition amount nodes of the starting layer where the target node is located are all measurable nodes, determining that the target node is predictable, and as shown in fig. 5, when the lower-layer decomposition amount nodes in the deduction tree are sequentially predictable nodes 1 to predictable nodes 7, determining that the target node 6 is necessarily measurable; when all the lower-layer decomposition volume nodes of the starting layer where the target node is located are non-measurable nodes, determining that the target node is non-predictable, and as shown in fig. 5, when the lower-layer decomposition volume nodes in the deduction tree are sequentially non-predictable nodes 12 to non-predictable nodes 20, determining that the target node 6 is not measurable; the lower-layer decomposition quantity nodes are all the lower-layer decomposition quantity nodes of the initial layer where the target node is located;
according to the predictability of the decomposition quantity nodes of the target nodes in the deduction tree and the logic driving relation between the decomposition quantity nodes and the target nodes, the predictability of the target nodes is determined, the accuracy of studying and judging the predictability of the target nodes is guaranteed, accurate analysis of relevant index variables of industrial equipment is achieved, and the efficiency and accuracy of data analysis are further improved.
In an optional embodiment of the present invention, after the step 131, the method may further include:
step 132b, if all the driving nodes of the current lower node of the traversed target node are not all in the preset prediction independent variable set
Figure 130277DEST_PATH_IMAGE005
And determining the current lower node as a result node.
In this embodiment, if all the driving nodes of the current lower node are not all in the preset set of predicted arguments
Figure 884082DEST_PATH_IMAGE005
Of middle, i.e. of said current lower layer resolution nodeIf unpredictable resolution nodes exist in the driving nodes, determining the corresponding lower-layer resolution nodes as result nodes, such as node 20 and node 15 in fig. 5, and node 25 and node 20 in fig. 6;
it should be appreciated that the resulting node is also predictable under certain conditions, for example, in FIG. 6, the resulting node 25 is predictable when the resulting node 25 is driven by only the driver node 10 and the driver node 9.
In an optional embodiment of the present invention, after the step 133, the method may further include:
step 134, if the set of predicted variables is set
Figure 380923DEST_PATH_IMAGE006
Including a set of predetermined predictive arguments
Figure 31347DEST_PATH_IMAGE005
An additional unmeasurable anchor quantity node, where the resulting node is predictable, determining that the target node is partially predictable.
In this embodiment, the set of variables is predicted when the resulting node is predicted, and based on the set of predicted variables
Figure 345653DEST_PATH_IMAGE006
And set of topological relationships
Figure 714318DEST_PATH_IMAGE002
If the undetectable anchoring quantity nodes exist in the deduction tree, determining that the target nodes are partially predictable; specifically, if only nodes 25, 19, 10, 9 exist for the nodes in the deduction tree, the target node 26 is determined to be partially predictable since the resulting node 25 is predictable, but the anchor node 29 is not measurable.
The above method will be described in an embodiment, as shown in fig. 7, the specific steps are as follows:
step 71, acquiring a target node in a system dynamics model of the industrial equipment;
step 72, constructing a deduction tree of the target node according to the topological relation between the plurality of nodes and the unmeasured intermediate quantity nodes and a preset prediction autovariate set;
step 721, in the system dynamics model, according to the topological relation between the plurality of nodes and the unmeasured intermediate nodes, sequentially decomposing the target node and all the associated nodes of the target node to obtain a decomposition amount node set
Figure 624636DEST_PATH_IMAGE007
And set of topological relationships between the resolution nodes
Figure 395146DEST_PATH_IMAGE002
Wherein, in the step (A),
Figure 490141DEST_PATH_IMAGE007
and
Figure 205156DEST_PATH_IMAGE002
composition diagram
Figure 294333DEST_PATH_IMAGE003
；
Step 7211, in said figure
Figure 919350DEST_PATH_IMAGE004
In the method, the graph is traversed from the layer where the target node is located in sequence
Figure 450825DEST_PATH_IMAGE004
Decomposing quantity nodes of each layer are sequentially subjected to decomposition circulation until the decomposing quantity nodes can not be decomposed again;
7212, if the current decomposition variable node is in the preset prediction autovariate set
Figure 653136DEST_PATH_IMAGE005
Adding the current decomposition amount node into the actual used nodeSet of measured variables
Figure 561049DEST_PATH_IMAGE006
(ii) a The anchor quantity node is a node corresponding to an index variable which is not influenced by other index variables in the system dynamics model;
step 72121, according to the graph
Figure 915939DEST_PATH_IMAGE004
Obtaining the downstream node independently influenced by the current decomposition quantity node, wherein the downstream node is in the preset prediction autovariate set
Figure 352736DEST_PATH_IMAGE005
Adding the downstream node into the prediction variable set actually used
Figure 183289DEST_PATH_IMAGE006
；
Step 72122, according to the graph
Figure 222789DEST_PATH_IMAGE004
Obtaining the upstream node of the current decomposition quantity node, if the upstream node is not in the graph
Figure 822398DEST_PATH_IMAGE004
Adding the upstream node to the graph
Figure 571042DEST_PATH_IMAGE004
；
722, according to the prediction variable set
Figure 623312DEST_PATH_IMAGE006
And topological relation set
Figure 607448DEST_PATH_IMAGE002
Obtaining a current graph, and determining the current graph as a deduction tree of the target node;
step 73, determining whether the target node is predictable according to the deduction tree;
step 731, traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence;
in step 732, if all the lower nodes of the target node are predictable nodes, the target node is determined to be predictable, otherwise, the target node is determined to be unpredictable.
In step 733, if all the driving nodes of the current lower node of the traversed target node are not all in the preset prediction autovariate set
Figure 186197DEST_PATH_IMAGE005
Determining the current lower node as a result node, adding the current lower node into a result node set, emptying the current result node set, and traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence until no new result node is generated;
step 7331, if the set of predicted variables
Figure 230376DEST_PATH_IMAGE006
Including a set of predetermined predictive arguments
Figure 642379DEST_PATH_IMAGE005
An additional unmeasurable anchor quantity node, where the resulting node is predictable, determining that the target node is partially predictable.
In the embodiment of the invention, the independent variables in the preset prediction independent variable set are screened through the nodes which are associated with the target nodes in the system dynamics model, the prediction variable set which is actually needed to be used is determined, the interference caused by unnecessary independent variables in the preset prediction independent variable set is avoided, and the accuracy of the subsequent target node predictability analysis is ensured; and further determining a deduction tree of a target node according to a topological relation and a prediction autovariate set in a system dynamic model, determining the predictability of the target node according to the predictability of the decomposition quantity node of the target node in the deduction tree and the logic driving relation between the decomposition quantity node and the target node, ensuring the accuracy of studying and judging the predictability of the target node, and realizing accurate analysis on related index variables of industrial equipment, thereby improving the accuracy and efficiency of data analysis.
As shown in fig. 8, an embodiment of the present invention further provides a variable prediction apparatus 80 of a system dynamics model of an industrial plant, including:
an obtaining module 81, configured to obtain a target node in a system dynamics model of an industrial device; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
a processing module 82, configured to construct a deduction tree of the target node according to the topological relations between the measurable nodes and the non-measurable nodes and a preset prediction autovariate set; determining whether the target node is predictable according to the deduction tree, wherein the preset prediction autovariate set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
Optionally, constructing a deduction tree of the target node according to the topological relations between the measurable nodes and the non-measurable nodes and a preset prediction autovariate set, where the deduction tree includes:
in the system dynamics model, according to the topological relations between the plurality of nodes and the unmeasured nodes, the target node and all the associated nodes of the target node are sequentially decomposed to obtain a decomposition amount node set
Figure 164627DEST_PATH_IMAGE007
And set of topological relationships between the resolution nodes
Figure 738828DEST_PATH_IMAGE002
Wherein, in the step (A),
Figure 812963DEST_PATH_IMAGE007
and
Figure 574246DEST_PATH_IMAGE002
composition diagram
Figure 41130DEST_PATH_IMAGE003
；
According to the diagram
Figure 204258DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 590240DEST_PATH_IMAGE005
And constructing a deduction tree of the target node.
Optionally, according to said figure
Figure 229032DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 93083DEST_PATH_IMAGE005
And constructing a deduction tree of the target node, including:
in the figure
Figure 251663DEST_PATH_IMAGE004
In the method, the graph is traversed from the layer where the target node is located in sequence
Figure 808546DEST_PATH_IMAGE004
Each layer of decomposition volume node;
if the current decomposition value node is in the preset prediction autovariate set
Figure 810000DEST_PATH_IMAGE005
Adding the current decomposition amount node into the actually used prediction variable set
Figure 336796DEST_PATH_IMAGE006
(ii) a The anchor quantity node is not in the system dynamics modelThe node corresponding to the index variable influenced by other index variables;
according to the prediction variable set
Figure 474517DEST_PATH_IMAGE006
And topological relation set
Figure 343247DEST_PATH_IMAGE002
And obtaining a current graph, and determining the current graph as a deduction tree of the target node.
Optionally, the processing module 82 is further configured to:
according to the diagram
Figure 300838DEST_PATH_IMAGE004
Obtaining the downstream nodes independently influenced by the current decomposition quantity node, wherein the downstream nodes are in the preset prediction autovariate set
Figure 37850DEST_PATH_IMAGE005
Adding the downstream node into the set of actually used predicted variables
Figure 154711DEST_PATH_IMAGE006
；
According to the diagram
Figure 53397DEST_PATH_IMAGE004
Obtaining the upstream node of the current decomposition quantity node, if the upstream node is not in the graph
Figure 642160DEST_PATH_IMAGE004
Adding the upstream node to the graph
Figure 651704DEST_PATH_IMAGE004
。
Optionally, determining whether the target node is predictable according to the deduction tree includes:
traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence;
and if all lower-layer nodes of the target node are predictable nodes, determining the target node to be predictable, otherwise, determining the target node to be unpredictable.
Optionally, the processing module 82 is further configured to:
if all the driving nodes of the current lower layer node of the traversed target node are not all in the preset prediction autovariate set
Figure 764016DEST_PATH_IMAGE005
And determining the current lower node as a result node.
Optionally, the processing module 82 is further configured to:
if the set of predicted variables is
Figure 958237DEST_PATH_IMAGE006
Including a set of predetermined predictive arguments
Figure 156000DEST_PATH_IMAGE005
An additional unmeasurable anchor quantity node, where the resulting node is predictable, determining that the target node is partially predictable.
It should be noted that the apparatus is an apparatus corresponding to the above method, and all the implementations in the above method embodiment are applicable to the embodiment of the apparatus, and the same technical effects can be achieved.
Embodiments of the present invention also provide a computing device, comprising: a processor, a memory storing a computer program which, when executed by the processor, performs the method as described above. All the implementation manners in the above method embodiment are applicable to this embodiment, and the same technical effect can be achieved.
Embodiments of the present invention also provide a computer-readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the method as described above. All the implementation manners in the above method embodiment are applicable to this embodiment, and the same technical effect can be achieved.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
Furthermore, it is to be noted that in the device and method of the invention, it is obvious that the individual components or steps can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of performing the series of processes described above may naturally be performed chronologically in the order described, but need not necessarily be performed chronologically, and some steps may be performed in parallel or independently of each other. It will be understood by those skilled in the art that all or any of the steps or elements of the method and apparatus of the present invention may be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or any combination thereof, which can be implemented by those skilled in the art using their basic programming skills after reading the description of the present invention.
Thus, the objects of the invention may also be achieved by running a program or a set of programs on any computing device. The computing device may be a general purpose device as is well known. The object of the invention is thus also achieved solely by providing a program product comprising program code for implementing the method or the apparatus. That is, such a program product also constitutes the present invention, and a storage medium storing such a program product also constitutes the present invention. It is to be understood that the storage medium may be any known storage medium or any storage medium developed in the future. It is also noted that in the apparatus and method of the present invention, it is apparent that each component or step can be decomposed and/or recombined. These decompositions and/or recombinations are to be regarded as equivalents of the present invention. Also, the steps of executing the series of processes described above may naturally be executed chronologically in the order described, but need not necessarily be executed chronologically. Some steps may be performed in parallel or independently of each other.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (10)

1. A method for predicting variables of a system dynamics model of an industrial plant, comprising:
acquiring a target node in a system dynamics model of industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
constructing a deduction tree of the target node according to the topological relation between the plurality of measurable nodes and the non-measurable nodes and a preset prediction autovariate set;
determining whether the target node is predictable according to the deduction tree, wherein the preset prediction autovariate set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
2. The method of predicting variables of a system dynamics model of an industrial plant according to claim 1, wherein constructing a deduction tree of the target node based on topological relationships between the plurality of measurable nodes and non-measurable nodes and a set of preset predicted independent variables comprises:
in the system dynamics model, according to the topological relations between the plurality of nodes and the unmeasured nodes, the target node and all the associated nodes of the target node are sequentially decomposed to obtain a decomposition amount node set
Figure 800515DEST_PATH_IMAGE001
And set of topological relationships between the resolution nodes
Figure 364351DEST_PATH_IMAGE002
Wherein, in the step (A),
Figure 420645DEST_PATH_IMAGE001
and
Figure 728129DEST_PATH_IMAGE002
composition diagram
Figure 759670DEST_PATH_IMAGE003
；
According to the diagram
Figure 556725DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 103244DEST_PATH_IMAGE005
And constructing a deduction tree of the target node.
3. The method of predicting variables for a system dynamics model of an industrial plant of claim 2, characterized in that it is based on the graph
Figure 683261DEST_PATH_IMAGE004
And presetting a set of predicted autovariables
Figure 897205DEST_PATH_IMAGE005
And constructing a deduction tree of the target node, including:
in the figure
Figure 865161DEST_PATH_IMAGE004
In the method, the graph is traversed from the layer where the target node is located in sequence
Figure 898976DEST_PATH_IMAGE004
Each layer of decomposition volume node;
if the current decomposition value node is in the preset prediction autovariate set
Figure 610580DEST_PATH_IMAGE005
Adding the current decomposition amount node into the actually used prediction variable set when the medium or current decomposition amount node is an immeasurable anchor quantitative node
Figure 681960DEST_PATH_IMAGE006
(ii) a The anchor quantity node is a node corresponding to an index variable which is not influenced by other index variables in the system dynamics model;
according to the prediction variable set
Figure 820817DEST_PATH_IMAGE006
And topological relation set
Figure 138666DEST_PATH_IMAGE002
And obtaining a current graph, and determining the current graph as a deduction tree of the target node.
4. The method of predicting variables for a system dynamics model of an industrial plant of claim 3, further comprising:
according to the diagram
Figure 591644DEST_PATH_IMAGE004
Obtaining a downstream node independently affected by the current resolution node, andthe downstream node predicts the independent variable set in the preset prediction
Figure 576917DEST_PATH_IMAGE005
Adding the downstream node into the prediction variable set actually used
Figure 558780DEST_PATH_IMAGE006
；
According to the diagram
Figure 363925DEST_PATH_IMAGE004
Obtaining the upstream node of the current decomposition quantity node, if the upstream node is not in the graph
Figure 417331DEST_PATH_IMAGE004
Adding the upstream node to the graph
Figure 460374DEST_PATH_IMAGE004
。
5. The method of predicting variables of a system dynamics model of an industrial plant of claim 4, wherein determining whether the target node is predictable from the deduction tree comprises:
traversing each layer of nodes of the deduction tree from the bottommost layer of the deduction tree to the upper layer in sequence;
and if all lower-layer nodes of the target node are predictable nodes, determining the target node to be predictable, otherwise, determining the target node to be unpredictable.
6. The method of predicting variables for a system dynamics model of an industrial plant of claim 5, further comprising:
if all the drive nodes of the current lower-layer node of the traversed target node are not in the preset prediction autovariate set
Figure 941034DEST_PATH_IMAGE005
And determining the current lower node as a result node.
7. The method of predicting variables for a system dynamics model of an industrial plant of claim 6, further comprising:
if the set of predicted variables is
Figure 436737DEST_PATH_IMAGE006
Including a set of predetermined predictive arguments
Figure 28255DEST_PATH_IMAGE005
An additional unmeasurable anchor quantity node, where the resulting node is predictable, determining that the target node is partially predictable.
8. A variable prediction apparatus of a system dynamics model of an industrial plant, comprising:
the acquisition module is used for acquiring a target node in a system dynamics model of the industrial equipment; the system dynamics model includes: describing a plurality of measurable nodes and non-measurable nodes corresponding to a plurality of index variables of the industrial equipment and topological relations between the measurable nodes and the non-measurable nodes; wherein, the observable index variable corresponds to the measurable node, and the non-observable index variable corresponds to the non-measurable node;
the processing module is used for constructing a deduction tree of the target node according to the topological relation between the measurable nodes and the non-measurable nodes and a preset prediction autovariate set; determining whether the target node is predictable according to the deduction tree, wherein the preset prediction autovariate set comprises: and a plurality of measurable nodes corresponding to a plurality of observable index variables in the system dynamics model.
9. A computing device, comprising: a processor, a memory storing a computer program which, when executed by the processor, performs the method of any of claims 1 to 7.
10. A computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method of any one of claims 1 to 7.
CN202210978259.8A 2022-08-16 2022-08-16 Variable prediction method, device and equipment of system dynamics model of industrial equipment Active CN115048826B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210978259.8A CN115048826B (en) 2022-08-16 2022-08-16 Variable prediction method, device and equipment of system dynamics model of industrial equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210978259.8A CN115048826B (en) 2022-08-16 2022-08-16 Variable prediction method, device and equipment of system dynamics model of industrial equipment

Publications (2)

Publication Number Publication Date
CN115048826A true CN115048826A (en) 2022-09-13
CN115048826B CN115048826B (en) 2022-10-21

Family

ID=83168121

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210978259.8A Active CN115048826B (en) 2022-08-16 2022-08-16 Variable prediction method, device and equipment of system dynamics model of industrial equipment

Country Status (1)

Country Link
CN (1) CN115048826B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109478055A (en) * 2016-03-01 2019-03-15 源讯网源公司 Using intelligent node for monitoring industrial process in general, intelligence system
CN110084326A (en) * 2019-05-13 2019-08-02 东北大学 A kind of industrial equipment method for detecting abnormality based on fuzzy set
CN112187632A (en) * 2020-09-23 2021-01-05 大唐广电科技（武汉）有限公司 Industrial equipment integration model and construction method thereof, edge gateway device and industrial equipment integration system
US20210119882A1 (en) * 2020-04-13 2021-04-22 Intel Corporation Methods and apparatus to determine topologies for networks
US20220066415A1 (en) * 2020-09-03 2022-03-03 Rockwell Automation Technologies, Inc. Industrial automation asset and control project analysis

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109478055A (en) * 2016-03-01 2019-03-15 源讯网源公司 Using intelligent node for monitoring industrial process in general, intelligence system
CN110084326A (en) * 2019-05-13 2019-08-02 东北大学 A kind of industrial equipment method for detecting abnormality based on fuzzy set
US20210119882A1 (en) * 2020-04-13 2021-04-22 Intel Corporation Methods and apparatus to determine topologies for networks
US20220066415A1 (en) * 2020-09-03 2022-03-03 Rockwell Automation Technologies, Inc. Industrial automation asset and control project analysis
CN112187632A (en) * 2020-09-23 2021-01-05 大唐广电科技（武汉）有限公司 Industrial equipment integration model and construction method thereof, edge gateway device and industrial equipment integration system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
栾燕等: "工业网关分类及关键指标分析", 《自动化仪表》 *

Also Published As

Publication number Publication date
CN115048826B (en) 2022-10-21

Similar Documents

Publication Publication Date Title
Schlegel et al. Towards a rigorous evaluation of XAI methods on time series
Bohlin Practical grey-box process identification: theory and applications
Liu et al. Wind turbine fault detection based on deep residual networks
Ayodeji et al. Causal augmented ConvNet: A temporal memory dilated convolution model for long-sequence time series prediction
JP5338492B2 (en) Input variable selection support device
Conrad et al. Benchmarking AutoML for regression tasks on small tabular data in materials design
Langary et al. Inference of chemical reaction networks based on concentration profiles using an optimization framework
Tan et al. Deviation contribution plots of multivariate statistics
Li et al. A review on physics-informed data-driven remaining useful life prediction: Challenges and opportunities
Nasrabadi et al. Robustness of efficiency scores in data envelopment analysis with interval scale data
CN117217419B (en) Method and system for monitoring full life cycle carbon emission of industrial production
Elhishi et al. Unboxing machine learning models for concrete strength prediction using XAI
CN115048826B (en) Variable prediction method, device and equipment of system dynamics model of industrial equipment
Pevec et al. Prediction intervals in supervised learning for model evaluation and discrimination
Mohamad et al. Modelling the financial performance of construction companies using neural network via genetic algorithm
CN116720946A (en) Credit risk prediction method, device and storage medium based on recurrent neural network
Wederhake et al. On the surplus accuracy of data-driven energy quantification methods in the residential sector
Haghighatkhah Test case prioritization using build history and test distances: an approach for improving automotive regression testing in continuous integration environments
Sakhrawi et al. Investigating the impact of functional size measurement on predicting software enhancement effort using correlation-based feature selection algorithm and SVR method
CN115130805A (en) Analysis device, analysis method, and computer-readable medium having program recorded thereon
Servadei et al. Cost optimization at early stages of design using deep reinforcement learning
US8010477B2 (en) Integrated problem solving system
Rojek et al. Chemical Mixing Simulations with Integrated AI Accelerator
Kunz et al. Cross-impact balance as an approach for the development of consistent storylines for the European energy market
Benchekroun et al. Development of a kiln petcoke mill predictive model based on a multi-regression XGBoost algorithm

Legal Events

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