An automated sealing method and sealing device

Abstract

The application provides an automatic sealing method and a sealing device, and relates to the technical field of automatic sealing. The method comprises the following steps: obtaining Q sealing process nodes; obtaining a sealing control feature distribution model; obtaining sealing scene information; obtaining a sealing control optimization space; obtaining a first sealing control particle; performing sealing fitness prediction on the first sealing control particle to obtain a first particle sealing fitness; determining whether the first particle sealing fitness meets a preset sealing fitness constraint; if the first particle sealing fitness meets the preset sealing fitness constraint, outputting the first sealing control particle as a sealing scheme, and performing automatic sealing according to the sealing scheme. The technical problem that the sealing quality is poor due to inaccurate sealing effect prediction of sealing control parameters and improper selection of sealing control parameters in the prior art is solved, and the technical effects of improving the sealing control accuracy and the sealing quality are achieved.

Description

Technical Field

[0001] This invention relates to the field of automatic sealing technology, specifically to an automated sealing method and sealing device. Background Technology

[0002] With the development of automated drilling technology and equipment in coal mines, the demand for automated construction of the entire gas drainage hole drilling process in coal mines is becoming increasingly urgent. Sealing methods and devices are a crucial aspect of gas drainage hole construction, affecting the realization of automated sealing and the quality of sealing. Existing technologies suffer from technical problems such as inaccurate prediction of sealing effects based on sealing control parameters and improper selection of these parameters, leading to poor sealing quality. Summary of the Invention

[0003] This invention provides an automated sealing method and sealing device to solve the technical problem in the prior art where poor sealing quality is caused by inaccurate prediction of sealing effect of sealing control parameters and improper selection of sealing control parameters.

[0004] According to a first aspect of the present invention, an automated hole sealing method is provided, comprising: obtaining hole sealing process flow information, and dividing nodes based on the hole sealing process flow information to obtain Q hole sealing process nodes, wherein Q is a positive integer greater than 1; performing hole sealing control feature mining based on the Q hole sealing process nodes to obtain a hole sealing control feature distribution model; obtaining hole sealing scene information; performing control parameter mining based on the hole sealing scene information and the hole sealing control feature distribution model to obtain a hole sealing control optimization space; obtaining a first hole sealing control particle based on the hole sealing control optimization space; performing hole sealing fit prediction on the first hole sealing control particle based on a hole sealing fit optimization function to obtain a first particle hole sealing fit; determining whether the first particle hole sealing fit satisfies a preset hole sealing fit constraint; if the first particle hole sealing fit satisfies the preset hole sealing fit constraint, outputting the first hole sealing control particle as a hole sealing scheme, and performing automated hole sealing according to the hole sealing scheme.

[0005] According to a second aspect of the present invention, an automated sealing device is provided, comprising: a process flow division module, wherein the process flow division module is used to obtain sealing process flow information and perform node division based on the sealing process flow information to obtain Q sealing process nodes, wherein Q is a positive integer greater than 1; a sealing control feature mining module, wherein the sealing control feature mining module is used to perform sealing control feature mining based on the Q sealing process nodes to obtain a sealing control feature distribution model; a sealing scene information acquisition module, wherein the sealing scene information acquisition module is used to obtain sealing scene information; and a control parameter mining module, wherein the control parameter mining module is used to perform control parameter mining based on the sealing scene information and according to the sealing control feature distribution model. The system includes: a hole sealing control optimization space; a hole sealing control optimization module, which obtains a first hole sealing control particle based on the hole sealing control optimization space; a hole sealing fit prediction module, which predicts the hole sealing fit of the first hole sealing control particle based on the hole sealing fit optimization function to obtain the hole sealing fit of the first particle; a fit judgment module, which judges whether the hole sealing fit of the first particle meets the preset hole sealing fit constraint; and an automated hole sealing module, which outputs the first hole sealing control particle as a hole sealing scheme if the hole sealing fit of the first particle meets the preset hole sealing fit constraint, and performs automated hole sealing according to the hole sealing scheme.

[0006] The beneficial effects that can be achieved by adopting one or more technical solutions according to the present invention are as follows:

[0007] The process involves obtaining sealing process flow information and dividing the process into nodes, resulting in Q sealing process nodes, where Q is a positive integer greater than 1. Sealing control features are mined from these Q nodes to obtain a sealing control feature distribution model and sealing scenario information. Based on this scenario information and the sealing control feature distribution model, control parameters are mined to obtain a sealing control optimization space. A first sealing control particle is then obtained from this optimization space. The sealing fit degree of the first sealing control particle is predicted using a sealing fit degree optimization function to obtain its sealing fit degree. It is then determined whether the sealing fit degree of the first particle meets a preset sealing fit degree constraint. If it does, the first sealing control particle is output as a sealing scheme, and automated sealing is performed according to this scheme. Therefore, by optimizing the sealing control parameters of the Q sealing process nodes, the accuracy of sealing control is improved, thereby enhancing the sealing quality. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in this invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. The accompanying drawings, which constitute a part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating an automated sealing method provided in an embodiment of the present invention.

[0010] Figure 2 This is a schematic diagram of an automated sealing device provided in an embodiment of the present invention.

[0011] Explanation of reference numerals in the attached diagram: 11. Process flow division module; 12. Sealing control feature mining module; 13. Sealing scene information acquisition module; 14. Control parameter mining module; 15. Sealing control optimization module; 16. Sealing fit prediction module; 17. Fit judgment module; 18. Automated sealing module. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0013] The terminology used in this specification is for describing embodiments and not for limiting the invention. As used in this specification, the singular terms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. When used in this specification, the terms “comprising” and / or “including” specify the presence of a step, operation, element, and / or component, but do not preclude the presence or addition of one or more other steps, operations, elements, components, and / or groups thereof.

[0014] Unless otherwise defined, all terms (including technical and scientific terms) used in this specification shall have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms, as defined in common dictionaries, should not be interpreted in an idealized or overly formal sense unless expressly defined herein. Throughout this specification, the same reference numerals denote the same elements.

[0015] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this invention are all information and data authorized by the user or fully authorized by all parties.

[0016] Example 1

[0017] Figure 1 A diagram illustrating an automated hole-sealing method provided in an embodiment of the present invention, the method comprising:

[0018] Obtain the sealing process flow information, and divide the nodes based on the sealing process flow information to obtain Q sealing process nodes, where Q is a positive integer greater than 1;

[0019] The sealing process flow information refers to the sequence of sealing processes, such as grasping the lower bag section, the intermediate standard section, the upper bag section, and grouting. It should be noted that this sealing process flow information refers to the current sealing process flow; therefore, users need to determine this based on actual conditions and upload it through their client. Based on this sealing process flow information, nodes are divided. That is, the sealing process flow information includes multiple sequential process operations. Dividing these operations according to their sequence yields Q sealing process nodes, where Q is a positive integer greater than 1.

[0020] Based on the Q sealing process nodes, sealing control feature mining is performed to obtain a sealing control feature distribution model;

[0021] In a preferred embodiment, it further includes:

[0022] The Q sealing process nodes are sorted according to the sealing process sequence to obtain the first sealing process node, the second sealing process node, ..., the qth sealing process node, where q is a positive integer and q belongs to Q. Sealing control feature mining is performed on the first sealing process node to obtain the first node sealing control feature distribution. Sealing control feature mining is then performed on the second sealing process node...the qth sealing process node to obtain the second node sealing control feature distribution...the qth node sealing control feature distribution. The sealing control feature distributions of the first node, the second node, and the qth node are integrated to generate the sealing control feature distribution model.

[0023] In a preferred embodiment, it further includes:

[0024] The sealing process node is recorded with sealing control data to obtain multiple sealing control records for the first node. Sealing control indicators are extracted from these records to obtain multiple sealing control indicators for the first node. Support is calculated for each of the sealing control indicators based on the records to obtain multiple indicator support scores. Confidence is calculated based on the support scores to obtain multiple indicator confidence scores. Based on preset confidence constraints, the sealing control indicators are filtered to obtain multiple sealing control confidence indicators that satisfy the preset confidence constraints. These confidence indicators are then added to the first node sealing control feature distribution.

[0025] Based on the Q sealing process nodes, sealing control feature mining is performed to obtain a sealing control feature distribution model. This model includes the sealing features of the Q sealing process nodes. The specific acquisition process is as follows:

[0026] The Q sealing process nodes are sorted according to the sealing process sequence to obtain the first sealing process node, the second sealing process node, ..., the qth sealing process node, where q is a positive integer and q belongs to Q. The sealing process sequence needs to be determined based on the actual sealing process used. Further, sealing control features are mined based on the first sealing process node. Sealing control features refer to the control characteristics of the sealing process performed by the first sealing process node, such as the number of intermediate standard short sections added. This is used as the sealing control feature distribution of the first node. Similarly, sealing control features are mined for the second sealing process node...the qth sealing process node using the same method to obtain the sealing control feature distribution of the second node...the qth node. The sealing control feature distributions of the first node, the second node...the qth node are combined to form the sealing control feature distribution model. This achieves the extraction of control features for each packaging process node, thereby improving the accuracy of sealing.

[0027] The process of mining sealing control features based on the first sealing process node to obtain the distribution of sealing control features of the first node is as follows: Sealing control records are collected for the first sealing process node, resulting in multiple first node sealing control records. These multiple first node sealing control records can be understood as historical sealing control record data. Simply put, sealing control record data for the first sealing process node over a past period (e.g., the past month) can be retrieved, such as records of the composition of the grouting liquid, the grouting volume, and the grouting speed during grouting. Further, sealing control indicators are extracted based on these multiple first node sealing control records. Sealing control indicators refer to the control parameters used during sealing control, such as the grouting volume and grouting speed. This yields multiple first node sealing control indicators, which broadly refer to the control parameters used during sealing control. Based on these multiple first node sealing control records, support is calculated for these multiple first node sealing control indicators. Specifically, the frequency of occurrence of different sealing control indicators among the multiple first node sealing control indicators is used as the support of multiple indicators. The confidence level is calculated based on the support of the multiple indicators. The confidence level is the ratio of the support to the total number of all first-node sealing control indicators. That is, the total number of first-node sealing control indicators is counted, and the multiple support values ​​are divided by the total number. The result is the confidence level of multiple indicators.

[0028] Further, preset indicator confidence constraints are set. These preset indicator confidence constraints are set by those skilled in the art based on historical experience. The preset indicator confidence constraints are compared with multiple indicator confidences to obtain multiple first-node sealing control confidence indicators whose indicator confidences satisfy the preset indicator confidence constraints. These indicators are then added to the first-node sealing control feature distribution, thereby realizing the analysis and acquisition of the first-node sealing control feature distribution, providing a foundation for subsequent automatic sealing, and thus improving sealing accuracy.

[0029] Obtain information about the sealing scene;

[0030] Based on the sealing scenario information, control parameters are mined according to the sealing control feature distribution model to obtain the sealing control optimization space.

[0031] Obtain sealing scene information, which refers to the actual sealing status, including sealing type, size, etc., which needs to be determined based on the actual situation.

[0032] Based on the sealing scenario information, control parameters are mined according to the sealing control feature distribution model to obtain the sealing control optimization space. The sealing control optimization space includes the optimization space of the sealing control parameters corresponding to Q sealing process nodes. The specific acquisition process is detailed below.

[0033] In a preferred embodiment, it further includes:

[0034] Based on the sealing scene information, the control domain of the sealing control feature distribution of the first node is calibrated to obtain the first sealing control domain; based on the sealing scene information, the control domain of the sealing control feature distribution of the second node...the sealing control feature distribution of the qth node is further calibrated to obtain the second sealing control domain...the qth sealing control domain; the first sealing control domain, the second sealing control domain...the qth sealing control domain are integrated to obtain the sealing scene control domain; based on the sealing scene control domain, multiple sealing control particles are obtained by random selection, and the multiple sealing control particles are added to the sealing control optimization space.

[0035] In a preferred embodiment, it further includes:

[0036] Using the sealing scene information and the first node sealing control feature distribution as retrieval constraints, and based on the retrieval constraints, a sealing control parameter retrieval is performed to obtain a first scene-sealing control record library; based on the first scene-sealing control record library, the control parameter extrema of the first node sealing control feature distribution are calibrated to obtain the first node control feature extrema calibration result; based on the first node control feature extrema calibration result, the first sealing control domain is generated.

[0037] Based on the sealing scenario information, the control domain of the sealing control feature distribution of the first node is calibrated to obtain the first sealing control domain. Simply put, the first sealing control domain refers to the value range of the control parameters of the first sealing process node, and it conforms to the sealing scenario information and the sealing control feature distribution of the first node, as detailed below:

[0038] Using the sealing scene information and the first node sealing control feature distribution as retrieval constraints, and based on these constraints, sealing control parameters are retrieved. Specifically, sealing control records corresponding to the scenes described in the sealing scene information and the first node sealing control feature distribution are extracted from historical sealing control records over a period of time to form a first scene-sealing control record library. This first scene-sealing control record library can be understood as a set of standard control data extracted from historical data. Further, based on this first scene-sealing control record library, the maximum and minimum values ​​of the control parameters in the first node sealing control feature distribution are calibrated. These values ​​are used as the calibration results for the first node control feature maximum and minimum values. Based on these calibration results, the first sealing control domain is generated. The first sealing control domain is the range of control parameters formed by the maximum and minimum values ​​of the control parameters in the calibration results. This achieves the analysis and acquisition of the sealing control domain, providing support for subsequent sealing, facilitating the optimization of sealing control parameters, and improving sealing performance.

[0039] Similarly, based on the sealing scene information, using the same method as obtaining the first sealing control domain, the control domain calibration is performed on the second node sealing control feature distribution...the qth node sealing control feature distribution to obtain the second sealing control domain...the qth sealing control domain. The first sealing control domain, the second sealing control domain...the qth sealing control domain are then combined as the sealing scene control domain. Further, random values ​​are taken from the sealing scene control domain, that is, multiple random values ​​are taken from the first sealing control domain, the second sealing control domain...the qth sealing control domain, to obtain multiple sealing control particles. These multiple sealing control particles must be within the control parameter range of the sealing scene control domain. These multiple sealing control particles are added to the sealing control optimization space, thereby obtaining the sealing control optimization space. Subsequent control optimization can then be performed within this space to improve the accuracy and effectiveness of the sealing control.

[0040] Based on the hole sealing control optimization space, the first hole sealing control particle is obtained;

[0041] Based on the hole sealing fit optimization function, the hole sealing fit of the first hole control particle is predicted to obtain the hole sealing fit of the first particle.

[0042] The first sealing control particle refers to any sealing control particle randomly selected in the sealing control optimization space.

[0043] The sealing fit degree of the first sealing control particle is predicted based on the sealing fit degree optimization function. The sealing fit degree optimization function is an expression used to evaluate and predict the sealing fit degree. The sealing fit degree of the first particle is calculated by the sealing fit degree optimization function to facilitate sealing optimization. The specific process is detailed below.

[0044] In a preferred embodiment, it further includes:

[0045] Multi-dimensional sealing quality prediction is performed on the first sealing control particle to obtain the sealing quality prediction result of the first particle, wherein the sealing quality prediction result of the first particle includes the first predicted sealing density and the first predicted sealing stability.

[0046] Construct the hole fitting degree optimization function, wherein the hole fitting degree optimization function is:

[0047] SHT = HDY F +HSY G ;

[0048]

[0049]

[0050] Among them, SHT represents the particle sealing fit, HDY represents the predicted sealing density, HSY represents the predicted sealing stability, F represents the sealing density-exponent, G represents the sealing stability-exponent, HDY0 represents the preset sealing density, and HSY0 represents the preset sealing stability.

[0051] The prediction result of the first particle sealing quality is input into the sealing fit optimization function to generate the sealing fit of the first particle.

[0052] In a preferred embodiment, it further includes:

[0053] Obtain sealing scene information records, sealing scheme records, sealing density records, and sealing stability records; use the sealing scene information records and sealing scheme records as input data, and the sealing density records as output supervision data to train a BP neural network to obtain a sealing density prediction channel that satisfies convergence constraints; use the sealing scene information records and sealing scheme records as input data, and the sealing stability records as output supervision data to train a BP neural network to obtain a sealing stability prediction channel that satisfies the convergence constraints; connect the sealing density prediction channel and the sealing stability prediction channel to generate a sealing quality prediction submodule, and perform multi-dimensional sealing quality prediction of the first sealing control particle based on the sealing quality prediction submodule.

[0054] Multi-dimensional sealing quality prediction is performed on the first sealing control particle to obtain the first particle sealing quality prediction result. This result includes a first predicted sealing density and a first predicted sealing stability. Sealing density refers to the sealing performance of the sealing hole, while sealing stability refers to the degree of stability of the mass after sealing, i.e., the degree of mass change. The smaller the degree of mass change, the higher the sealing stability. Sealing density and sealing stability are important reference indicators for sealing quality testing. The specific process for obtaining the first particle sealing quality prediction result is as follows:

[0055] The system obtains sealing scene information records, sealing scheme records, sealing density records, and sealing stability records. These records are all historical data and have a one-to-one correspondence. The sealing scene information records and sealing scheme records refer to historical sealing scenes and their corresponding sealing schemes. The sealing scheme records include sealing control parameters, i.e., historical sealing control particles. The sealing density records and sealing stability records refer to the sealing quality inspection records corresponding to the sealing scene information records and sealing scheme records. Further, using the sealing scene information record and the sealing scheme record as input data, and the sealing density record as output supervision data, a BP neural network is trained to obtain a sealing density prediction channel that satisfies the convergence constraint. Simply put, a BP neural network is used to construct the network structure of the sealing density prediction channel. The sealing scene information record and the sealing scheme record are input into the sealing density prediction channel, and the sealing density record is used for supervision and adjustment to ensure that the output result is consistent with the sealing density record. After training all data, an accuracy test is performed to obtain a sealing density prediction channel whose accuracy satisfies the convergence constraint. Similarly, using the sealing scene information record and the sealing scheme record as input data, and the sealing stability record as output supervision data, a BP neural network is trained to obtain a sealing stability prediction channel that satisfies the convergence constraint. The method for obtaining the sealing stability prediction channel is the same as that for the sealing density prediction channel, and therefore will not be elaborated here. The sealing density prediction channel and the sealing stability prediction channel are connected as a sealing quality prediction submodule. Based on the sealing quality prediction submodule, multi-dimensional sealing quality prediction of the first sealing control particle is performed. This involves inputting the first sealing control particle and the sealing scene information into the sealing quality prediction submodule, and then performing sealing quality prediction through the sealing density prediction channel and the sealing stability prediction channel within the sealing quality prediction submodule. The submodule outputs a first predicted sealing density and a first predicted sealing stability. This achieves sealing quality prediction, facilitating the subsequent selection of the sealing control particle with the best sealing quality, thereby improving the overall sealing quality.

[0056] The sealing fit optimization function is further constructed, wherein the sealing fit optimization function is:

[0057] SHT = HDY F +HSY G ;

[0058]

[0059]

[0060] Among them, SHT represents the particle sealing fit, HDY represents the predicted sealing density, HSY represents the predicted sealing stability, F represents the sealing density exponent, G represents the sealing stability exponent, HDY0 represents the preset sealing density, and HSY0 represents the preset sealing stability. The preset sealing density and preset sealing stability need to be determined in combination with the actual situation. They refer to the sealing density and sealing stability when the sealing quality is qualified. Generally, when carrying out the sealing process, relevant professional technicians will give the qualified standards, which can be determined by the user.

[0061] By inputting the first predicted sealing density and the first predicted sealing stability from the first particle sealing quality prediction result into the sealing fit optimization function, the first particle sealing fit can be output.

[0062] This enables the analysis of the sealing fit. The greater the sealing fit of the first particle, the better the sealing effect of the first sealing control particle. This facilitates the optimization of sealing control and improves the sealing quality.

[0063] Determine whether the sealing fit of the first particle meets the preset sealing fit constraint.

[0064] If the sealing fit of the first particle meets the preset sealing fit constraint, the first sealing control particle is output as a sealing scheme, and automated sealing is performed according to the sealing scheme.

[0065] Determine whether the sealing fit of the first particle meets the preset sealing fit constraint. The preset sealing fit constraint refers to the sealing fit under the condition that the sealing is qualified, and can be set by those skilled in the art in combination with the actual situation.

[0066] If the sealing fit of the first particle satisfies the preset sealing fit constraint, the first sealing control particle is output as a sealing scheme, and automated sealing is performed according to the sealing scheme to ensure that the sealing quality meets the requirements. Conversely, if the sealing fit of the first particle does not satisfy the preset sealing fit constraint, a second sealing control particle other than the first sealing control particle is randomly selected from the sealing control optimization space, and the sealing fit is calculated to obtain the sealing fit of the second particle. It is then further determined whether the sealing fit of the second particle satisfies the preset sealing fit constraint. If it does, the second sealing control particle is used as the sealing scheme. If it does not, iterative optimization continues in the sealing control optimization space until the preset sealing fit constraint is satisfied.

[0067] Based on the above analysis, the beneficial effects that one or more technical solutions provided by the present invention can achieve are as follows:

[0068] The process involves obtaining sealing process flow information and dividing the process into nodes, resulting in Q sealing process nodes, where Q is a positive integer greater than 1. Sealing control features are mined from these Q nodes to obtain a sealing control feature distribution model and sealing scenario information. Based on this scenario information and the sealing control feature distribution model, control parameters are mined to obtain a sealing control optimization space. A first sealing control particle is then obtained from this optimization space. The sealing fit degree of the first sealing control particle is predicted using a sealing fit degree optimization function to obtain its sealing fit degree. It is then determined whether the sealing fit degree of the first particle meets a preset sealing fit degree constraint. If it does, the first sealing control particle is output as a sealing scheme, and automated sealing is performed according to this scheme. Therefore, by optimizing the sealing control parameters of the Q sealing process nodes, the accuracy of sealing control is improved, thereby enhancing the sealing quality.

[0069] Example 2

[0070] Based on the same inventive concept as the automated sealing method in the foregoing embodiments, such as Figure 2 As shown, the present invention also provides an automated sealing device, the device comprising:

[0071] The process flow division module 11 is used to obtain sealing process flow information and perform node division based on the sealing process flow information to obtain Q sealing process nodes, where Q is a positive integer greater than 1.

[0072] A hole sealing control feature mining module 12 is used to mine hole sealing control features based on the Q hole sealing process nodes to obtain a hole sealing control feature distribution model.

[0073] A hole sealing scene information acquisition module 13 is used to acquire hole sealing scene information;

[0074] The control parameter mining module 14 is used to mine control parameters based on the sealing scene information and the sealing control feature distribution model to obtain the sealing control optimization space.

[0075] A hole sealing control optimization module 15 is used to obtain a first hole sealing control particle based on the hole sealing control optimization space.

[0076] The sealing fit prediction module 16 is used to predict the sealing fit of the first sealing control particle based on the sealing fit optimization function, so as to obtain the sealing fit of the first particle.

[0077] The fit degree judgment module 17 is used to judge whether the fit degree of the first particle sealing hole meets the preset sealing hole fit degree constraint.

[0078] An automated sealing module 18 is configured to output the first sealing control particle as a sealing scheme if the sealing fit of the first particle meets the preset sealing fit constraint, and to perform automated sealing according to the sealing scheme.

[0079] Furthermore, the hole sealing control feature mining module 12 is also used for:

[0080] The Q sealing process nodes are sorted according to the sealing process sequence to obtain the first sealing process node, the second sealing process node, ... the qth sealing process node, where q is a positive integer and q belongs to Q;

[0081] Based on the first sealing process node, sealing control feature mining is performed to obtain the sealing control feature distribution of the first node.

[0082] Continue to mine the sealing control features of the second sealing process node...the qth sealing process node to obtain the sealing control feature distribution of the second node...the sealing control feature distribution of the qth node;

[0083] The sealing control feature distribution of the first node, the sealing control feature distribution of the second node, ... the sealing control feature distribution of the qth node are integrated to generate the sealing control feature distribution model.

[0084] Furthermore, the hole sealing control feature mining module 12 is also used for:

[0085] The sealing control record of the first sealing process node is collected to obtain multiple sealing control records of the first node;

[0086] Based on the multiple first node sealing control records, sealing control indicators are extracted to obtain multiple first node sealing control indicators.

[0087] Based on the multiple first node sealing control records, the support of the multiple first node sealing control indicators is calculated to obtain the support of multiple indicators.

[0088] Confidence scores are calculated based on the support scores of the multiple indicators to obtain the confidence scores of the multiple indicators.

[0089] Based on the preset index confidence level constraint, the multiple first node sealing control indices are screened according to the multiple index confidence levels to obtain multiple first node sealing control confidence indices that satisfy the preset index confidence level constraint, and the multiple first node sealing control confidence indices are added to the first node sealing control feature distribution.

[0090] Furthermore, the control parameter mining module 14 is also used for:

[0091] Based on the sealing scenario information, the control domain of the sealing control feature distribution of the first node is calibrated to obtain the first sealing control domain.

[0092] Based on the hole sealing scenario information, the control domain of the second node hole sealing control feature distribution...the qth node hole sealing control feature distribution is further calibrated to obtain the second hole sealing control domain...the qth hole sealing control domain;

[0093] By integrating the first sealing control domain, the second sealing control domain, ... the qth sealing control domain, a sealing scene control domain is obtained;

[0094] Based on the hole sealing scene control domain, random values ​​are selected to obtain multiple hole sealing control particles, and these multiple hole sealing control particles are added to the hole sealing control optimization space.

[0095] Furthermore, the control parameter mining module 14 is also used for:

[0096] Using the sealing scene information and the distribution of sealing control features of the first node as retrieval constraints, and based on the retrieval constraints, the sealing control parameters are retrieved to obtain the first scene-sealing control record library;

[0097] Based on the first scenario-sealing control record library, the maximum and minimum values ​​of control parameters of the first node sealing control feature distribution are calibrated to obtain the calibration results of the maximum and minimum values ​​of the first node control features.

[0098] Based on the calibration results of the maximum and minimum values ​​of the control features of the first node, the first sealing control domain is generated.

[0099] Furthermore, the sealing fit prediction module 16 is also used for:

[0100] Multi-dimensional sealing quality prediction is performed on the first sealing control particle to obtain the sealing quality prediction result of the first particle, wherein the sealing quality prediction result of the first particle includes the first predicted sealing density and the first predicted sealing stability.

[0101] Construct the hole fitting degree optimization function, wherein the hole fitting degree optimization function is:

[0102] SHT = HDY F +HSY G ;

[0103]

[0104]

[0105] Among them, SHT represents the particle sealing fit, HDY represents the predicted sealing density, HSY represents the predicted sealing stability, F represents the sealing density-exponent, G represents the sealing stability-exponent, HDY0 represents the preset sealing density, and HSY0 represents the preset sealing stability.

[0106] The prediction result of the first particle sealing quality is input into the sealing fit optimization function to generate the sealing fit of the first particle.

[0107] Furthermore, the sealing fit prediction module 16 is also used for:

[0108] Obtain records of sealing scenario information, sealing scheme, sealing density, and sealing stability;

[0109] Using the sealing scene information record and the sealing scheme record as input data, and the sealing density record as output supervision data, the BP neural network is trained to obtain a sealing density prediction channel that satisfies the convergence constraint.

[0110] Using the sealing scenario information record and the sealing scheme record as input data, and the sealing stability record as output supervision data, the BP neural network is trained to obtain a sealing stability prediction channel that satisfies the convergence constraint.

[0111] Connect the sealing density prediction channel and the sealing stability prediction channel to generate a sealing quality prediction submodule, and perform multi-dimensional sealing quality prediction of the first sealing control particle based on the sealing quality prediction submodule.

[0112] The specific example of the automated sealing method in the aforementioned embodiment 1 is also applicable to the automated sealing device in this embodiment. Through the detailed description of the automated sealing method described above, those skilled in the art can clearly understand the automated sealing device in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0113] It should be understood that various forms of the process shown above can be used, with steps rearranged, added, or deleted, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0114] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. An automated sealing method, characterized in that, The method includes: Obtain the sealing process flow information, and divide the nodes based on the sealing process flow information to obtain Q sealing process nodes, where Q is a positive integer greater than 1; Based on the Q sealing process nodes, sealing control feature mining is performed to obtain a sealing control feature distribution model; Obtain information about the sealing scene; Based on the sealing scenario information, control parameters are mined according to the sealing control feature distribution model to obtain the sealing control optimization space. Based on the hole sealing control optimization space, the first hole sealing control particle is obtained; Based on the hole sealing fit optimization function, the hole sealing fit of the first hole control particle is predicted to obtain the hole sealing fit of the first particle. Determine whether the sealing fit of the first particle meets the preset sealing fit constraint. If the sealing fit of the first particle meets the preset sealing fit constraint, the first sealing control particle is output as a sealing scheme, and automated sealing is performed according to the sealing scheme. Based on the Q sealing process nodes, sealing control feature mining is performed to obtain a sealing control feature distribution model, including: The Q sealing process nodes are sorted according to the sealing process sequence to obtain the first sealing process node, the second sealing process node, ... the qth sealing process node, where q is a positive integer and q belongs to Q; Based on the first sealing process node, sealing control feature mining is performed to obtain the sealing control feature distribution of the first node. Continue to mine the sealing control features of the second sealing process node...the qth sealing process node to obtain the sealing control feature distribution of the second node...the sealing control feature distribution of the qth node; The sealing control feature distribution of the first node, the sealing control feature distribution of the second node, ... the sealing control feature distribution of the qth node are integrated to generate the sealing control feature distribution model; Based on the first sealing process node, sealing control feature mining is performed to obtain the sealing control feature distribution of the first node, including: The sealing control record of the first sealing process node is collected to obtain multiple sealing control records of the first node; Based on the multiple first node sealing control records, sealing control indicators are extracted to obtain multiple first node sealing control indicators. Based on the multiple first node sealing control records, the support of the multiple first node sealing control indicators is calculated to obtain the support of multiple indicators. Confidence scores are calculated based on the support scores of the multiple indicators to obtain the confidence scores of the multiple indicators. Based on the preset index confidence level constraint, the multiple first node sealing control indices are screened according to the multiple index confidence levels to obtain multiple first node sealing control confidence indices that satisfy the preset index confidence level constraint, and the multiple first node sealing control confidence indices are added to the first node sealing control feature distribution.

2. The method as described in claim 1, characterized in that, Based on the sealing scenario information, control parameters are mined according to the sealing control feature distribution model to obtain the sealing control optimization space, including: Based on the sealing scenario information, the control domain of the sealing control feature distribution of the first node is calibrated to obtain the first sealing control domain. Based on the hole sealing scenario information, the control domain of the second node hole sealing control feature distribution...the qth node hole sealing control feature distribution is further calibrated to obtain the second hole sealing control domain...the qth hole sealing control domain; By integrating the first sealing control domain, the second sealing control domain, ... the qth sealing control domain, a sealing scene control domain is obtained; Based on the hole sealing scene control domain, random values ​​are selected to obtain multiple hole sealing control particles, and these multiple hole sealing control particles are added to the hole sealing control optimization space.

3. The method as described in claim 2, characterized in that, Based on the sealing scenario information, the control domain of the sealing control feature distribution of the first node is calibrated to obtain the first sealing control domain, including: Using the sealing scene information and the distribution of sealing control features of the first node as retrieval constraints, and based on the retrieval constraints, the sealing control parameters are retrieved to obtain the first scene-sealing control record library; Based on the first scenario-sealing control record library, the maximum and minimum values ​​of control parameters of the first node sealing control feature distribution are calibrated to obtain the calibration results of the maximum and minimum values ​​of the first node control features. Based on the calibration results of the maximum and minimum values ​​of the control features of the first node, the first sealing control domain is generated.

4. The method as described in claim 1, characterized in that, Based on the sealing fit optimization function, the sealing fit of the first sealing control particle is predicted to obtain the sealing fit of the first particle, including: Multi-dimensional sealing quality prediction is performed on the first sealing control particle to obtain the sealing quality prediction result of the first particle, wherein the sealing quality prediction result of the first particle includes the first predicted sealing density and the first predicted sealing stability. Construct the hole fitting degree optimization function, wherein the hole fitting degree optimization function is: ； ； ； Among them, SHT represents the particle sealing fit, HDY represents the predicted sealing density, HSY represents the predicted sealing stability, F represents the sealing density-exponent, G represents the sealing stability-exponent, HDY0 represents the preset sealing density, and HSY0 represents the preset sealing stability. The prediction result of the first particle sealing quality is input into the sealing fit optimization function to generate the sealing fit of the first particle.

5. The method as described in claim 4, characterized in that, Multi-dimensional sealing quality prediction is performed on the first sealing control particle, including: Obtain records of sealing scenario information, sealing scheme, sealing density, and sealing stability; Using the sealing scene information record and the sealing scheme record as input data, and the sealing density record as output supervision data, the BP neural network is trained to obtain a sealing density prediction channel that satisfies the convergence constraint. Using the sealing scenario information record and the sealing scheme record as input data, and the sealing stability record as output supervision data, the BP neural network is trained to obtain a sealing stability prediction channel that satisfies the convergence constraint. Connect the sealing density prediction channel and the sealing stability prediction channel to generate a sealing quality prediction submodule, and perform multi-dimensional sealing quality prediction of the first sealing control particle based on the sealing quality prediction submodule.

6. An automated sealing device, characterized in that, The apparatus is used to perform the steps of any one of the automated sealing methods according to claims 1 to 5, the apparatus comprising: The process flow division module is used to obtain sealing process flow information and perform node division based on the sealing process flow information to obtain Q sealing process nodes, where Q is a positive integer greater than 1. A hole sealing control feature mining module is used to mine hole sealing control features based on the Q hole sealing process nodes to obtain a hole sealing control feature distribution model. A hole-sealing scene information acquisition module, wherein the hole-sealing scene information acquisition module is used to acquire hole-sealing scene information; A control parameter mining module is used to mine control parameters based on the sealing scenario information and the sealing control feature distribution model to obtain the sealing control optimization space. A hole sealing control optimization module is used to obtain a first hole sealing control particle based on the hole sealing control optimization space. A sealing fit prediction module is used to predict the sealing fit of the first sealing control particle based on the sealing fit optimization function, so as to obtain the sealing fit of the first particle. A fit degree judgment module is used to determine whether the fit degree of the first particle sealing hole meets the preset sealing hole fit degree constraint. An automated sealing module is used to output the first sealing control particle as a sealing scheme if the sealing fit of the first particle meets the preset sealing fit constraint, and to perform automated sealing according to the sealing scheme.

Images