CN116228010A

CN116228010A - Information adjustment method, device, electronic equipment and computer readable medium

Info

Publication number: CN116228010A
Application number: CN202310159727.3A
Authority: CN
Inventors: 余远航; 张洪恩; 刘锋; 周龙龙; 何阳; 熊琨; 唐浩; 苏炯; 沈默; 朱玉强; 刘忠; 蔡德润; 杨行; 李忠杰
Original assignee: Zhongping Information Technology Co ltd; China Pingmei Shenma Holding Group Co ltd
Current assignee: Zhongping Information Technology Co ltd; China Pingmei Shenma Holding Group Co ltd
Priority date: 2023-02-23
Filing date: 2023-02-23
Publication date: 2023-06-06

Abstract

Embodiments of the present disclosure disclose information adjustment methods, apparatuses, electronic devices, and computer-readable media. One embodiment of the method comprises the following steps: acquiring target coke refining requirement information; determining a corresponding first cluster center from the first cluster center set; determining coke refining theoretical information; inputting at least one piece of matched coal theoretical index information into a matched coal information expansion network model to generate a matched coal theoretical index information set; controlling a target coking device to realize coke refining; responding to the end of coke refining, acquiring an actual operation parameter set and an actual index information set of the matched coal in the coking process; inputting the corresponding parameter information set and the actual index information set of the matched coal into a coking energy consumption prediction model to output coking energy consumption prediction information; and adjusting the first clustering centers in the first clustering center set. According to the embodiment, the adjusted first clustering center set can be utilized, so that refining of coke can be efficiently realized, and waste of resources is avoided.

Description

Information adjustment method, device, electronic equipment and computer readable medium

Technical Field

Embodiments of the present disclosure relate to the field of computer technology, and in particular, to an information adjustment method, an apparatus, an electronic device, and a computer readable medium.

Background

Currently, coke is an important energy source and is commonly used in various industrial industries (e.g., iron-making industry). For the refining of coke, the following methods are generally adopted: first, index information in the coke refining process is determined empirically by a person skilled in the art. And then, according to the index information, refining the coke.

However, the inventors have found that when refining coke in the above manner, there are often the following technical problems:

firstly, aiming at cokes with different quality requirements, more reasonable and accurate index information cannot be determined efficiently and flexibly, so that resources are wasted;

second, since the environmental characteristics of the environment where the coke is located are similar to the characteristics of the coke itself, the segmentation of the coke image cannot be accurately and effectively realized by the conventional image segmentation model.

The above information disclosed in this background section is only for enhancement of understanding of the background of the inventive concept and, therefore, may contain information that does not form the prior art that is already known to those of ordinary skill in the art in this country.

Disclosure of Invention

The disclosure is in part intended to introduce concepts in a simplified form that are further described below in the detailed description. The disclosure is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

Some embodiments of the present disclosure propose information adjustment methods, apparatuses, electronic devices, and computer-readable media to solve one or more of the technical problems mentioned in the background section above.

In a first aspect, some embodiments of the present disclosure provide an information adjustment method, including: acquiring target coke refining requirement information aiming at target coking equipment; determining a first cluster center corresponding to the target coke refining requirement information from a preset first cluster center set as a target cluster center; determining coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information comprises the following steps: at least one matched coal theoretical index information and a coking process required operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis; inputting the at least one piece of matched coal theoretical index information into a matched coal information expansion network model to generate a matched coal theoretical index information set; controlling the target coking equipment to realize coke refining according to the matched coal theoretical index information set and the coking process required operation parameter set; responding to the end of coke refining, acquiring an actual operation parameter set of the coking process and an actual index information set of the matched coal, which are determined by using a monitoring instrument; inputting a parameter information set corresponding to the actual operation parameter set of the coking process and an actual index information set of the matched coal into a coking energy consumption prediction model to output coking energy consumption prediction information; and adjusting a first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information.

In a second aspect, some embodiments of the present disclosure provide an information adjustment apparatus, including: a first acquisition unit configured to acquire target coke refining requirement information for a target coking plant; a first determining unit configured to determine, from a first cluster center set configured in advance, a first cluster center corresponding to the target coke refining requirement information as a target cluster center; a second determining unit configured to determine coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information includes: at least one matched coal theoretical index information and a coking process required operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis; the first input unit is configured to input the at least one piece of matched coal theoretical index information into the matched coal information expansion network model so as to generate a matched coal theoretical index information set; a control unit configured to control the target coking plant to effect coke refining based on the set of fitted coal theoretical index information and the set of coking process required operating parameters; a second acquisition unit configured to acquire an actual operation parameter set of the coking process and an actual index information set of the coal in response to the end of the coke refining, which are determined by the monitoring instrument; a second input unit configured to input a parameter information set corresponding to the actual operation parameter set of the coking process and a coal-blending actual index information set into a coking energy consumption prediction model to output coking energy consumption prediction information; and an adjustment unit configured to adjust the first cluster center in the first cluster center set based on the coking energy consumption prediction information and the target coke refining requirement information.

In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, which when executed by one or more processors, cause the one or more processors to implement the method as described in any of the implementations of the first aspect.

In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the program when executed by a processor implements a method as described in any of the implementations of the first aspect.

The above embodiments of the present disclosure have the following advantageous effects: by utilizing the adjusted first clustering center set, the information adjustment method of some embodiments of the present disclosure can efficiently refine coke and avoid resource waste. In particular, the reason for the inefficiency of the associated coke refining is that: aiming at cokes with different quality requirements, more reasonable and accurate index information cannot be determined efficiently and flexibly, so that resources are wasted. Based on this, the information adjustment method of some embodiments of the present disclosure first acquires target coke refining requirement information for a target coking plant for subsequent determination of coke refining theoretical information for the target coke refining requirement information. Then, a first cluster center corresponding to the above target coke refining requirement information is determined from a preconfigured first cluster center set as a target cluster center. Here, by the first cluster centers for each type of coke refining requirement information being configured in advance, the most reasonable target cluster center corresponding to the target coke refining requirement information and most matched with the target coking equipment performance can be quickly and accurately determined, and then the coke refining theoretical information matched with the target coke refining requirement information can be generated. Then, according to the target cluster center, the coke refining theoretical information matched with the target coke refining requirement information can be accurately determined. Wherein, the theoretical information of coke refining comprises: at least one piece of matched coal theoretical index information and a coking process requirement operation parameter set, wherein the at least one piece of matched coal theoretical index information is index information after index principal component analysis. Further, the at least one piece of matched coal theoretical index information is input into a matched coal information expansion network model to generate a matched coal theoretical index information set. The index information of the network model is expanded through the matched coal information, so that at least one matched coal theoretical index information is more perfect, and more refining parameters can be referred in the subsequent coke refining process. And then controlling the target coking equipment to accurately and efficiently refine the coke according to the matched coal theoretical index information set and the coking process required operation parameter set. Further, in response to the end of the coke refining, a set of actual operating parameters of the coking process and a set of actual index information of the blended coal determined by the monitoring instrument are obtained for subsequent determination of actual coking energy consumption by the coke refining. And finally, adjusting the first clustering center in the first clustering center set according to the coking energy consumption prediction information and the target coke refining requirement information. Here, by comparing the energy consumption between the coking energy consumption prediction information and the coking energy consumption requirement information in the target coke refining requirement information, whether the energy consumption exceeds the standard or not in the whole refining process of refining the coke according to the target cluster center is determined. Therefore, the first clustering centers in the first clustering center set are adjusted through the judgment of energy consumption, so that the first clustering center set is more accurate. Each first cluster center can accurately correspond to a unique coke refining requirement. In sum, by the application of the first clustering center set and the update of the first clustering center set, the adjusted first clustering center set is utilized, so that the coke can be efficiently refined, and the waste of resources is avoided.

Drawings

The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale.

FIG. 1 is a flow chart of some embodiments of an information adjustment method according to the present disclosure;

FIG. 2 is a schematic structural view of some embodiments of an information adjustment device according to the present disclosure;

fig. 3 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain 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 construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure.

It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings. Embodiments of the present disclosure and features of embodiments may be combined with each other without conflict.

It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.

It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.

The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Referring to fig. 1, a flow 100 of some embodiments of an information adjustment method according to the present disclosure is shown. The information adjustment method comprises the following steps:

Step 101, obtaining target coke refining requirement information for target coking equipment.

In some embodiments, the execution subject (e.g., electronic device) of the above-described information adjustment method may acquire the target coke refining requirement information for the target coking device through a wired connection or a wireless connection. The target coking plant may be a plant for refining coke. In practice, the above-described target coking plants may include, but are not limited to, at least one of the following: a coal blending and feeding device, a coke oven, a raw gas treatment device and a coke quenching device. The target coke refining requirement information may be coke index information of refined coke required by a refining requirement party. For example, the coke index information may include: index information of coke ash index, index information of coke sulfur index, standard information of coke abrasion resistance and standard information of coke crushing strength.

And 102, determining a first cluster center corresponding to the target coke refining requirement information from a preset first cluster center set as a target cluster center.

In some embodiments, the executing entity may determine a first cluster center corresponding to the target coke refining requirement information from a preconfigured first cluster center set as a target cluster center. Each cluster center can be a coding vector which aims at the corresponding coke refining requirement information and characterizes the corresponding characteristic information of the coke refining theoretical information. Each first cluster center has corresponding coke refining requirement information.

In practice, for the first cluster center as the encoding vector, the execution body may input the first cluster center into the coke refining pre-estimated information generating model, so as to output the coke refining pre-estimated information generating model (i.e. refined coke index information) corresponding to the coke refining theoretical information corresponding to the first cluster center. The coke refining estimated information generation model may be a model for generating refined coke corresponding to each coke index information. For example, the coke index generation model may be an RBF prediction model.

In practice, the first cluster center set may be generated by:

in the first step, a coke refining theory information set is obtained.

And secondly, inputting each piece of coke refining theoretical information in the coke refining theoretical information set into the coding network model to output a coding vector, and obtaining a coding vector set. The information coding network model can be a network model for carrying out information coding on the coking theoretical information. For example, the coding network model may be a transform coding model.

And thirdly, clustering the coded vector set to obtain a first clustering center set. Wherein, each first cluster center has a corresponding code vector cluster.

In some optional implementations of some embodiments, the determining, from the preconfigured first cluster center set, a first cluster center corresponding to the target coke refining requirement information as a target cluster center may include the steps of:

and determining the theoretical information of the coke refining requirement corresponding to each first cluster center in the first cluster center set, and obtaining a theoretical information set of the coke refining requirement.

And secondly, inputting each piece of coke refining requirement theoretical information in the coke refining requirement theoretical information set into a coke refining estimated information generation model so as to output coke refining estimated information. The coke refining estimation information generation model may be a model for generating coke index information of refined coke.

And thirdly, screening at least one first clustering center of which the information difference between the corresponding coke refining estimated information and the target coke refining requirement information meets the preset information difference condition from the first clustering center. The preset information difference condition may be that a difference value between corresponding coke index information is within a corresponding interval range.

And a fourth step of generating weighted summation vectors between every two first clustering centers in the at least one first clustering center to obtain a summation vector set.

And fifthly, generating verification information representing the feasibility proportion of the summation vector corresponding to the coking scheme for each summation vector in the summation vector set. Wherein the feasibility ratio may characterize the probability of success in performing a coking regimen with the target coking plant.

For example, for each summation vector in the summation vector set, the execution body may input the summation vector into a feasibility ratio generation model, generate a feasibility ratio corresponding to the summation vector, and obtain corresponding verification information. The feasibility ratio generating model can be a multi-layer convolutional neural network model.

And sixthly, removing the summation vector with the feasibility ratio smaller than a preset ratio corresponding to the verification information representation coking scheme from the summation vector set to obtain a vector set after removal. For example, the predetermined ratio is 60%.

Seventh, aggregating the vector set after removal and the at least one first clustering center to generate an aggregate vector set.

And eighth step, determining the coke refining energy consumption requirement information and the coke refining value resource consumption information corresponding to each polymerization vector in the polymerization vector set to obtain a coke refining energy consumption requirement information set and a corresponding coke refining value resource consumption information set.

And a ninth step of creating a table for the coke refining energy consumption information set and the coke refining value resource consumption information set.

And tenth, transmitting the table to a client of a refining requester for selecting theoretical information corresponding to the coke refining requirement by the clustering center.

And step 103, determining coke refining theoretical information according to the target clustering center.

In some embodiments, the executive may determine the coke refining theory information based on the target cluster center. Wherein, the theoretical information of coke refining comprises: at least one of the coordinated coal theoretical index information and the coking process requires a set of operating parameters. The at least one piece of matched coal theoretical index information is the index information after the principal component analysis. The coal blend theoretical index information may be the theoretical index information of the blend used for refining coke. In practice, the at least one coal blend theoretical index information may include, but is not limited to, at least one of: ash content index information, moisture index information, sulfur content index information, volatile content index information, coking time index information and matching coal corresponding proportion index information. The coking process required operating parameter may be a refining operating parameter in a coke refining process. In practice, the coking process required operating parameters may include, but are not limited to, at least one of the following: the temperature parameter of the side flue, the temperature parameter of the coke side flue, the suction parameter of the side flue and the suction parameter of the coke side flue. The index principal component analysis may be a principal component analysis method (PCA, principal Components Analysis) for a theoretical index of the blended coal.

As an example, the above-described execution subject may input the target cluster center to the decoding model to output coke refining theory information. In practice, the decoding model may be a transducer decoding model.

And 104, inputting the at least one piece of matched coal theoretical index information into a matched coal information expansion network model to generate a matched coal theoretical index information set.

In some embodiments, the executing entity may input the at least one coal blending theory index information into a coal blending information expansion network model to generate a coal blending theory index information set. The coal blending information expansion network model can be a model for generating index information corresponding to coal blending. In practice, the network model for the coal information expansion can be a multi-head attention mechanism model. The number of the matched coal theoretical index information included in the matched coal theoretical index information set is greater than the number of the matched coal theoretical index information included in at least one matched coal theoretical index information.

And 105, controlling the target coking equipment to realize coke refining according to the matched coal theoretical index information set and the coking process required operation parameter set.

In some embodiments, the executive may control the target coking plant to effect coke refining based on the set of fitted coal theoretical index information and the set of coking process desired operating parameters.

As an example, the executing body may perform index adjustment on indexes corresponding to the blended coal according to the set of theoretical index information of the blended coal, and adjust operation parameters in the coking process according to the set of operation parameters required by the coking process, so as to control the target coking equipment, thereby realizing the refining of coke.

And 106, responding to the end of coke refining, and acquiring an actual operation parameter set of the coking process and an actual index information set of the matched coal, which are determined by using a monitoring instrument.

In some embodiments, the executive may obtain a set of actual operating parameters of the coking process and a set of actual index information of the blending coal determined using the monitoring instrument in response to the coke refining ending. The monitoring instrument can be an instrument for detecting the actual index information of the matched coal and/or the actual operation parameters of the coking process.

Step 107, inputting the parameter information set corresponding to the actual operation parameter set of the coking process and the actual index information set of the blending coal into the coking energy consumption prediction model to output coking energy consumption prediction information.

In some embodiments, the execution entity may input a parameter information set corresponding to the actual operation parameter set of the coking process and a blended coal actual index information set to the coking energy consumption prediction model to output coking energy consumption prediction information. The coking energy consumption prediction model may be a model for predicting coking energy consumption information. Coking energy consumption may be the energy information consumed in the overall process of refining coke. In practice, the coking energy consumption prediction model may be a Residual Network (ResNet) model.

And step 108, adjusting the first clustering center in the first clustering center set according to the coking energy consumption prediction information and the target coke refining requirement information.

In some embodiments, the executing entity may adjust the first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information.

In some alternative implementations of some embodiments, the target coke refining requirement information includes: coke refining energy consumption requirement information. The coke refining energy consumption requirement information may be maximum energy consumption requirement information of refining coke.

Optionally, the adjusting the first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information may include the following steps:

in the first step, a vector for the coking energy consumption prediction information is generated in response to determining that the energy consumption value corresponding to the coking energy consumption prediction information is greater than the energy consumption value corresponding to the coke refining energy consumption requirement information.

As an example, first, the above-described execution subject may generate coke refining actual information from a coking process actual operation parameter set and a blending coal actual index information set. The coke refining actual information is then input to the code network model to output the above vectors.

And secondly, adding the vector into a vector cluster corresponding to the target cluster center to serve as a target vector cluster.

And thirdly, determining the vector distance between the vector and the target clustering center.

Wherein the vector distance may be a cosine distance.

And step four, generating a vector region range by taking a preset multiple value of the vector distance as a radius and taking the target clustering center as a center. For example, the predetermined multiple value may be 2 times.

And fifthly, determining a vector set in the vector area range from the target vector cluster.

And sixthly, clustering the vector sets to obtain a second aggregation center set.

And seventh, determining the weighted summation vector corresponding to each second aggregation center in the second aggregation center set.

And eighth step, determining a vector with the nearest cosine distance corresponding to the weighted sum vector from the vector set as a nearest cluster center.

And ninth, according to the nearest cluster center, carrying out cluster center updating on the first cluster center set to obtain a candidate cluster center set.

As an example, first, the execution subject may replace the cluster center corresponding to the first cluster center set with the nearest cluster center, to obtain a replaced cluster center set. And then, carrying out cluster center updating processing on the replacement cluster center set again by utilizing the vector set corresponding to the replacement cluster center set to obtain a candidate cluster center set.

And tenth, generating an adjusted cluster center set according to the candidate cluster center set.

As an example, the execution subject described above may determine the candidate cluster center set as the adjusted cluster center set.

Optionally, generating the adjusted cluster center set according to the candidate cluster center set may include the following steps:

first, a cluster partition label information set is obtained. The cluster division tag information may be tag information corresponding to each cluster. I.e. the individual tag information differences between every two clusters are relatively large. According to each label, the clustering can be realized. In practice, the clustering of the partitioned tag information sets includes: at least one of the remaining coal index tags. The label of at least one other index label of the blended coal is completely different from at least one label corresponding to at least one theoretical index information of the blended coal. In practice, at least one of the remaining blended coal index tags may comprise: the fineness label of coal, the coalification degree label, the lithology label and the expansion pressure label of coal are matched.

And a second step of determining a label level corresponding to each cluster partition label information in the cluster partition label information set. Each label level is provided with a corresponding cluster division label information group.

For example, the at least one remaining blended coal index tag includes: the first matched coal theoretical index label, the second matched coal theoretical index label and the third matched coal theoretical index label. The label level includes: a first level, a second level, and a third level. The label grade corresponding to the first matched coal theoretical index label is a first grade. The label grade corresponding to the second matched coal theoretical index label is a second grade. And the label grade corresponding to the label of the third matched coal theoretical index label is a third grade.

And thirdly, inputting each candidate cluster center in the candidate cluster center set into a cluster partition label generation model to generate at least one cluster partition label content aiming at the candidate cluster center. The cluster partition label generation model may be a model for generating at least one other coal index label corresponding to at least one other coal index information. In practice, the cluster partition label generation model may be a generation formula and an countermeasure neural network model.

Fourth, for each two candidate cluster centers in the candidate cluster center set, the following determination steps are performed:

sub-step 1, for each tag level, the following scale generation step is performed:

and a first sub-step of determining at least one cluster partition label information with a cluster partition label content difference between every two candidate cluster centers in response to determining that the corresponding cluster partition label content has a difference.

And a second sub-step of determining a label level difference ratio corresponding to the label information of the at least one cluster division. The label level difference ratio may be a ratio of the number of information corresponding to the label information divided by at least one cluster to the number of information corresponding to the label information set divided by the cluster.

And 2, generating merging information representing whether each two candidate cluster centers are to-be-merged or not according to the obtained label level difference proportion set.

As an example, in response to determining that the label level difference ratio corresponding to the label level with the highest level is greater than 60%, merging information is generated, wherein the merging information characterizes that each two candidate cluster centers are cluster centers to be merged. And generating merging information representing that every two candidate cluster centers are not to be cluster centers to be merged in response to the fact that the label level difference proportion corresponding to the label level with the highest level is less than 60% and the label level difference proportion corresponding to the label levels of the rest preset numbers is less than 70%. And generating merging information representing that each two candidate cluster centers are to-be-merged in response to the fact that the label level difference proportion corresponding to the label level with the highest level is less than 60% and the label level difference proportion corresponding to the label levels of the rest preset numbers is greater than 70%. Wherein the predetermined number may be the number of tag levels minus a value of 2.

And fifthly, determining the candidate cluster center set as an adjusted cluster center set in response to determining that the merging information corresponding to each two candidate cluster centers characterizes that the corresponding cluster centers are not merged.

Optionally, the steps further include:

in the first step, at least two candidate cluster centers to be combined are determined in response to determining that the combination information corresponding to the two candidate cluster centers is present to characterize the combination of the corresponding cluster centers.

And a second step of determining whether repeated candidate cluster centers exist in the at least two candidate cluster centers.

And thirdly, responding to the fact that the cluster does not exist, and carrying out corresponding cluster combination on corresponding cluster center pairs in at least two candidate cluster centers to be combined according to the obtained combination information set to obtain at least one combined cluster.

And fourth, for each merging cluster in the at least one merging cluster, determining a weighted summation vector corresponding to two clustering centers corresponding to the merging cluster as a merging cluster center corresponding to the merging cluster.

And fifthly, carrying out cluster updating on clusters corresponding to the first cluster center set according to the at least one combined cluster to obtain updated clusters.

As an example, the execution body may replace a corresponding cluster subset in the cluster with at least one merged cluster, to obtain the updated cluster.

And sixthly, generating the cluster center set after adjustment according to the updated cluster set and the obtained combined cluster center set.

As an example, the execution entity may re-perform the calculation and selection of the cluster center according to the updated cluster set and the merged cluster center set, so as to generate the adjusted cluster center set.

In some alternative implementations of some embodiments, after step 108, the steps further include:

in a first step, a coke image for a refined coke set is acquired.

And secondly, inputting the coke image into a coke segmentation model to generate a coke sub-image, and obtaining a coke sub-image set. The number of pixels of the coke object corresponding to each coke sub-image in the coke sub-image set is greater than a predetermined number. The coke segment model may be a neural network model that segments out coke sub-images. For example, the coke segment model may be a Mask R-CNN model.

And thirdly, performing image selective supplementary processing on each coke sub-image in the coke sub-image set according to the coke position information corresponding to each coke sub-image so as to generate a supplementary image and obtain a supplementary image set.

As an example, first, in response to determining that a coke site exhibited by a coke sub-image is a tangential site, a pixel gradient map of a pixel region corresponding to the tangential site is determined. And then, according to the pixel gradient map, performing pixel supplementation on the corresponding pixel region of the cut part to obtain a supplemented image. In response to determining that the coke sites exhibited by the coke sub-images are angular sites, the coke sub-images are not pixel-supplemented.

And step four, inputting each supplementary image in the supplementary image set into a feature extraction model to generate image feature information, and obtaining an image feature information set. The feature extraction model may be a model that extracts features of an image. In practice, the feature extraction model includes: and (3) a texture feature extraction sub-model and a granularity feature extraction sub-model. The feature extraction model may be a plurality of convolutional neural network modules. Each convolutional neural network module is composed of at least one convolutional neural network in series.

And fifthly, clustering the image characteristic information set to obtain a cluster center set.

Sixth, for each cluster center in the cluster center set, performing the following generating steps:

and 1, inputting the cluster center into a coke appearance information generation model to output coke granularity uniformity degree information, coke granularity transverse and longitudinal crack information, coke section information and coke edge angle information. The coke appearance information generation model may be a model that generates coke appearance information. In practice, the coke appearance information generation model may be a multi-class output residual network model.

And 2, generating first coke quality verification information according to the coke granularity uniformity degree information, the coke transverse and longitudinal crack information, the coke section information and the coke edge angle information.

As an example, in response to determining that the coke particle size uniformity information meets a preset particle size condition, the coke transverse and longitudinal crack information meets a preset transverse and longitudinal crack condition, and the coke angular information meets a preset angular condition, verification information characterizing that the coke quality is acceptable is generated. The preset granularity condition may be that the uniformity degree of the corresponding coke granularity is in a preset interval. The preset transverse and longitudinal crack condition may be that the number of transverse and longitudinal cracks corresponding to the coke is less than a predetermined number and the transverse and longitudinal crack size is less than a predetermined size. The preset angular condition may be that the angular definition of the corresponding coke is greater than a predetermined definition level and the sharpness of the angular is greater than a predetermined sharpness level.

And fifthly, generating quality verification information for the refined coke set according to the obtained first coke quality verification information set.

As an example, first, a duty cycle of the first set of coke quality verification information corresponding to verification information characterizing coke quality pass is determined. Then, in response to determining that the duty ratio is greater than 60%, verification information is generated that characterizes the quality of the refined coke set as acceptable.

Optionally, the performing body inputs the coke image to a coke segmentation model to generate a coke sub-image, and obtains a coke sub-image set, and may include the steps of:

According to a preset pixel division threshold, a binarization model included in a coke segmentation model is utilized to carry out binarization processing on a coke image, and a binarization image is obtained. For example, the preset pixelation threshold may be 30.

And a second step of determining a binarized image area with a corresponding value of 1 and a pixel number greater than a preset number from the binarized image to obtain at least one binarized image area.

And thirdly, determining the area boundary shape corresponding to each of the at least one binarized image area.

And a fourth step of removing the binarized image area with the corresponding shape not being the preset shape from the at least one binarized image area to obtain a first removed binarized image area set. In practice, the preset shape may be a triangular shape.

And fifthly, determining density information with the corresponding value of 1 in each first removed binarized image area in the first removed binarized image area set.

And sixthly, removing the first removed binarized image area with the corresponding density information smaller than a preset value from the first removed binarized image area set to obtain a second removed binarized image area set.

And seventhly, inputting the image subset corresponding to the second removed binarized image region set into a coke identification model included in the coke segmentation model to output a coke identification result set. The coke identification model may be a model that identifies whether the image object is coke. For example, the coke identification model may be a multi-layer convolutional neural network model.

And eighth, removing images which are not coke and are represented by the corresponding coke identification result from the image subsets, and obtaining the removed image subsets.

And a ninth step of inputting each removed image in the removed image subset to an image segmentation model included in the coke segmentation model to output a coke sub-image set. The image segmentation model may be a model that segments the coke objects in the image. For example, the image segmentation model may be the Faster RCNN model.

The foregoing first to ninth steps, as an invention point of the present disclosure, solve the second technical problem mentioned in the background art, that is, "since the environmental features of the environment where the coke is located are similar to the features of the coke itself, the segmentation of the coke image cannot be accurately and effectively achieved through the conventional image segmentation model. Based on the method, through binarization processing of the coke image, image content, corresponding to tone and black, in the coke image can be screened out preliminarily, so that relevant background characteristic information is removed, and the subsequent coke identification model is prevented from learning more useless characteristic information. In addition, through screening of the corresponding shapes and the corresponding densities of the binarized image areas, article information with similar color tone to the coke and ground area information with lower coke density in the coke image can be removed, so that the removed image subset corresponding to the binarized image area set is more reflected to be related to the coke. Based on the method, more background characteristic information in the coke image can be removed by utilizing the image segmentation model, so that the coke segmentation is more efficient and accurate.

Optionally, the generating quality verification information for the refined coke set according to the obtained first coke quality verification information set may include the following steps:

in a first step, char landing audio for a subset of refined char is obtained. Wherein the subset of refined cokes is a predetermined number of cokes randomly selected from the set of refined cokes.

And secondly, inputting the coke landing audio to a coke landing sound belonging type determining model to generate landing sound types for each refined coke in the refined coke subset, and obtaining a landing sound type set. The category determination model to which the coke landing sound belongs may be a model that determines a category of the landing sound to which the coke landing sound corresponds. The floor sound categories may include: clear and crisp category, clunk category. The water content of the corresponding coke is different, the landing sound is different, and the landing sound is more smoky as the water content is larger. In practice, the coke landing sound belonging category determination model may be DNN (Deep-Learning Neural Network, deep neural network).

And thirdly, generating second coke quality verification information according to the floor sound class set.

As an example, first, the above-described execution subject may determine the floor sound category ratio corresponding to the floor sound category as the crisp category. Then, in response to determining that the floor sound class ratio is greater than 60%, verification information is generated that characterizes the coke subset quality as acceptable.

Fourth, generating quality verification information according to the first coke quality verification information set and the second coke quality verification information.

As an example, in response to determining that the first set of coke quality verification information characterizes quality of the refined coke set as being acceptable, quality verification information characterizing quality of the refined coke set as being acceptable is generated. In response to determining that the first set of coke quality verification information characterizes the quality of the refined coke set as failed and the second set of coke quality verification information characterizes the quality of the subset of coke as failed, quality verification information characterizing the quality of the refined coke set as failed is generated.

With further reference to fig. 2, as an implementation of the method shown in the above figures, the present disclosure provides some embodiments of an information adjustment device, which correspond to those method embodiments shown in fig. 1, and which are particularly applicable in various electronic apparatuses.

As shown in fig. 2, an information adjustment apparatus 200 includes: a first acquisition unit 201, a first determination unit 202, a second determination unit 203, a first input unit 204, a control unit 205, a second acquisition unit 206, a second input unit 207, and an adjustment unit 208. Wherein, the first acquisition unit 201 is configured to acquire target coke refining requirement information for a target coking plant; a first determining unit 202 configured to determine, from a first cluster center set configured in advance, a first cluster center corresponding to the above-described target coke refining requirement information as a target cluster center; a second determining unit 203 configured to determine coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information includes: at least one matched coal theoretical index information and a coking process required operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis; a first input unit 204 configured to input the at least one piece of fitted coal theoretical index information into a fitted coal information expansion network model to generate a set of fitted coal theoretical index information; a control unit 205 configured to control the target coking plant to effect coke refining based on the set of fitted coal theoretical index information and the set of coking process demand operating parameters; a second acquisition unit 206 configured to acquire an actual operation parameter set of the coking process and an actual index information set of the coal to be blended determined by the monitoring instrument in response to the end of the coke refining; a second input unit 207 configured to input a parameter information set corresponding to the actual operation parameter set of the coking process and a coal-blending actual index information set to the coking energy consumption prediction model to output coking energy consumption prediction information; and an adjustment unit 208 configured to adjust the first cluster center among the first cluster centers based on the coking energy consumption prediction information and the target coke refining requirement information.

It will be appreciated that the elements described in the information adjustment device 200 correspond to the various steps in the method described with reference to fig. 1. Thus, the operations, features and advantages described above for the method are equally applicable to the information adjustment device 200 and the units contained therein, and are not described herein.

Referring now to fig. 3, a schematic diagram of an electronic device (e.g., electronic device) 300 suitable for use in implementing some embodiments of the present disclosure is shown. The electronic device shown in fig. 3 is merely an example and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.

As shown in fig. 3, the electronic device 300 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 301 that may perform various suitable actions and processes in accordance with programs stored in a read-only memory 302 or programs loaded from a storage 308 into a random access memory 303. In the RAM 303, various programs and data required for the operation of the electronic apparatus 300 are also stored. The processing means 301, the read only memory 302 and the random access memory 303 are connected to each other by a bus 304. An input/output interface 305 is also connected to the bus 304.

In general, the following devices may be connected to the input/output interface 305: input devices 306 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 307 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 308 including, for example, magnetic tape, hard disk, etc.; and communication means 309. The communication means 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. While fig. 3 shows an electronic device 300 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead. Each block shown in fig. 3 may represent one device or a plurality of devices as needed.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via communications device 309, or from storage device 308, or from read only memory 302. The above-described functions defined in the methods of some embodiments of the present disclosure are performed when the computer program is executed by the processing means 301.

It should be noted that, in some embodiments of the present disclosure, the computer readable medium may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

In some implementations, the clients, servers may communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol ), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring target coke refining requirement information aiming at target coking equipment; determining a first cluster center corresponding to the target coke refining requirement information from a preset first cluster center set as a target cluster center; determining coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information comprises the following steps: at least one matched coal theoretical index information and a coking process required operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis; inputting the at least one piece of matched coal theoretical index information into a matched coal information expansion network model to generate a matched coal theoretical index information set; controlling the target coking equipment to realize coke refining according to the matched coal theoretical index information set and the coking process required operation parameter set; responding to the end of coke refining, acquiring an actual operation parameter set of the coking process and an actual index information set of the matched coal, which are determined by using a monitoring instrument; inputting a parameter information set corresponding to the actual operation parameter set of the coking process and an actual index information set of the matched coal into a coking energy consumption prediction model to output coking energy consumption prediction information; and adjusting a first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information.

Computer program code for carrying out operations for some embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in some embodiments of the present disclosure may be implemented by means of software, or may be implemented by means of hardware. The described units may also be provided in a processor, for example, described as: a processor includes a first acquisition unit, a first determination unit, a second determination unit, a first input unit, a control unit, a second acquisition unit, a second input unit, and an adjustment unit. The names of these units are not limited to the unit itself in some cases, and for example, the first acquisition unit may also be described as "a unit that acquires target coke refining requirement information for a target coking plant".

The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.

The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above technical features, but encompasses other technical features formed by any combination of the above technical features or their equivalents without departing from the spirit of the invention. Such as the above-described features, are mutually substituted with (but not limited to) the features having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. An information adjustment method, comprising:

acquiring target coke refining requirement information aiming at target coking equipment;

determining a first cluster center corresponding to the target coke refining requirement information from a preset first cluster center set, and taking the first cluster center as a target cluster center;

determining coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information comprises: at least one matched coal theoretical index information and a coking process requirement operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis;

inputting the at least one piece of matched coal theoretical index information into a matched coal information expansion network model to generate a matched coal theoretical index information set;

controlling the target coking equipment to realize coke refining according to the matched coal theoretical index information set and the coking process required operation parameter set;

responding to the end of coke refining, acquiring an actual operation parameter set of the coking process and an actual index information set of the matched coal, which are determined by using a monitoring instrument;

inputting a parameter information set corresponding to the actual operation parameter set of the coking process and an actual index information set of the matched coal into a coking energy consumption prediction model to output coking energy consumption prediction information;

And adjusting a first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information.

2. The method of claim 1, wherein the target coke refining requirement information comprises: coke refining energy consumption requirement information; and

the adjusting the first clustering center in the first clustering center set according to the coking energy consumption prediction information and the target coke refining requirement information comprises the following steps:

generating a coking energy consumption prediction information vector for the coking energy consumption in response to determining that the energy consumption value corresponding to the coking energy consumption prediction information is greater than the energy consumption value corresponding to the coke refining energy consumption requirement information;

adding the vector into a vector cluster corresponding to the target cluster center to serve as a target vector cluster;

determining a vector distance between the vector and the target cluster center;

generating a vector region range by taking a preset multiple value of the vector distance as a radius and the target clustering center as a center;

determining a vector set in the vector area range from the target vector cluster;

clustering the vector sets to obtain a second aggregation center set;

Determining a weighted summation vector corresponding to each second aggregation center in the second aggregation center set;

determining a vector with the nearest cosine distance corresponding to the weighted sum vector from the vector set as a nearest clustering center;

according to the nearest cluster center, carrying out cluster center updating on the first cluster center set to obtain a candidate cluster center set;

and generating an adjusted cluster center set according to the candidate cluster center set.

3. The method of claim 2, wherein the generating an adjusted cluster center set from the candidate cluster center set comprises:

acquiring a cluster partition label information set;

determining a label level corresponding to each cluster partition label information in the cluster partition label information set, wherein each label level has a corresponding cluster partition label information set;

inputting each candidate cluster center in the candidate cluster center set to a cluster partition label generation model to generate at least one cluster partition label content for the candidate cluster center;

for each two candidate cluster centers in the candidate cluster center set, performing the following determination steps:

For each tag level, the following scale generation step is performed:

determining at least one cluster partition label information with the cluster partition label content difference between every two candidate cluster centers in response to determining that the corresponding cluster partition label content has the difference;

determining a label level difference proportion corresponding to the label information of the at least one cluster partition;

generating merging information representing whether each two candidate cluster centers are cluster centers to be merged or not according to the obtained label level difference proportion set;

and determining the candidate cluster center set as an adjusted cluster center set in response to determining that the merging information corresponding to each two candidate cluster centers characterizes that the corresponding cluster centers are not merged.

4. The method of claim 3, wherein the determining, from a preconfigured first cluster center set, a first cluster center corresponding to the target coke refining requirement information as a target cluster center comprises:

determining theoretical information of coke refining requirements corresponding to each first cluster center in the first cluster center sets to obtain theoretical information sets of coke refining requirements;

inputting each piece of coke refining requirement theoretical information in the coke refining requirement theoretical information set into a coke refining estimated information generation model so as to output coke refining estimated information;

Screening at least one first cluster center from the first cluster center set, wherein the information difference between the corresponding coke refining estimated information and the target coke refining requirement information meets the preset information difference condition;

generating a weighted summation vector between every two first cluster centers in the at least one first cluster center to obtain a summation vector set;

for each summation vector in the summation vector set, generating verification information representing the feasibility proportion of the summation vector corresponding to the coking scheme;

removing summation vectors with the feasibility proportion of the corresponding verification information representing the coking scheme smaller than a preset proportion from the summation vector set to obtain a removed vector set;

aggregating the removed vector set and the at least one first clustering center to generate an aggregate vector set;

determining coke refining energy consumption requirement information and coke refining value resource consumption information corresponding to each polymerization vector in the polymerization vector set to obtain a coke refining energy consumption requirement information set and a corresponding coke refining value resource consumption information set;

generating a table for the coke refining energy consumption requirement information set and the coke refining value resource consumption information set;

And sending the table to a client of a refining requester for selecting theoretical information corresponding to the coke refining requirement by the clustering center.

5. The method of claim 4, wherein the method further comprises:

acquiring a coke image for a refined coke set;

inputting the coke image into a coke segmentation model to generate a coke sub-image, and obtaining a coke sub-image set;

performing image selective supplementary processing on each coke sub-image in the coke sub-image set according to coke position information corresponding to each coke sub-image to generate a supplementary image, thereby obtaining a supplementary image set;

inputting each supplementary image in the supplementary image set to a feature extraction model to generate image feature information, and obtaining an image feature information set;

clustering the image characteristic information set to obtain a cluster center set;

for each cluster center in the cluster center set, performing the generating step of:

inputting the cluster center into a coke appearance information generation model to output coke granularity uniformity degree information, coke transverse and longitudinal crack information, coke section information and coke edge angle information;

generating first coke quality verification information according to the coke granularity uniformity degree information, the coke granularity transverse and longitudinal crack information, the coke section information and the coke edge angle information;

And generating quality verification information for the refined coke set according to the obtained first coke quality verification information set.

6. The method of claim 5, wherein the generating quality-verification information for the refined coke set from the resulting first coke quality-verification information set comprises:

acquiring coke landing audio for a refined coke subset, wherein the refined coke subset is a predetermined number of cokes randomly selected from the refined coke set;

inputting the coke landing audio to a coke landing sound belonging type determining model to generate a landing sound type for each refined coke in the refined coke subset, so as to obtain a landing sound type set;

generating second coke quality verification information according to the floor sound class set;

and generating quality verification information according to the first coke quality verification information set and the second coke quality verification information.

7. The method of claim 6, wherein the method further comprises:

determining at least two candidate cluster centers to be combined in response to determining that the combination information corresponding to the two candidate cluster centers exists to characterize the combination of the corresponding cluster centers;

Determining whether there are duplicate candidate cluster centers in the at least two candidate cluster centers;

in response to determining that the cluster does not exist, corresponding cluster combination is carried out on corresponding cluster center pairs in at least two candidate cluster centers to be combined according to the obtained combination information set, and at least one combined cluster is obtained;

for each merging cluster in the at least one merging cluster, determining weighted summation vectors corresponding to two cluster centers corresponding to the merging cluster as merging cluster centers corresponding to the merging cluster;

according to the at least one merging cluster, cluster updating is carried out on clusters corresponding to the first cluster center set, and updated clusters are obtained;

and generating the adjusted cluster center set according to the updated cluster set and the obtained combined cluster center set.

8. An information adjustment device, comprising:

a first acquisition unit configured to acquire target coke refining requirement information for a target coking plant;

a first determination unit configured to determine, from a preconfigured first cluster center set, a first cluster center corresponding to the target coke refining requirement information as a target cluster center;

a second determination unit configured to determine coke refining theoretical information according to the target cluster center, wherein the coke refining theoretical information includes: at least one matched coal theoretical index information and a coking process requirement operation parameter set, wherein the at least one matched coal theoretical index information is index information after index principal component analysis;

A first input unit configured to input the at least one fitted coal theoretical index information to a fitted coal information expansion network model to generate a set of fitted coal theoretical index information;

a control unit configured to control the target coking plant to effect coke refining based on the set of fitted coal theoretical index information and the set of coking process required operating parameters;

a second acquisition unit configured to acquire an actual operation parameter set of the coking process and an actual index information set of the coal in response to the end of the coke refining, which are determined by the monitoring instrument;

a second input unit configured to input a parameter information set corresponding to the coking process actual operation parameter set and a coal blending actual index information set to a coking energy consumption prediction model to output coking energy consumption prediction information;

and an adjustment unit configured to adjust a first cluster center in the first cluster center set according to the coking energy consumption prediction information and the target coke refining requirement information.

9. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1-7.

10. A computer readable medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of any of claims 1-7.