CN117723514B - Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method - Google Patents
Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method Download PDFInfo
- Publication number
- CN117723514B CN117723514B CN202410172597.1A CN202410172597A CN117723514B CN 117723514 B CN117723514 B CN 117723514B CN 202410172597 A CN202410172597 A CN 202410172597A CN 117723514 B CN117723514 B CN 117723514B
- Authority
- CN
- China
- Prior art keywords
- coal quality
- coal
- quality analysis
- sample
- initial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003245 coal Substances 0.000 title claims abstract description 358
- DGAQECJNVWCQMB-PUAWFVPOSA-M Ilexoside XXIX Chemical compound C[C@@H]1CC[C@@]2(CC[C@@]3(C(=CC[C@H]4[C@]3(CC[C@@H]5[C@@]4(CC[C@@H](C5(C)C)OS(=O)(=O)[O-])C)C)[C@@H]2[C@]1(C)O)C)C(=O)O[C@H]6[C@@H]([C@H]([C@@H]([C@H](O6)CO)O)O)O.[Na+] DGAQECJNVWCQMB-PUAWFVPOSA-M 0.000 title claims abstract description 36
- 239000011734 sodium Substances 0.000 title claims abstract description 36
- 229910052708 sodium Inorganic materials 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 309
- 238000012549 training Methods 0.000 claims abstract description 122
- 238000001228 spectrum Methods 0.000 claims abstract description 38
- 238000012216 screening Methods 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims description 45
- 238000012360 testing method Methods 0.000 claims description 28
- 230000008569 process Effects 0.000 claims description 26
- 238000007781 pre-processing Methods 0.000 claims description 20
- 238000013101 initial test Methods 0.000 claims description 16
- 238000002203 pretreatment Methods 0.000 claims description 9
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 claims description 8
- 239000011593 sulfur Substances 0.000 claims description 8
- 229910052717 sulfur Inorganic materials 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 239000010883 coal ash Substances 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 description 5
- 230000002285 radioactive effect Effects 0.000 description 5
- 239000002956 ash Substances 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000000862 absorption spectrum Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- -1 moisture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000010238 partial least squares regression Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to the field of coal quality analysis, and particularly discloses a multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm and a detection method. The multi-index coal quality analyzer comprises a sodium light source, a detector, an intelligent analysis chip and a touch screen. The touch screen is used for displaying a plurality of candidate indexes, an initial sample set and a primary screening coal quality analysis model set. The intelligent analysis chip is used for training the first number of coal quality analysis models by using the corresponding actual training sample set to obtain the first number of actual coal quality analysis models; analyzing the characteristic spectrum by using a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal analysis report according to the first quantity of coal analysis results. Therefore, the safety risk brought to operators is avoided, and the coal quality analysis result and the finally generated coal quality analysis report are more accurate.
Description
Technical Field
The invention relates to the field of coal quality analysis, in particular to a multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm and a detection method.
Background
The multi-index coal quality analyzer is an instrument capable of rapidly analyzing multiple indexes of coal such as ash content, moisture content, sulfur content and the like.
In the detection and analysis process of the existing multi-index coal quality analyzer, the following technical problems often exist:
firstly, a radioactive source (such as a gamma radioactive source) is needed in the detection process, the radioactive source needs to be strictly approved in the links of purchasing, installing, checking and accepting, and the like, and the requirement on operators is high, if the operation is improper, the safety risk is easily brought to the operators; in addition, the existing multi-index coal quality analyzer has the problems of low configuration flexibility and low detection speed;
Secondly, in the actual analysis and inspection process, the production place and the types of the coal are different, and the characteristic difference of the coal of different types is larger, so that when the types of the coal corresponding to the training sample are different from those of the coal in actual detection, the accuracy of the coal analysis result output by the coal analysis model and the finally generated coal analysis report are reduced;
Thirdly, in the actual analysis and inspection process, the existing multi-index coal analyzers are independently operated, so that each multi-index coal analyzer needs to independently complete the processes of model training and the like, and different multi-index coal analyzers are difficult to realize data sharing, so that the efficiency is low.
Disclosure of Invention
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary 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.
The invention provides a multi-index coal quality analyzer and a detection method based on a sodium light source and an intelligent algorithm, which are used for solving one or more of the technical problems mentioned in the background art section.
The invention provides a multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm, which comprises the following components: the sodium light source is used for emitting electromagnetic waves to the coal to be tested so as to excite the coal to be tested to emit a characteristic spectrum; the detector is used for detecting the characteristic spectrum and sending the characteristic spectrum to the intelligent analysis chip; the intelligent analysis chip is provided with a candidate coal quality analysis model set and an initial sample set, wherein each coal quality analysis model in the candidate coal quality analysis model set is constructed based on different intelligent algorithms; the touch screen is used for displaying a plurality of candidate indexes, wherein the plurality of candidate indexes comprise coal ash, coal moisture and coal sulfur, at least one index selected by a user from the plurality of candidate indexes is received, the at least one index forms a target index set, and the touch screen is used for sending the target index set to the intelligent analysis chip; the intelligent analysis chip is used for screening a plurality of coal quality analysis models from the candidate coal quality analysis model set according to the target index set, and the screened plurality of coal quality analysis models form a primary screening coal quality analysis model set; the touch screen is also used for displaying a sample set editor, and the sample set editor is used for receiving sample editing information input by a user and dividing an initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information; the touch screen is also used for displaying a primary screening coal quality analysis model set and the historical prediction accuracy of each coal quality analysis model in the primary screening coal quality analysis model set, so that a user selects a first number of coal quality analysis models from the primary screening coal quality analysis model set according to the historical prediction accuracy, and the first number of coal quality analysis models form a target coal quality analysis model set; the touch screen is also used for receiving first configuration information input by a user and sending the first configuration information to the intelligent analysis chip, and the intelligent analysis chip is used for configuring corresponding relations for the first number of coal quality analysis models and the first number of actual training sample sets according to the first configuration information so that each coal quality analysis model in the first number of coal quality analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one; the intelligent analysis chip is also used for training the first number of coal quality analysis models by utilizing the corresponding actual training sample set to obtain the first number of actual coal quality analysis models; analyzing the characteristic spectrum by using a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal analysis report according to the first quantity of coal analysis results.
Optionally, a preprocessing method set is further deployed on the intelligent analysis chip, the touch screen is further used for displaying the preprocessing method set, receiving a first number of preprocessing methods selected by a user from the preprocessing method set, receiving second configuration information input by the user and sending the second configuration information to the intelligent analysis chip, and the intelligent analysis chip is used for configuring a corresponding relation between the first number of preprocessing methods and the first number of actual training sample sets according to the second configuration information, so that each preprocessing method in the first number of preprocessing methods corresponds to each actual training sample set in the first number of actual training sample sets one by one; the intelligent analysis chip is further used for preprocessing the first number of actual training sample sets by using a first number of preprocessing methods before the corresponding actual training sample sets are used for respectively training the first number of coal analysis models, so as to obtain the processed first number of actual training sample sets, and training the first number of coal analysis models by using the processed first number of actual training sample sets.
Optionally, the sodium light source comprises a plurality of sub-light sources, each of the plurality of sub-light sources having a different energy intensity. Electromagnetic waves with different energy intensities have different action cross sections with each element in coal, so that different characteristic performances are presented, and more information of the coal can be contained.
Optionally, the first number of actual training sample sets and the second number of test sample sets are divided by: receiving a first quantity, a second quantity and a dividing proportion input by a user through a touch screen; dividing the initial sample set according to the dividing proportion to obtain a first sample set and a second sample set, wherein the ratio of the number of samples in the first sample set to the number of samples in the second sample set is matched with the dividing proportion; taking the first quantity as the quantity of clustering clusters in the clustering process, clustering the first sample set to obtain a first quantity of clustering clusters and the mass center of each clustering cluster, and taking each clustering cluster as an actual training sample set to obtain a first quantity of actual training sample set; and taking the second quantity as the quantity of the clustering clusters in the clustering process, clustering the second sample sets to obtain the second quantity of the clustering clusters and the mass center of each clustering cluster, and taking each clustering cluster as a test sample set to obtain the second quantity of the test sample sets.
Optionally, the coal quality analysis report is generated by: determining the distance between the characteristic spectrum and the mass center of each practical training sample set in the first practical training sample sets, and configuring weights for the corresponding practical coal quality analysis models according to the distances, wherein the distances and the weights are in inverse proportion; and generating a coal quality analysis report according to the weight of each actual coal quality analysis model and the first number of coal quality analysis results.
The invention provides a multi-index coal quality detection method, which is applied to the multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm, and comprises the following steps: displaying a plurality of candidate indexes, wherein the plurality of candidate indexes comprise ash content of coal, moisture content of coal and sulfur content of coal, and receiving at least one index selected by a user from the plurality of candidate indexes, and the at least one index forms a target index set; screening a plurality of coal quality analysis models from the candidate coal quality analysis model set according to the target index set, wherein the screened plurality of coal quality analysis models form a primary screening coal quality analysis model set; the method comprises the steps of displaying a sample set editor, wherein the sample set editor is used for receiving sample editing information input by a user and dividing an initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information; displaying a primary screening coal quality analysis model set and the historical prediction accuracy of each coal quality analysis model in the primary screening coal quality analysis model set, so that a user selects a first number of coal quality analysis models from the primary screening coal quality analysis model set according to the historical prediction accuracy, and the first number of coal quality analysis models form a target coal quality analysis model set; receiving first configuration information input by a user, and configuring a corresponding relation for a first number of coal quality analysis models and a first number of actual training sample sets according to the first configuration information so that each coal quality analysis model in the first number of coal quality analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one; respectively training a first number of coal quality analysis models by using corresponding actual training sample sets to obtain a first number of actual coal quality analysis models; analyzing the characteristic spectrum by using a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal analysis report according to the first quantity of coal analysis results.
The invention has the following beneficial effects:
1. The sodium light source is used as an excitation light source, and as the sodium light source is a non-radioactive source, the approval of links such as purchasing, installation, acceptance and the like is avoided, manpower and material resources are saved, and safety risks brought to operators are avoided; in addition, the user can configure the content such as the index to be detected, the number of the actual training sample set and the test sample set, the corresponding relation between the coal analysis model and the actual training sample set and the like through the touch screen, so that more personalized and flexible model configuration and report generation can be realized; in the process, by using an intelligent analysis chip and carrying a candidate coal quality analysis model set constructed based on different intelligent algorithms, the detection time is greatly shortened, and the data show that the multi-index coal quality analyzer can complete detection within 10 ms;
2. The first number of actual training sample sets and the second number of test sample sets are generated through clustering, so that the distance between samples in each actual training sample set or each test sample set is relatively close, namely, samples corresponding to coal belonging to the same kind are ensured. Therefore, the actual coal quality analysis model obtained through training of the actual training sample set can fully learn the mapping relation between the characteristic spectrum of the coal of the same kind and the index of the coal, and the accuracy of the output coal quality analysis result is not affected because of large difference of the characteristics of different kinds of coal. Furthermore, the accuracy of the coal quality analysis result is improved, and the generated coal quality analysis report is more accurate by adjusting the weight in the process of generating the coal quality analysis report. Specifically, as the distance between the characteristic spectrum and the mass center of each actual training sample set can represent the similarity degree between the characteristic spectrum and each actual training sample set, when the distance between the characteristic spectrum and the mass center of a certain actual training sample set is closer, the characteristic spectrum is more similar to the type of coal corresponding to the actual training sample set, so that the weight of an actual coal analysis model corresponding to the actual training sample set can be increased, and the generated coal analysis report is more accurate;
3. By acquiring the candidate coal quality analysis model set and the initial sample set from the cloud platform, data sharing among different multi-index coal quality analyzers in the target area can be realized, including sharing of the sample set and sharing of the coal quality analysis model, so that the overall efficiency is improved. In the process, the updating sample group is selected through variance, so that the actual coal quality analysis model can better learn the characteristics of the coal of the type, and the reason is that: the smaller the variance is, the more concentrated the samples are in the updated sample group, so that the characteristics of the corresponding kinds of coal can be better represented. It should be noted that, although the generalization capability of a single model is reduced to a certain extent by such processing, as the first number of actual coal quality analysis models exist and different actual coal quality analysis models are used for processing the characteristic spectrums of different types of coal, the defect of insufficient generalization capability of the single model can be overcome, and the overall accuracy of the multi-index coal quality analyzer is improved.
Drawings
The above and other features, advantages and aspects of embodiments of the present invention 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 schematic diagram of a multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm of the present invention;
FIG. 2 is a flow chart of the multi-index coal quality detection method of the present invention.
Detailed Description
The invention will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the invention have been illustrated in the accompanying drawings, it is to be understood that the invention 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 invention are for illustration purposes only and are not intended to limit the scope of the present invention.
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 invention and features of the embodiments may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like herein are merely used for distinguishing between different devices, modules, or units and not for limiting the order or interdependence of the functions performed by such 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 skilled in the art will appreciate that "one or more" is intended to be construed as "one or more" unless the context clearly indicates otherwise.
The names of messages or information interacted between the devices of the present invention are for illustrative purposes only and are not intended to limit the scope of such messages or information.
The invention will be described in detail below with reference to the drawings in connection with embodiments.
As shown in fig. 1, a schematic structural diagram of the multi-index coal quality analyzer based on the sodium light source and the intelligent algorithm of the present invention is shown. The multi-index coal quality analyzer comprises a sodium light source 101, a detector 102, an intelligent analysis chip 103 and a touch screen 104. Optionally, the multi-index coal quality analyzer further comprises a cabinet 105. It is to be understood that the configuration and shape of the cabinet 105 are exemplary and that a variety of configurations and shapes of cabinets may be provided as desired. The cabinet 105 may not be provided according to actual needs.
The sodium light source 101 is used for emitting electromagnetic waves to the coal to be tested so as to excite the coal to be tested to emit a characteristic spectrum. In practical applications, the sodium light source 101 may be a high-pressure sodium light source or a low-pressure sodium light source. A special coating is added outside the sodium light source 101, and the emitted electromagnetic wave is focused in a fixed range; the electromagnetic wave emitted by the sodium light source 101 is reflected, refracted, transmitted and diffracted for a plurality of times among coal particles, and in the process, organic molecules in the coal absorb waves with specific energy levels. Specifically, only a specific wave having energy that is an integer multiple of the energy level difference of electrons in the molecule is absorbed, thereby causing the molecule to vibrate, resulting in a characteristic absorption spectrum, i.e., a characteristic spectrum. Because the characteristic spectrum carries a large amount of information of coal, the information of coal quality and the like of the coal can be obtained by analyzing the characteristic spectrum.
As an example, as shown in fig. 1, the sodium light source 101 may be disposed above the cabinet 105 of the multi-index coal quality analyzer, and electromagnetic waves emitted by the sodium light source 101 may be emitted to the coal to be measured through an opening disposed above the cabinet 105, and the coal to be measured may be disposed on a bracket inside the cabinet 105.
Alternatively, the sodium light source 101 may comprise a plurality of sub-light sources, each of the plurality of sub-light sources having a different energy intensity. One or more sub-light sources can be used as excitation light sources according to actual needs.
As an example, the detector 102 may be disposed within the cabinet 105 opposite the sodium light source 101. The detector 102 is configured to detect the characteristic spectrum and send the characteristic spectrum to the intelligent analysis chip 103. The detector 102 is in communication connection with the intelligent analysis chip 103, and the detector 102 can communicate with the intelligent analysis chip 103 in a wired or wireless connection mode so as to realize data interaction between the two. For example, the detector 102 may include a CCD image sensor.
As an example, the intelligent analysis chip 103 may be disposed within the cabinet 105, for example, on an inner wall of the bottom of the cabinet 105. The intelligent analysis chip 103 is provided with a candidate coal quality analysis model set and an initial sample set, wherein each coal quality analysis model in the candidate coal quality analysis model set is constructed based on different intelligent algorithms. In practical applications, the intelligent analysis chip 103 may be an AI chip (artificial intelligence chip). Among them, intelligent algorithms include, but are not limited to: partial least squares regression algorithm (PLS), ridge regression algorithm, lasso regression algorithm, elastic network regression algorithm, etc. Each sample in the initial sample set comprises a sample characteristic spectrum and a sample coal quality analysis result corresponding to the sample characteristic spectrum. Sample coal quality analysis results include ash, moisture, and sulfur content.
On the basis, each coal analysis model in the candidate coal analysis model set is a model which is pre-trained, and the practical training time can be shortened by pre-training each coal analysis model, so that a coal analysis report can be quickly generated, and the analysis and detection efficiency is improved.
The touch screen 104 can be used to display a plurality of candidate metrics including coal ash, coal moisture, and coal sulfur. Based on this, the user may select at least one index from the multiple candidate indexes according to actual needs, and the touch screen 104 may receive the at least one index selected by the user and send the at least one index to the intelligent analysis chip. Wherein at least one index constitutes a target index set.
Based on this, the intelligent analysis chip 104 can screen a plurality of coal analysis models from the candidate coal analysis model set according to the target index set, and the screened plurality of coal analysis models form a primary screening coal analysis model set. Specifically, since the detectable indexes of each coal quality analysis model are different, for any one of the candidate coal quality analysis models, if the detectable index of the candidate coal quality analysis model matches with any one of the target index sets, the coal quality analysis model is selected, thereby selecting a plurality of coal quality analysis models. The screened multiple coal quality analysis models form a primary screening coal quality analysis model set.
The touch screen 104 is further configured to display a sample set editor, where the sample set editor is configured to receive sample editing information input by a user, and divide an initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information. Therefore, the user can flexibly configure the number of the actual training sample set and the test sample set according to actual needs.
As an example, the sample editing information includes a division ratio, on the basis of which an initial sample set is divided according to the division ratio, to obtain a first sample set and a second sample set, and a ratio between the number of samples in the first sample set and the number of samples in the second sample set is matched with the division ratio; the first sample set is divided equally into a first number of actual training sample sets and the second sample set is divided equally into a second number of test sample sets.
On this basis, the touch screen 104 is further configured to display a set of preliminary screening coal quality analysis models and a historical prediction accuracy of each coal quality analysis model in the set of preliminary screening coal quality analysis models, so that a user selects a first number of coal quality analysis models from the set of preliminary screening coal quality analysis models according to the historical prediction accuracy, where the first number of coal quality analysis models form a set of target coal quality analysis models.
Wherein the historical prediction accuracy of each coal quality analysis model may be an average prediction accuracy of the coal quality analysis model over a period of time (e.g., the past week). Therefore, the user can preferentially select the coal analysis model with larger historical prediction accuracy.
The touch screen 104 is further configured to receive first configuration information input by a user and send the first configuration information to the intelligent analysis chip, where the intelligent analysis chip is configured to configure a corresponding relationship for the first number of coal quality analysis models and the first number of actual training sample sets according to the first configuration information, so that each coal quality analysis model in the first number of coal quality analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one. The first configuration information may represent a correspondence between the coal quality analysis model and the actual training sample set. For example, the first configuration information may include a plurality of tuples, each of which is composed of a number of the coal quality analysis model and a number of the actual training sample set corresponding to each other.
The intelligent analysis chip 103 is further configured to respectively train the first number of coal quality analysis models by using the corresponding actual training sample set, so as to obtain the first number of actual coal quality analysis models; analyzing the characteristic spectrum by using a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal analysis report according to the first quantity of coal analysis results.
Because the corresponding relation is configured, each coal quality analysis model can be trained by using the actual training sample set corresponding to each coal quality analysis model, and an actual coal quality analysis model is obtained. And further, respectively inputting the characteristic spectrums into a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results. In the training process, the peak value of the sample characteristic spectrum of the sample in the actual training sample set is used as the input of the actual coal quality analysis model.
On this basis, the results of the first number of coal analyses are averaged and the average is reported as a coal analysis.
Optionally, a preprocessing method set is further deployed on the intelligent analysis chip, the touch screen is further used for displaying the preprocessing method set, receiving a first number of preprocessing methods selected by a user from the preprocessing method set, receiving second configuration information input by the user and sending the second configuration information to the intelligent analysis chip, and the intelligent analysis chip is used for configuring a corresponding relation between the first number of preprocessing methods and the first number of actual training sample sets according to the second configuration information, so that each preprocessing method in the first number of preprocessing methods corresponds to each actual training sample set in the first number of actual training sample sets one by one; the intelligent analysis chip is further used for preprocessing the first number of actual training sample sets by using a first number of preprocessing methods before the corresponding actual training sample sets are used for respectively training the first number of coal analysis models, so as to obtain the processed first number of actual training sample sets, and training the first number of coal analysis models by using the processed first number of actual training sample sets. Among the pretreatment methods in the pretreatment method set, but not limited to: mean centering, normalization, first derivative and SG smoothing, standard normal transformation SNV, and the like.
In some embodiments, the sodium light source is used as the excitation light source, and as the sodium light source is a non-radioactive source, the approval of links such as purchasing, installation, acceptance and the like is avoided, manpower and material resources are saved, and safety risks brought to operators are avoided. In addition, the user can configure the content such as the index to be detected, the number of the actual training sample set and the test sample set, the corresponding relation between the coal analysis model and the actual training sample set and the like through the touch screen, so that more personalized and flexible model configuration and report generation can be realized.
In some embodiments, in order to further solve the second technical problem described in the background section, that is, "in the actual analysis and inspection process, the production place and the type of coal are different, but the difference of the characteristics of different types of coal is larger, so when the type of coal corresponding to the training sample is different from the type of coal in the actual inspection, the accuracy of the coal analysis result output by the coal analysis model and the finally generated coal analysis report is reduced", in some embodiments of the present invention, the first number of actual training sample sets and the second number of test sample sets are obtained by dividing the following steps:
Step one, receiving a first quantity, a second quantity and a dividing proportion input by a user through a touch screen;
Dividing the initial sample set according to the dividing proportion to obtain a first sample set and a second sample set, wherein the ratio of the number of samples in the first sample set to the number of samples in the second sample set is matched with the dividing proportion;
Step three, using the first quantity as the quantity of clustering clusters in the clustering process, clustering the first sample sets to obtain a first quantity of clustering clusters and the mass center of each clustering cluster, and using each clustering cluster as an actual training sample set to obtain a first quantity of actual training sample sets;
and step four, taking the second quantity as the quantity of clustering clusters in the clustering process, clustering the second sample sets to obtain the second quantity of clustering clusters and the mass center of each clustering cluster, and taking each clustering cluster as a test sample set to obtain the second quantity of test sample sets.
On the basis, the coal quality analysis report is generated by the following steps:
Firstly, determining the distance between a characteristic spectrum and the mass center of each practical training sample set in a first number of practical training sample sets, and configuring weights for corresponding practical coal quality analysis models according to the distances, wherein the distances and the weights are in inverse proportion;
And secondly, generating a coal quality analysis report according to the weight of each actual coal quality analysis model and the first number of coal quality analysis results.
In these embodiments, the first number of actual training sample sets and the second number of test sample sets are generated by clustering, thereby ensuring that the distance between the samples in each actual training sample set and each test sample set is relatively close, i.e., samples corresponding to coal belonging to the same class. Therefore, the actual coal quality analysis model obtained through training of the actual training sample set can fully learn the mapping relation between the characteristic spectrum of the coal of the same kind and the index of the coal, and the accuracy of the output coal quality analysis result is not affected because of large difference of the characteristics of different kinds of coal. Furthermore, the accuracy of the coal quality analysis result is improved, and the generated coal quality analysis report is more accurate by adjusting the weight in the process of generating the coal quality analysis report. Specifically, as the distance between the characteristic spectrum and the centroid of each actual training sample set can represent the similarity degree between the characteristic spectrum and each actual training sample set, when the distance between the characteristic spectrum and the centroid of a certain actual training sample set is closer, the characteristic spectrum is more similar to the type of coal corresponding to the actual training sample set, so that the weight of an actual coal analysis model corresponding to the actual training sample set can be increased, and the generated coal analysis report is more accurate. In the process, by using the intelligent analysis chip and carrying the candidate coal quality analysis model set constructed based on different intelligent algorithms, the detection time is greatly shortened, and the data show that the multi-index coal quality analyzer can complete detection within 10 ms.
In some embodiments, in order to further solve the third technical problem described in the background section, that is, "in the actual analysis and inspection process, the existing multi-index coal analyzers are independently operated, so that each multi-index coal analyzer needs to independently complete model training, and different multi-index coal analyzers are difficult to realize data sharing, resulting in low efficiency", according to the present invention, in some embodiments, the multi-index coal analyzers based on a sodium light source and an intelligent algorithm are in communication connection with a cloud platform, and the cloud platform can communicate with and interact data with a plurality of multi-index coal analyzers in a target area (for example, the same factory).
On the basis, in the deployment stage, the intelligent analysis chip can acquire a candidate coal quality analysis model set and an initial sample set from the cloud platform, wherein the initial sample set comprises three parts, the first part is a sample to be grouped, the second part is a grouped sample, and the third part is a plurality of initial test sample sets. The grouped samples comprise a plurality of initial sample groups, each initial sample group corresponds to a type of coal, wherein the grouped samples and the plurality of initial test sample groups are generated in a detection process by the rest multi-index coal quality analyzers in the target area and are uploaded to the cloud platform. The candidate coal quality analysis model set comprises two parts, wherein one part is a plurality of original coal quality analysis models, the other part is a plurality of trained initial coal quality analysis models, and the plurality of trained initial coal quality analysis models are generated in the detection process of the rest multi-index coal quality analyzers in the target area and uploaded to the cloud platform.
On this basis, the intelligent analysis chip can execute the following processing steps:
First, sample information is acquired, wherein the sample information comprises the number of samples in the samples to be grouped, the number of initial sample groups included in the grouped samples, and the types of coal corresponding to each initial sample group.
Second, determining whether the number of the initial sample groups is greater than or equal to the first number included in the sample editing information; if the number of the initial sample groups is greater than or equal to the first number included in the sample editing information, calculating the distance between each sample in the samples to be grouped and the centroid of each initial sample group, adding the sample into the initial sample group with the minimum corresponding distance until all samples in the samples to be grouped are traversed, and obtaining a plurality of updated sample groups.
Third, a variance of each of the plurality of updated sample groups is determined, and a first number of updated sample groups is selected as a first number of actual training sample sets in order of the variance from small to large.
Fourth, if the number of the initial sample groups is smaller than the first number included in the sample editing information, a preset distance threshold is obtained, for each sample in the samples to be grouped, the distance between the sample and the mass center of each initial sample group is calculated, the initial sample group with the smallest corresponding distance is determined to be a target initial sample group, the distance between the initial sample group and the sample is determined to be a target distance, and if the target distance is smaller than the distance threshold, the sample is added into the target initial sample group, and an updated sample group is obtained; if the target distance is greater than the distance threshold, adding the sample into the newly built sample group; after multiple iterations, the obtained updated sample set and the newly-built sample set jointly form the first number of actual training sample sets. In practice, the distance threshold may be continuously adjusted to adjust the number of newly created sample sets until a first number of actual training sample sets are obtained. The distance threshold is reduced when the sum of the number of newly built sample groups and the number of updated sample groups is less than the first number.
Fifthly, if the number of the initial test sample sets is smaller than the second number included in the sample editing information, dividing one or more initial test sample sets in the plurality of initial test sample sets until a second number of test sample sets are obtained; if the number of the initial test sample sets is greater than the second number included in the sample editing information, merging one or more initial test sample sets in the plurality of initial test sample sets until a second number of test sample sets is obtained;
Sixthly, determining whether a plurality of trained initial coal quality analysis models in the candidate coal quality analysis model set contain all indexes in the target index set, and if so, screening a plurality of coal quality analysis models from the plurality of trained initial coal quality analysis models; determining whether the number of the types of the coals corresponding to the screened multiple coal quality analysis models is greater than or equal to a preset number, if so, forming a primary screening coal quality analysis model set by the screened multiple coal quality analysis models, and if not, selecting one or more coal quality analysis models from the original coal quality analysis models, wherein the selected one or more coal quality analysis models and the screened multiple coal quality analysis models form the primary screening coal quality analysis model set.
In some embodiments, by acquiring the candidate coal quality analysis model set and the initial sample set from the cloud platform, data sharing among different multi-index coal quality analyzers in the target area can be realized, including sharing of the sample set and sharing of the coal quality analysis model, so that overall efficiency is improved. In the process, the updating sample group is selected through variance, so that the actual coal quality analysis model can better learn the characteristics of the coal of the type, and the reason is that: the smaller the variance is, the more concentrated the samples are in the updated sample group, so that the characteristics of the corresponding kinds of coal can be better represented. It should be noted that, although the generalization capability of a single model is reduced to a certain extent by such processing, as the first number of actual coal quality analysis models exist and different actual coal quality analysis models are used for processing the characteristic spectrums of different types of coal, the defect of insufficient generalization capability of the single model can be overcome, and the overall accuracy of the multi-index coal quality analyzer is improved.
Referring to fig. 2, a flow chart of the multi-index coal quality detection method of the present invention is shown. The multi-index coal quality detection method comprises the following steps:
step 201, displaying a plurality of candidate indexes, wherein the plurality of candidate indexes comprise coal ash, coal moisture and coal sulfur, and receiving at least one index selected by a user from the plurality of candidate indexes, and the at least one index forms a target index set;
Step 202, screening a plurality of coal quality analysis models from a candidate coal quality analysis model set according to a target index set, wherein the screened plurality of coal quality analysis models form a primary screening coal quality analysis model set;
step 203, displaying a sample set editor, wherein the sample set editor is used for receiving sample editing information input by a user and dividing an initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information;
Step 204, displaying a primary screening coal quality analysis model set and a historical prediction accuracy of each coal quality analysis model in the primary screening coal quality analysis model set, so that a user selects a first number of coal quality analysis models from the primary screening coal quality analysis model set according to the historical prediction accuracy, wherein the first number of coal quality analysis models form a target coal quality analysis model set;
Step 205, receiving first configuration information input by a user, and configuring a corresponding relation for a first number of coal quality analysis models and a first number of actual training sample sets according to the first configuration information, so that each coal quality analysis model in the first number of coal quality analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one;
step 206, training the first number of coal quality analysis models by using the corresponding actual training sample sets respectively to obtain the first number of actual coal quality analysis models; analyzing the characteristic spectrum by using a first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal analysis report according to the first quantity of coal analysis results.
In some embodiments, the user may configure, through the touch screen, the content such as the index to be detected, the number of the actual training sample set and the test sample set, the correspondence between the coal analysis model and the actual training sample set, and so on, thereby realizing more personalized and flexible model configuration and report generation.
The above description is only illustrative of the few preferred embodiments of the present invention and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the invention referred to in the present invention is not limited to the specific combinations of the technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the inventive concept described above. Such as the above-mentioned features and the technical features disclosed in the present invention (but not limited to) having similar functions are replaced with each other.
Claims (6)
1. Multi-index coal quality analyzer based on sodium light source and intelligent algorithm, its characterized in that includes:
The sodium light source is used for emitting electromagnetic waves to the coal to be tested so as to excite the coal to be tested to emit a characteristic spectrum;
the detector is used for detecting the characteristic spectrum and sending the characteristic spectrum to the intelligent analysis chip;
The intelligent analysis chip is provided with a candidate coal quality analysis model set and an initial sample set, wherein each coal quality analysis model in the candidate coal quality analysis model set is constructed based on different intelligent algorithms;
The touch screen is used for displaying a plurality of candidate indexes, wherein the candidate indexes comprise coal ash, coal moisture and coal sulfur, at least one index selected by a user from the candidate indexes is received, the at least one index forms a target index set, and the touch screen is used for sending the target index set to the intelligent analysis chip; the intelligent analysis chip is used for screening a plurality of coal analysis models from the candidate coal analysis model set according to the target index set, and the screened plurality of coal analysis models form a primary screening coal analysis model set;
The touch screen is further used for displaying a sample set editor, and the sample set editor is used for receiving sample editing information input by a user and dividing the initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information;
The touch screen is further used for displaying the primary screening coal quality analysis model set and the historical prediction accuracy of each coal quality analysis model in the primary screening coal quality analysis model set, so that a user selects a first number of coal quality analysis models from the primary screening coal quality analysis model set according to the historical prediction accuracy, and the first number of coal quality analysis models form a target coal quality analysis model set;
the touch screen is further used for receiving first configuration information input by a user and sending the first configuration information to the intelligent analysis chip, and the intelligent analysis chip is used for configuring corresponding relations for the first number of coal quality analysis models and the first number of actual training sample sets according to the first configuration information so that each coal quality analysis model in the first number of coal quality analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one;
the intelligent analysis chip is further used for training the first number of coal quality analysis models by using a corresponding actual training sample set to obtain a first number of actual coal quality analysis models; respectively analyzing the characteristic spectrum by using the first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; generating a coal quality analysis report according to the first number of coal quality analysis results;
The intelligent analysis chip is in communication connection with the cloud platform, the candidate coal quality analysis model set and the initial sample set are obtained from the cloud platform in a deployment stage of the intelligent analysis chip, wherein the initial sample set comprises three parts, the first part is a sample to be grouped, the second part is a grouped sample, and the third part is a plurality of initial test sample sets; the grouped samples comprise a plurality of initial sample groups, each initial sample group corresponds to a type of coal, wherein the grouped samples and the plurality of initial test sample groups are generated in a detection process by other multi-index coal quality analyzers in a target area and are uploaded to a cloud platform; the candidate coal quality analysis model set comprises two parts, wherein one part is a plurality of original coal quality analysis models, the other part is a plurality of trained initial coal quality analysis models, and the plurality of trained initial coal quality analysis models are generated in the detection process of the rest multi-index coal quality analyzers in the target area and uploaded to the cloud platform;
the intelligent analysis chip is also used for executing the following processing steps:
Firstly, acquiring sample information, wherein the sample information comprises the number of samples in samples to be grouped, the number of initial sample groups included in the grouped samples and the types of coal corresponding to each initial sample group;
Second, determining whether the number of the initial sample groups is greater than or equal to the first number included in the sample editing information; if the number of the initial sample groups is greater than or equal to the first number included in the sample editing information, calculating the distance between each sample in the samples to be grouped and the centroid of each initial sample group, adding the sample into the initial sample group with the minimum corresponding distance until traversing all samples in the samples to be grouped to obtain a plurality of updated sample groups;
third, determining a variance of each of the plurality of updated sample groups, and selecting a first number of updated sample groups as a first number of actual training sample sets according to an order in which the variances are from small to large;
Fourth, if the number of the initial sample groups is smaller than the first number included in the sample editing information, a preset distance threshold is obtained, for each sample in the samples to be grouped, the distance between the sample and the mass center of each initial sample group is calculated, the initial sample group with the smallest corresponding distance is determined to be a target initial sample group, the distance between the initial sample group and the sample is determined to be a target distance, and if the target distance is smaller than the distance threshold, the sample is added into the target initial sample group, and an updated sample group is obtained; if the target distance is greater than the distance threshold, adding the sample into the newly built sample group; after multiple iterations, the obtained updated sample group and the newly-built sample group jointly form the first number of actual training sample sets; in the process, the distance threshold value is continuously adjusted to adjust the number of newly built sample groups until a first number of actual training sample sets are obtained; wherein the distance threshold is reduced when the sum of the number of newly built sample groups and the number of updated sample groups is less than the first number;
Fifthly, if the number of the initial test sample sets is smaller than the second number included in the sample editing information, dividing one or more initial test sample sets in the plurality of initial test sample sets until a second number of test sample sets are obtained; if the number of the initial test sample sets is greater than the second number included in the sample editing information, merging one or more initial test sample sets in the plurality of initial test sample sets until a second number of test sample sets is obtained;
Sixthly, determining whether a plurality of trained initial coal quality analysis models in the candidate coal quality analysis model set contain all indexes in the target index set, and if so, screening a plurality of coal quality analysis models from the plurality of trained initial coal quality analysis models; determining whether the number of the types of the coals corresponding to the screened multiple coal quality analysis models is greater than or equal to a preset number, if so, forming a primary screening coal quality analysis model set by the screened multiple coal quality analysis models, and if not, selecting one or more coal quality analysis models from the original coal quality analysis models, wherein the selected one or more coal quality analysis models and the screened multiple coal quality analysis models form the primary screening coal quality analysis model set.
2. The multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm according to claim 1, wherein a pretreatment method set is further deployed on the intelligent analysis chip, the touch screen is further used for displaying the pretreatment method set, receiving a first number of pretreatment methods selected by a user from the pretreatment method set, receiving second configuration information input by the user and sending the second configuration information to the intelligent analysis chip, and the intelligent analysis chip is used for configuring a corresponding relation for the first number of pretreatment methods and the first number of actual training sample sets according to the second configuration information so that each pretreatment method in the first number of pretreatment methods corresponds to each actual training sample set in the first number of actual training sample sets one by one;
The intelligent analysis chip is further used for preprocessing the first number of actual training sample sets by the first number of preprocessing methods before the corresponding actual training sample sets are used for respectively training the first number of coal analysis models, so as to obtain the processed first number of actual training sample sets, and training the first number of coal analysis models by the processed first number of actual training sample sets.
3. The multi-index coal quality analyzer based on a sodium light source and intelligent algorithm of claim 2, wherein the sodium light source comprises a plurality of sub-light sources, each of the plurality of sub-light sources having a different energy intensity.
4. The multi-index coal quality analyzer based on a sodium light source and an intelligent algorithm according to claim 3, wherein the first number of actual training sample sets and the second number of test sample sets are divided by:
receiving a first quantity, a second quantity and a dividing proportion input by a user through the touch screen;
Dividing the initial sample set according to the dividing proportion to obtain a first sample set and a second sample set, wherein the ratio of the number of samples in the first sample set to the number of samples in the second sample set is matched with the dividing proportion;
Taking the first quantity as the quantity of clustering clusters in the clustering process, clustering the first sample sets to obtain a first quantity of clustering clusters and the mass center of each clustering cluster, and taking each clustering cluster as an actual training sample set to obtain the first quantity of actual training sample sets;
And taking the second quantity as the quantity of clustering clusters in the clustering process, clustering the second sample sets to obtain a second quantity of clustering clusters and the mass center of each clustering cluster, and taking each clustering cluster as a test sample set to obtain the second quantity of test sample sets.
5. The sodium light source and intelligent algorithm-based multi-index coal quality analyzer of claim 4, wherein the coal quality analysis report is generated by:
Determining the distance between the characteristic spectrum and the mass center of each practical training sample set in the first number of practical training sample sets, and configuring weights for the corresponding practical coal quality analysis models according to the distance, wherein the distance and the weights are in an inverse proportion relation;
and generating the coal analysis report according to the weight of each actual coal analysis model and the first number of coal analysis results.
6. The multi-index coal quality detection method applied to the sodium light source and intelligent algorithm-based multi-index coal quality analyzer as claimed in claim 1 is characterized by comprising the following steps:
Displaying a plurality of candidate indexes, wherein the plurality of candidate indexes comprise coal ash, coal moisture and coal sulfur, and receiving at least one index selected by a user from the plurality of candidate indexes, and the at least one index forms a target index set;
screening a plurality of coal quality analysis models from the candidate coal quality analysis model set according to the target index set, wherein the screened plurality of coal quality analysis models form a primary screening coal quality analysis model set;
The method comprises the steps of displaying a sample set editor, wherein the sample set editor is used for receiving sample editing information input by a user and dividing an initial sample set into a first number of actual training sample sets and a second number of test sample sets according to the sample editing information;
Displaying the primary screening coal quality analysis model set and the historical prediction accuracy of each coal quality analysis model in the primary screening coal quality analysis model set, so that a user selects a first number of coal quality analysis models from the primary screening coal quality analysis model set according to the historical prediction accuracy, and the first number of coal quality analysis models form a target coal quality analysis model set;
Receiving first configuration information input by a user, and configuring a corresponding relation for the first number of coal analysis models and the first number of actual training sample sets according to the first configuration information so that each coal analysis model in the first number of coal analysis models corresponds to each actual training sample set in the first number of actual training sample sets one by one;
Respectively training the first number of coal quality analysis models by using corresponding actual training sample sets to obtain first number of actual coal quality analysis models; respectively analyzing the characteristic spectrum by using the first number of actual coal quality analysis models to obtain a first number of coal quality analysis results; and generating a coal quality analysis report according to the first quantity of coal quality analysis results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410172597.1A CN117723514B (en) | 2024-02-07 | 2024-02-07 | Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410172597.1A CN117723514B (en) | 2024-02-07 | 2024-02-07 | Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117723514A CN117723514A (en) | 2024-03-19 |
| CN117723514B true CN117723514B (en) | 2024-05-17 |
Family
ID=90205660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410172597.1A Active CN117723514B (en) | 2024-02-07 | 2024-02-07 | Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117723514B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102358505A (en) * | 2011-06-28 | 2012-02-22 | 太原理工大学 | Conveyor belt longitudinal-tearing online monitoring early warning device |
| CN103884670A (en) * | 2014-03-13 | 2014-06-25 | 西安交通大学 | Smoke component quantitative analysis method based on near infrared spectrum |
| CN106650825A (en) * | 2016-12-31 | 2017-05-10 | 中国科学技术大学 | Automotive exhaust emission data fusion system |
| CN107730473A (en) * | 2017-11-03 | 2018-02-23 | 中国矿业大学 | A kind of underground coal mine image processing method based on deep neural network |
| CN108133232A (en) * | 2017-12-15 | 2018-06-08 | 南京航空航天大学 | A kind of Radar High Range Resolution target identification method based on statistics dictionary learning |
| CN114354654A (en) * | 2022-01-07 | 2022-04-15 | 中国矿业大学 | DW-KNN-based rapid nondestructive detection method for coal moisture content |
| CN114414521A (en) * | 2022-01-20 | 2022-04-29 | 淮北师范大学 | Milk main component measuring method based on infrared multispectral sensor |
| CN114636687A (en) * | 2022-03-14 | 2022-06-17 | 国能神皖能源有限责任公司 | Small sample coal quality characteristic analysis system and method based on deep migration learning |
| CN115964612A (en) * | 2023-01-17 | 2023-04-14 | 江苏地质矿产设计研究院(中国煤炭地质总局检测中心) | Coal bed resource estimation method based on triangulation algorithm |
| CN116256803A (en) * | 2023-03-09 | 2023-06-13 | 中国矿业大学 | Coal mine microseismic region positioning method integrating mining information and geological information |
| CN116559095A (en) * | 2023-05-09 | 2023-08-08 | 北京易兴元石化科技有限公司 | Table type coal quality analysis spectrometer and coal quality analysis method |
| CN116792151A (en) * | 2023-03-23 | 2023-09-22 | 中国矿业大学 | Deep learning-based intelligent advanced early warning method for coal and gas outburst |
| CN117216558A (en) * | 2023-09-01 | 2023-12-12 | 广东能源集团科学技术研究院有限公司 | Coal quality analysis model training method and device, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107818313B (en) * | 2017-11-20 | 2019-05-14 | 腾讯科技（深圳）有限公司 | Vivo identification method, device and storage medium |
-
2024
- 2024-02-07 CN CN202410172597.1A patent/CN117723514B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102358505A (en) * | 2011-06-28 | 2012-02-22 | 太原理工大学 | Conveyor belt longitudinal-tearing online monitoring early warning device |
| CN103884670A (en) * | 2014-03-13 | 2014-06-25 | 西安交通大学 | Smoke component quantitative analysis method based on near infrared spectrum |
| CN106650825A (en) * | 2016-12-31 | 2017-05-10 | 中国科学技术大学 | Automotive exhaust emission data fusion system |
| CN107730473A (en) * | 2017-11-03 | 2018-02-23 | 中国矿业大学 | A kind of underground coal mine image processing method based on deep neural network |
| CN108133232A (en) * | 2017-12-15 | 2018-06-08 | 南京航空航天大学 | A kind of Radar High Range Resolution target identification method based on statistics dictionary learning |
| CN114354654A (en) * | 2022-01-07 | 2022-04-15 | 中国矿业大学 | DW-KNN-based rapid nondestructive detection method for coal moisture content |
| CN114414521A (en) * | 2022-01-20 | 2022-04-29 | 淮北师范大学 | Milk main component measuring method based on infrared multispectral sensor |
| CN114636687A (en) * | 2022-03-14 | 2022-06-17 | 国能神皖能源有限责任公司 | Small sample coal quality characteristic analysis system and method based on deep migration learning |
| CN115964612A (en) * | 2023-01-17 | 2023-04-14 | 江苏地质矿产设计研究院(中国煤炭地质总局检测中心) | Coal bed resource estimation method based on triangulation algorithm |
| CN116256803A (en) * | 2023-03-09 | 2023-06-13 | 中国矿业大学 | Coal mine microseismic region positioning method integrating mining information and geological information |
| CN116792151A (en) * | 2023-03-23 | 2023-09-22 | 中国矿业大学 | Deep learning-based intelligent advanced early warning method for coal and gas outburst |
| CN116559095A (en) * | 2023-05-09 | 2023-08-08 | 北京易兴元石化科技有限公司 | Table type coal quality analysis spectrometer and coal quality analysis method |
| CN117216558A (en) * | 2023-09-01 | 2023-12-12 | 广东能源集团科学技术研究院有限公司 | Coal quality analysis model training method and device, electronic equipment and storage medium |
Non-Patent Citations (6)
| Title |
|---|
| Approach to simulation optimization of time-varying parameters system based on neural network;Wu, S;《Systems Engineering and Electronics》;20240111;472-80 * |
| MATLAB实现BP神经网络的煤炭需求预测;余良晖;《煤炭经济研究》;20071015(第10期);34-36 * |
| Research on transportation organization mode of just-in-time block train for regional circulation operation based on coal storage and distribution base and locomotive circulation mode;Zhu Liang;《China Railway Science》;20170701;第38卷(第4期);124-31 * |
| 张少波 ; .煤矿能耗现状分析及节能技术发展方向.煤炭科学技术.2009,(第05期),1-4. * |
| 煤矿能耗现状分析及节能技术发展方向;张少波;;煤炭科学技术;20090525(第05期);1-4 * |
| 量子化学在煤微观结构研究中的应用进展;蒙雅莹;《山东化工》;20210508;第50卷(第9期);66-67+69 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117723514A (en) | 2024-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aartsen et al. | Development of a general analysis and unfolding scheme and its application to measure the energy spectrum of atmospheric neutrinos with IceCube: IceCube Collaboration | |
| CN106815643B (en) | Infrared spectroscopy Model Transfer method based on random forest transfer learning | |
| Graur et al. | A unified explanation for the supernova rate-galaxy mass dependence based on supernovae detected in Sloan galaxy spectra | |
| Zaritsky et al. | Spectral classification of galaxies along the hubble sequence | |
| Novak et al. | Constraints on submicrojansky radio number counts based on evolving VLA-COSMOS luminosity functions | |
| Douglas | Capturing geographically-varying uncertainty in earthquake ground motion models or what we think we know may change | |
| CN115187527B (en) | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum | |
| CN101923039A (en) | The method of one or more parameters of control flow cytometry type measuring system | |
| JP2010520471A (en) | Ensemble method and apparatus for classification of materials and quantification of components of mixtures | |
| CN103344600A (en) | Characteristic wavelength selecting method for near infrared spectrum in ant colony optimization algorithm | |
| CN111428561A (en) | Broadband electromagnetic interference source identification method and system, storage medium and equipment | |
| CN105335379B (en) | The method and apparatus to sort to the combination of mutation, test case, random seed in mutation test | |
| KR20210134077A (en) | Die screening using inline defect information | |
| Boudart et al. | Machine learning algorithm for minute-long burst searches | |
| CN117723514B (en) | Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method | |
| Dinsmore et al. | Luminosity functions consistent with a pulsar-dominated Galactic Center excess | |
| CN117056680A (en) | Data noise reduction and signal detection method, device and system and storage medium | |
| CN105654174A (en) | System and method for prediction | |
| Krumholz et al. | SLUG IV: a novel forward-modelling method to derive the demographics of star clusters | |
| Bentley et al. | Random Forest Model of Ultralow‐Frequency Magnetospheric Wave Power | |
| CN110031416B (en) | Gas concentration detection device and method | |
| CN114858958B (en) | Method and device for analyzing mass spectrum data in quality evaluation and storage medium | |
| CN109239006A (en) | A kind of substance identification based on humidity compensation model, device and storage medium | |
| CN113642629B (en) | Visualization method and device for improving reliability of spectroscopy analysis based on random forest | |
| TW202142878A (en) | A system and a method for obtaining a performance metric of a device under test based on one or more nearfield measurement results |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |