CN114354654B - DW-KNN-based rapid nondestructive testing method for coal moisture content - Google Patents
DW-KNN-based rapid nondestructive testing method for coal moisture content Download PDFInfo
- Publication number
- CN114354654B CN114354654B CN202210016888.2A CN202210016888A CN114354654B CN 114354654 B CN114354654 B CN 114354654B CN 202210016888 A CN202210016888 A CN 202210016888A CN 114354654 B CN114354654 B CN 114354654B
- Authority
- CN
- China
- Prior art keywords
- coal
- sample
- microwave
- moisture content
- training
- 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
Images
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention relates to a DW-KNN-based rapid nondestructive detection method for coal moisture content. Which comprises the following steps: step 1, constructing a coal moisture content detection model based on DW-KNN; when a DW-KNN-based coal moisture content detection model is constructed, the manufactured training data set comprises a coal training sample moisture content label of a coal training sample and coal training sample microwave S parameter spectrum information of the coal training sample under the coal training sample moisture content label; and 2, providing a coal sample to be detected for the moisture content, acquiring coal sample microwave S parameter spectrum information of which the coal sample is consistent with the coal training sample microwave S parameter spectrum information, and processing the acquired coal sample microwave S parameter spectrum information by using a DW-KNN-based coal moisture content detection model to obtain and output the moisture content of the coal sample. The invention realizes the detection of the moisture content of the coal under the condition of no damage, and improves the detection precision and the robustness.
Description
Technical Field
The invention relates to a coal moisture content rapid nondestructive testing method, in particular to a DW-KNN-based coal moisture content rapid nondestructive testing method.
Background
Moisture is one of four basic indexes (moisture, ash content, volatile matter and fixed carbon) for measuring the economic value of coal. In the coal coking and combustion process, the problems of low coal utilization efficiency, environmental pollution, energy waste and the like can be caused by too high or too low coal moisture content, so that the rapid and accurate detection of the coal moisture content has important significance.
The measurement method of the coal moisture content mainly comprises a direct measurement method and an indirect measurement method. The direct measurement method is a standard weight method, commonly called weight loss method, and mainly comprises a nitrogen drying method and an air drying method. The standard gravimetric method is a laboratory measurement method, and can obtain higher measurement precision; however, this method is time-consuming and requires the original properties of the sample to be destroyed.
Conventional indirect measurement methods mainly include neutron methods, conductivity methods, capacitance methods, infrared reflection methods, microwave methods, and the like. The principle of measuring hydrogen in coal by a neutron method is based on measuring the content of hydrogen in a sample. However, in addition to water, many chemical impurities in coal also contain hydrogen. Furthermore, the instruments are expensive and there is a risk of using radioactive neutron sources. The conductivity method and the capacitance method are influenced by various factors such as temperature, density, electrolyte content and the like, and have poor precision under the condition of no compensation strategy. The infrared reflection method has good effect on detecting the moisture content of the mixed liquid sample. For the detection of solids, only surface moisture (penetration depth within microns) of small particle samples can be detected. The microwave method has the advantages of safety, stability, no damage, non-contact and the like, and the most common method is a free space transmission measurement method. This method usually selects only a single point frequency or discrete frequency band of microwaves as a calibration signal, which results in signal loss that is closely related to moisture content; and the fitting method is simple, the detection precision is low, and the robustness is not high.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a DW-KNN-based coal moisture content rapid nondestructive detection method, which realizes the detection of the coal moisture content under the nondestructive condition and improves the detection precision and the robustness.
According to the technical scheme provided by the invention, the DW-KNN-based coal moisture content rapid nondestructive detection method comprises the following steps:
step 1, constructing a DW-KNN-based coal moisture content detection model, wherein when the DW-KNN-based coal moisture content detection model is constructed, a manufactured training data set comprises a coal training sample moisture content label of a coal training sample and coal training sample microwave S parameter spectrum information of the coal training sample under the coal training sample moisture content label, and the manufactured training data set is used for training the DW-KNN model to construct and obtain the DW-KNN-based coal moisture content detection model;
and 2, providing a coal sample to be detected for water content, acquiring coal sample microwave S parameter spectrum information of which the coal sample is consistent with the coal training sample microwave S parameter spectrum information, and processing the acquired coal sample microwave S parameter spectrum information by using the DW-KNN-based coal water content detection model to obtain and output the water content of the coal sample.
In step 1, when a training data set is manufactured, the method specifically comprises the following steps:
step 1.1, collecting raw coal samples in different areas, screening all the collected raw coal samples, uniformly and flatly paving the screened raw coal samples in corresponding carriers, and drying the raw coal samples in the carriers to obtain experimental coal samples in different areas;
step 1.2, carrying out sealing cooling, water spraying stirring and sealing standing treatment on the obtained experimental coal samples in different areas in sequence to obtain coal training samples under a preset coal training sample moisture content label;
and step 1.3, performing microwave test on the obtained coal training samples to obtain coal sample microwave S parameter spectrum information of the coal training samples in corresponding areas under the coal training sample moisture content labels.
In step 1.2, when water is sprayed and stirred, the water spraying quality determined according to the preset coal training sample water content label for any experimental coal sample is as follows:
wherein M is add Is the mass of the water spray, M c&m Is the mass of the mixture of the experimental coal sample and water, MC before Is the moisture content, MC, of the experimental coal sample after Is a coal sample moisture content label of a coal training sample.
In step 1.3, when the coal training samples are subjected to microwave testing, spectrum data of corresponding frequency points at the frequency of 8.05 GHz-12.01 GHz is acquired and obtained for any coal training sample to obtain coal sample microwave S parameter spectrum information of the coal training samples, wherein all coal sample microwave S parameter spectrum data of one coal training sample form a microwave S parameter spectrum initial curve.
Preprocessing microwave S parameter spectrum initial curves of all coal training samples from the same area to obtain coal sample microwave S parameter spectrum information of the coal training samples under a coal training sample moisture content label after preprocessing, wherein the preprocessing comprises the following steps:
step 1.3.1, performing linear fitting on any microwave S parameter spectrum initial curve, calculating the corresponding spectrum data distance between the microwave S parameter spectrum initial curve and the linear fitting straight line of the microwave S parameter spectrum initial curve to obtain a spectrum data linear fitting distance matrix D,
wherein, A [ m-1] [ i ] is the spectral data of the ith frequency point of the m-1 th microwave S parameter spectrum initial curve, A [ m-1] [ i ] is the fitted spectral data of the ith frequency point of a linear fitted straight line corresponding to the m-1 th microwave S parameter spectrum initial curve, n is the number of the frequency points corresponding to each microwave S parameter spectrum initial curve obtained by testing, and m is the number of the microwave S parameter spectrum initial curves;
step 1.3.2, comparing the linear fitting distance of the spectrum data in the linear fitting distance matrix D of the spectrum data with a preset distance threshold, and eliminating a microwave S parameter spectrum initial curve which is greater than the distance threshold to obtain a coal sample microwave S parameter spectrum curve corresponding to the current area;
and the microwave S parameter spectrum curves of the coal samples in all the areas form the microwave S parameter spectrum information of the coal training samples of the training data set.
The preset distance threshold value is 1.5 multiplied by dis _ m, wherein dis _ m is a distance median after the spectral data of the region where the current raw coal sample is located are subjected to linear fitting distance sorting.
Dividing the manufactured training data set according to the acquisition region of the coal training sample to obtain a plurality of groups of training data subsets;
and inputting the training data subset into a DW-KNN model when k neighbors exist according to a leave-one-out cross validation method, respectively calculating cross validation mean values of evaluation indexes when the k neighbors exist at different numbers, selecting the k neighbors when the cross validation mean value of the evaluation indexes is optimal, and constructing and obtaining a DW-KNN-based coal moisture content detection model according to the k neighbors when the selected evaluation index cross validation mean value is optimal.
When the DW-KNN-based coal moisture content detection model processes microwave S parameter spectrum information of a coal sample to obtain the moisture content of the coal sample, the DW-KNN-based coal moisture content detection method comprises the following steps:
step 2.1, calculating microwave S parameter spectrum information of the coal sample and a qth coal training sample delta q Euclidean distance of
Wherein the content of the first and second substances,
spectral data corresponding to the p-th frequency point of the microwave S parameter spectral information of the coal sample delta qp For the qth coal training sample δ q Spectrum data corresponding to the pth frequency point of the microwave S parameter spectrum information of the coal training sample;
step 2.2, all the Euclidean distances obtained by calculation
Sequencing according to an ascending order, and determining coal training samples respectively corresponding to the first k Euclidean distances;
step 2.3, according to the first k Euclidean distances determined by selection
And corresponding coal training samples, and calculating and outputting the moisture content of the coal specimen as follows:
wherein the content of the first and second substances,
moisture content of coal specimen, psi (delta) j ) To determine the jth coal training sample delta in the first k coal training samples j Coal training sample moisture content tag, σ j To determine the jth coal training sample delta in the first k coal training samples j The weight of the distance of (a) is,
is the corresponding jth Euclidean distance in the determined first k Euclidean distances.
In step 1.1, a sieve with 13mm sieve pores is adopted to sieve the raw coal sample, the coal sample with the moisture content to be detected is also required to be sieved by the sieve with 13mm sieve pores, and the coal sample microwave S parameter spectrum information of the coal sample which is consistent with the coal training sample microwave S parameter spectrum information is obtained for the sieved coal sample.
The microwave testing system comprises a vector network analyzer and two horn antennas matched with the vector network analyzer, a sample container used for containing a coal sample or a coal specimen is arranged between the two horn antennas which are just correspondingly arranged, and the vector network analyzer is electrically connected with the testing main controller.
The invention has the advantages that: the method comprises the steps of constructing a DW-KNN-based coal moisture content detection model, obtaining coal specimen microwave S parameter spectrum information of which the coal specimen is consistent with the coal training sample microwave S parameter spectrum information, and processing the obtained coal specimen microwave S parameter spectrum information by using the DW-KNN-based coal moisture content detection model to obtain and output the moisture content of the coal specimen, namely, realizing the detection of the coal moisture content under the lossless condition, and improving the detection precision and the robustness.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of a microwave testing system according to the present invention.
Description of reference numerals: 1-test main controller, 2-vector network analyzer, 3-horn antenna and 4-sample container.
Detailed Description
The invention is further illustrated by the following specific figures and examples.
As shown in fig. 1: in order to realize the detection of the coal moisture content under the lossless condition and improve the detection precision and the robustness, the rapid nondestructive detection method of the coal moisture content comprises the following steps:
step 1, constructing a DW-KNN-based coal moisture content detection model, wherein when the DW-KNN-based coal moisture content detection model is constructed, a manufactured training data set comprises a coal training sample moisture content label of a coal training sample and coal training sample microwave S parameter spectrum information of the coal training sample under the coal training sample moisture content label, and the manufactured training data set is used for training the DW-KNN model to construct and obtain the DW-KNN-based coal moisture content detection model;
in particular, the DW-KNN (Distance Weighted K-nearest neighbor) model is a K-nearest neighbor algorithm based on Distance weighting, and is a commonly used model in the prior art, and the details of the DW-KNN are consistent with the prior art and well known to those skilled in the art. And (5) constructing a coal moisture content detection model by using the DW-KNN model to obtain the DW-KNN-based coal moisture content detection model. When the DW-KNN model is used for building the DW-KNN-based coal moisture content detection model, a training data set needs to be made, the DW-KNN model is trained by using the made training data set, and the DW-KNN-based coal moisture content detection model is obtained after training.
And for a training data set, including a coal training sample moisture content label of a coal training sample and microwave S parameter spectrum information of the coal training sample under the coal training sample moisture content label, wherein the coal training sample moisture content label of the coal training sample is the current moisture content value of the coal training sample. The microwave S parameter spectrum information of the coal training samples under the moisture content labels of the coal training samples, specifically, the microwave S parameter spectrum information of the coal training samples corresponds to the moisture content labels of the coal training samples one by one.
The manufacturing process of the training data set and the specific process of constructing the coal moisture content detection model based on the DW-KNN are specifically described by the following expression.
In specific implementation, the method specifically comprises the following steps of:
step 1.1, collecting raw coal samples in different areas, screening all the collected raw coal samples, uniformly and flatly paving the screened raw coal samples in corresponding carriers, and drying the raw coal samples in the carriers to obtain experimental coal samples in different areas;
in the embodiment of the invention, the area of the raw coal sample can be selected according to actual needs, and is not described herein again. Typically, a sample of raw coal may be obtained by manual collection. Respectively screening collected raw coal samples from different areas by using a sieve with 13mm sieve pores, and uniformly spreading the raw coal samples in a high-temperature resistant aluminum tray after screening; placing the aluminum plate into a blast drying oven to be dried for 2 hours at 108 ℃ to obtain an experimental coal sample; that is, the carrier may be made of a high temperature resistant aluminum plate, and of course, the carrier may also be made in other forms, which may be specifically selected according to the needs, and the description thereof is omitted here.
In specific implementation, 500 +/-10 g of screened raw coal sample is uniformly and flatly paved in a high-temperature-resistant aluminum disc, and the thickness of the raw coal sample is not more than 2 times of the maximum particle diameter of the raw coal sample.
Step 1.2, carrying out sealing cooling, water spraying stirring and sealing standing treatment on the obtained experimental coal samples in different areas in sequence to obtain coal training samples under a preset coal training sample moisture content label;
in the embodiment of the invention, the experimental coal sample is taken out from the blast drying box and then immediately put into a barrel with a sealing cover for sealing and cooling treatment, and meanwhile, the mass of the dried coal sample is weighed for subsequent calculation; after the coal sample is cooled to normal temperature, opening a sealing cover, spraying a certain amount of water to the coal sample, and slowly stirring the coal sample to fully mix the coal and the water; and sealing the coal sample again, and placing the coal sample in a room temperature environment for more than 6 hours to fully diffuse water in the coal, thereby finally obtaining the coal training sample.
In specific implementation, all experimental processes are carried out in an indoor normal-temperature environment; the mass of the dried coal sample is obtained by weighing the total weight of the dried coal and the barrel minus the net weight of the barrel; when the steps of spraying water to the experimental coal sample and stirring are carried out, the contact time of the coal sample and the air is reduced as much as possible, and sealing treatment is carried out in time, so that the interference of moisture in the air on the experimental result is prevented; the process of the steps of spraying water and stirring the experimental coal sample can be executed by adopting the technical means commonly used in the technical field, and the steps can be specifically selected according to actual needs.
In the embodiment of the invention, when water is sprayed and stirred, the water spraying quality of any experimental coal sample determined according to the preset coal training sample moisture content label is as follows:
wherein M is add Is the mass of the water spray, M c&m Is the mass of the mixture of the experimental coal sample and water, MC before Is the moisture content, MC, of the experimental coal sample after Is a coal sample moisture content label of a coal training sample.
During concrete implementation, a coal training sample moisture content label can be preset according to actual requirements, the preset coal training sample moisture content label can be reasonably set according to the condition of a raw coal sample, and after the coal training sample moisture content label is set, the water spraying quality can be determined, so that the coal training sample moisture content label of the required coal training sample and the manufactured coal training sample can be finally prepared. Since it is difficult to control the quality of sprayed water in actual experiments, the moisture content of a new coal sample can be recalculated by the mass difference between before and after spraying water to correct errors.
When water is sprayed, the moisture content label of the preset coal training sample can be finally reached through multiple water spraying operations. When the water spraying operation is carried out for multiple times, the moisture content MC of the experimental coal sample before The water content MC of the experimental coal sample is the water content after the last water spraying, such as the first water spraying and stirring after the sealing and the cooling before Is 0 or a value close to 0. After the first water spraying and before the second water spraying, the corresponding moisture content MC of the coal testing sample can be obtained before And the rest can be analogized, and the description is omitted here.
And step 1.3, performing microwave test on the obtained coal training samples to obtain coal sample microwave S parameter spectrum information of the coal training samples in corresponding areas under the coal training sample moisture content labels.
Specifically, the microwave testing system for microwave testing is further included, the microwave testing system is used for implementing microwave testing on the coal training sample, a specific implementation case of the microwave testing system is shown in fig. 2, specifically: the microwave testing system comprises a vector network analyzer 2 and two horn antennas 3 matched with the vector network analyzer 2, a sample container 4 used for containing a coal sample or a coal specimen is arranged between the two horn antennas 3 which are correspondingly arranged, and the vector network analyzer 2 is electrically connected with a testing main controller 1.
Specifically, the main test controller 1 may adopt a conventional computer, which is mainly used to control the vector network analyzer 2 to receive and transmit microwave signals and visualize the spectrum information of the S parameters of the microwaves, the main test controller 1 is connected to the vector network analyzer 2 through a data transmission line, the vector network analyzer 2 may adopt a conventional portable analyzer, and the specific situation of the vector network analyzer 2 is consistent with the conventional situation, which is well known to those skilled in the art and is not described herein again. Two signal ports of the vector network analyzer 2 are respectively connected with the rectangular horn antenna 3 through coaxial cables, the two horn antennas 3 are oppositely arranged in a mouth-to-mouth mode and are installed on a horizontal linear sliding rail, the distance between the antenna ports of the horn antennas 3 is 10cm, the horn antennas 3 can adopt the existing common mode, and the specific connection matching mode of the horn antennas 3 and the vector network analyzer 2 is consistent with the existing mode.
The coal sample can be stored through the sample container 4, the sample container 4 is placed in the central area of the central axis of the horn antenna 3, in the embodiment of the invention, microwave signals with 8.05-12.01GHz frequency transmitted and received by the vector network analyzer 2 totally contain 133 frequency points, and 133 spectrum data can be obtained through the 133 frequency points of the vector network analyzer 2. The horn antenna 3 is an EIA standard WR90 waveguide type horn antenna with a gain of 20dB, and the sample container 4 is a rectangular container of PMMA material with a thickness of 3 mm.
In the concrete implementation, when the coal training samples are subjected to microwave testing, spectrum data of corresponding frequency points under the frequency of 8.05 GHz-12.01 GHz is acquired and obtained for any coal training sample so as to obtain coal sample microwave S parameter spectrum information of the coal training samples, wherein all coal training sample microwave S parameter spectrum data of one coal training sample form a microwave S parameter spectrum initial curve.
As can be seen from the above description, for any coal training sample, under the effect of 133 frequency points, 133 spectrum data, that is, the microwave S parameter spectrum data of 133 coal training samples, can be obtained, so that one initial microwave S parameter spectrum curve can be formed by using the microwave S parameter spectrum data of 133 coal training samples of one coal training sample. In a specific implementation, the spectrum data at a frequency point specifically refers to the insertion loss parameter S21, where the specific situation of the insertion loss parameter S21 is consistent with the prior art, and is not described herein again.
During specific implementation, a microwave S parameter spectrum initial curve can be generated after microwave testing is carried out on coal training samples collected in different areas or under different coal training sample moisture content labels, and the total number of the microwave S parameter spectrum initial curves can be selected and determined according to actual needs.
Further, preprocessing the initial microwave S parameter spectrum curves of all coal training samples from the same region to obtain coal sample microwave S parameter spectrum information of the coal training samples under the coal training sample moisture content label after preprocessing, wherein the preprocessing comprises the following steps:
step 1.3.1, performing linear fitting on any microwave S parameter spectrum initial curve, calculating the corresponding spectrum data distance between the microwave S parameter spectrum initial curve and the linear fitting straight line of the microwave S parameter spectrum initial curve to obtain a spectrum data linear fitting distance matrix D,
wherein, A [ m-1] [ i ] is the spectral data of the ith frequency point of the m-1 th microwave S parameter spectrum initial curve, A [ m-1] [ i ] is the fitted spectral data of the ith frequency point of a linear fitted straight line corresponding to the m-1 th microwave S parameter spectrum initial curve, n is the number of the frequency points corresponding to each microwave S parameter spectrum initial curve obtained by testing, and m is the number of the microwave S parameter spectrum initial curves;
step 1.3.2, comparing the linear fitting distance of the spectrum data in the linear fitting distance matrix D of the spectrum data with a preset distance threshold, and eliminating a microwave S parameter spectrum initial curve which is greater than the distance threshold to obtain a coal sample microwave S parameter spectrum curve corresponding to the current region;
and the coal sample microwave S parameter spectrum curves of all the areas form coal training sample microwave S parameter spectrum information of the training data set.
Specifically, any initial curve of the S-parameter spectrum of the microwave may be linearly fitted by using a common technical means in the field, and a specific linear fitting manner and the like may be selected as needed, which is not described herein again. After linear fitting, a linear fitting straight line which is just corresponding to a microwave S parameter spectrum initial curve can be obtained. As can be seen from the above description, each frequency point corresponds to a spectrum data, and thus after a linear fit straight line is obtained, spectrum data on the initial curve of the microwave S-parameter spectrum and corresponding spectrum data on the linear fit straight line can be determined for any frequency point, and a linear fit distance matrix D of the spectrum data can be obtained through the spectrum data on all the initial curves of the microwave S-parameter spectrum and corresponding spectrum data on the linear fit straight line.
In the embodiment of the present invention, as can be seen from the above description, the number n of frequency points on each initial curve of the microwave S parameter spectrum is 133, and the number m of the initial curves of the microwave S parameter spectrum can be specifically selected according to actual needs, and the above description can be specifically referred to.
According to the spectrum data linear fitting distance matrix D, the spectrum data linear fitting distance corresponding to any coal training sample can be compared with a preset distance threshold, and a microwave S parameter spectrum initial curve larger than the distance threshold is eliminated, so that the needed coal sample microwave S parameter spectrum information is obtained. In specific implementation, the preset distance threshold is 1.5 × dis _ m, where dis _ m is a distance median after linear fitting distance sorting of spectral data of a region where the current raw coal sample is located.
In specific implementation, when a raw coal sample corresponds to a plurality of acquisition regions, the initial microwave S parameter spectrum curves of all coal training samples in each region need to be preprocessed, so that after preprocessing is completed, the microwave S parameter spectrum information of the coal sample is formed by the microwave S parameter spectrum curves of the coal samples in all regions, wherein the microwave S parameter spectrum information of the coal sample is specifically a plurality of microwave S parameter spectrum curves of the coal sample.
Further, dividing the manufactured training data set according to the acquisition region of the coal training sample to obtain a plurality of groups of training data subsets;
and inputting the training data subset into a DW-KNN model when k neighbors exist according to a leave-one-out cross validation method, respectively calculating cross validation mean values of evaluation indexes when the k neighbors exist at different numbers, selecting the k neighbors when the cross validation mean value of the evaluation indexes is optimal, and constructing and obtaining a DW-KNN-based coal moisture content detection model according to the k neighbors when the selected evaluation index cross validation mean value is optimal.
In the embodiment of the invention, as the raw coal samples are collected in different areas, after the training data set is manufactured through the steps, the training data set is divided according to the collection areas of the coal training samples, so as to obtain a plurality of groups of training data subsets. The leave-one cross verification method specifically comprises the following steps: the specific implementation and process of leaving a cross validation method are consistent with the prior art, which are well known in the art and will not be described herein again.
Respectively calculating the cross validation mean values of evaluation indexes including a decision coefficient R when k neighbor numbers are input into a training data subset according to a DW-KNN model when k neighbor numbers are input by a leave-one-cross validation method commonly used in the technical field 2 The Mean Absolute Error (MAE) and the Root Mean Square Error (RMSE) are calculated and recorded respectively, specifically, the k neighbor number is gradually increased from small to small, and the cross validation Mean values of the evaluation indexes at different k neighbor numbers are calculated and recorded respectively; and when the evaluation indexes do not change greatly any more, stopping operation, selecting k nearest neighbor when the evaluation index cross validation mean value is optimal as the optimal selection of the model, and obtaining the DW-KNN-based coal moisture content detection model. In specific implementation, when the evaluation index cross validation mean value is optimal, generally, the coefficient R is determined 2 Is maximized, i.e. the coefficient R can be determined 2 Determining whether the evaluation index cross validation mean value is optimal or not according to the value taking condition; of course, a technical means may also be adopted to determine that the cross validation mean of the evaluation index is optimal, and specifically, a manner and a process for determining that the cross validation mean of the evaluation index is optimal are well known to those skilled in the art.
In the embodiment of the invention, the k neighbor number is increased by 1 from 3 each time until the evaluation index has no longer changed greatly, and the specific process is consistent with the prior art and is well known by the technical field.
And 2, providing a coal sample to be detected for water content, acquiring coal sample microwave S parameter spectrum information of which the coal sample is consistent with the coal training sample microwave S parameter spectrum information, and processing the acquired coal sample microwave S parameter spectrum information by using the DW-KNN-based coal water content detection model to obtain and output the water content of the coal sample.
In specific implementation, the microwave S parameter spectrum information of the coal specimen, which is consistent with the microwave S parameter spectrum information of the coal training sample, is obtained, specifically, for the coal specimen providing the moisture content to be detected, a sieve with 13mm sieve meshes is adopted to sieve the coal specimen, and the microwave S parameter spectrum information of the coal specimen, which is consistent with the microwave S parameter spectrum information of the coal training sample, is obtained for the sieved coal specimen; the coal specimen microwave S parameter spectrum information is consistent with the coal training sample microwave S parameter spectrum information, namely the coal specimen microwave S parameter spectrum information and the coal training sample microwave S parameter spectrum information are tested under the same frequency point to obtain corresponding spectrum data, and when the number n of the corresponding frequency points of each microwave S parameter spectrum initial curve obtained through the test is 133, the coal specimen microwave S parameter spectrum information and the coal training sample microwave S parameter spectrum information both comprise spectrum data corresponding to 133 frequency points.
When the DW-KNN-based coal moisture content detection model processes microwave S parameter spectrum information of a coal specimen to obtain the moisture content of the coal specimen, the DW-KNN-based coal moisture content detection method comprises the following steps:
step 2.1, calculating microwave S parameter spectrum information of the coal sample and a qth coal training sample delta q Euclidean distance of
Wherein the content of the first and second substances,
spectral data corresponding to the p-th frequency point of the microwave S parameter spectral information of the coal sample delta qp For the qth coal training sample δ q Spectrum data corresponding to the p-th frequency point of the microwave S parameter spectrum information of the coal training sample;
specifically, when the total number of the coal sample microwave S parameter spectrum curves included in the coal sample microwave S parameter spectrum information is Q, the value range of Q is 1 to Q.
Step 2.2, all Euclidean distances obtained by calculation
Sequencing in an ascending order, and determining coal training samples respectively corresponding to the first k Euclidean distances;
specifically, by step 1, Q Euclidean distances can be determined
To the Euclidean distance
And after sequencing according to the ascending order, selecting the first k Euclidean distances and the coal training samples corresponding to the first k Euclidean distances according to the determined k neighbor numbers, and after determining the first k coal training samples, determining the moisture content label of the coal training sample of each coal training sample and the microwave S parameter spectrum information of the coal training samples under the moisture content label of the coal training samples.
Step 2.3, according to the first k Euclidean distances determined by selection
And corresponding coal training samples, and calculating and outputting the moisture content of the coal specimen as follows:
wherein the content of the first and second substances,
moisture content of coal specimen, psi (delta) j ) To determine the jth coal training sample delta in the first k coal training samples j Coal training sample moisture content tag, σ j To determine the jth coal training sample delta in the first k coal training samples j The weight of the distance of (a) is,
is the corresponding jth Euclidean distance in the determined first k Euclidean distances.
In the embodiment of the invention, the moisture content of the coal sample can be obtained by processing the microwave S parameter spectrum information of the coal sample.
Claims (5)
1. A DW-KNN-based coal moisture content rapid nondestructive testing method is characterized by comprising the following steps:
step 1, constructing a DW-KNN-based coal moisture content detection model, wherein when the DW-KNN-based coal moisture content detection model is constructed, a manufactured training data set comprises a coal training sample moisture content label of a coal training sample and coal training sample microwave S parameter spectrum information of the coal training sample under the coal training sample moisture content label, and the manufactured training data set is used for training the DW-KNN model to construct and obtain the DW-KNN-based coal moisture content detection model;
step 2, providing a coal sample to be detected for moisture content, acquiring coal sample microwave S parameter spectrum information of which the coal sample is consistent with coal training sample microwave S parameter spectrum information, and processing the acquired coal sample microwave S parameter spectrum information by using the DW-KNN-based coal moisture content detection model to obtain and output the moisture content of the coal sample;
in step 1, when a training data set is manufactured, the method specifically comprises the following steps:
step 1.1, collecting raw coal samples in different areas, screening all the collected raw coal samples, uniformly and flatly paving the screened raw coal samples in corresponding carriers, and drying the raw coal samples in the carriers to obtain experimental coal samples in different areas;
step 1.2, carrying out sealing cooling, water spraying stirring and sealing standing treatment on the obtained experimental coal samples in different areas in sequence to obtain coal training samples under a preset coal training sample moisture content label;
step 1.3, performing microwave testing on the obtained coal training samples to obtain coal sample microwave S parameter spectrum information of the coal training samples in corresponding areas under the coal training sample moisture content label;
in the step 1.3, when the coal training samples are subjected to microwave testing, spectrum data of corresponding frequency points under the frequency of 8.05 GHz-12.01 GHz are acquired and obtained for any coal training sample to obtain coal sample microwave S parameter spectrum information of the coal training sample, wherein all coal sample microwave S parameter spectrum data of one coal training sample form a microwave S parameter spectrum initial curve;
preprocessing microwave S parameter spectrum initial curves of all coal training samples from the same area to obtain coal sample microwave S parameter spectrum information of the coal training samples under a coal training sample moisture content label after preprocessing, wherein the preprocessing comprises the following steps:
step 1.3.1, performing linear fitting on any microwave S parameter spectrum initial curve, calculating the corresponding spectrum data distance between the microwave S parameter spectrum initial curve and the linear fitting straight line of the microwave S parameter spectrum initial curve to obtain a spectrum data linear fitting distance matrix D,
wherein, A [ m-1] [ i ] is the spectral data of the ith frequency point of the m-1 th microwave S parameter spectrum initial curve, A [ m-1] [ i ] is the fitted spectral data of the ith frequency point of a linear fitted straight line corresponding to the m-1 th microwave S parameter spectrum initial curve, n is the number of the frequency points corresponding to each microwave S parameter spectrum initial curve obtained by testing, and m is the number of the microwave S parameter spectrum initial curves;
step 1.3.2, comparing the linear fitting distance of the spectrum data in the linear fitting distance matrix D of the spectrum data with a preset distance threshold, and eliminating a microwave S parameter spectrum initial curve which is greater than the distance threshold to obtain a coal sample microwave S parameter spectrum curve corresponding to the current region;
the microwave S parameter spectrum curves of the coal samples in all the areas form the microwave S parameter spectrum information of the coal training samples of the training data set;
dividing the manufactured training data set according to the acquisition region of the coal training sample to obtain a plurality of groups of training data subsets;
inputting a training data subset into a DW-KNN model when k neighbors exist according to a leave-one-out cross validation method, respectively calculating cross validation mean values of evaluation indexes when the k neighbors exist at different numbers, selecting the k neighbors when the cross validation mean value of the evaluation indexes is optimal, and constructing and obtaining a DW-KNN-based coal moisture content detection model according to the k neighbors when the selected evaluation index cross validation mean value is optimal;
when the DW-KNN-based coal moisture content detection model processes microwave S parameter spectrum information of a coal specimen to obtain the moisture content of the coal specimen, the DW-KNN-based coal moisture content detection method comprises the following steps:
step 2.1, calculating microwave S parameter spectrum information of the coal sample and a qth coal training sample delta q Euclidean distance of
Wherein the content of the first and second substances,
spectral data corresponding to the p-th frequency point of the microwave S parameter spectral information of the coal sample delta qp For the qth coal training sample δ q Spectrum data corresponding to the p-th frequency point of the microwave S parameter spectrum information of the coal training sample;
step 2.2, all the Euclidean distances obtained by calculation
Sequencing in an ascending order, and determining coal training samples respectively corresponding to the first k Euclidean distances;
step 2.3, according to the first k Euclidean distances determined by selection
And corresponding coal training samples, and calculating and outputting the moisture content of the coal specimen as follows:
wherein, the first and the second end of the pipe are connected with each other,
moisture content of coal specimen, psi (delta) j ) To determine the jth coal training sample delta in the first k coal training samples j Coal training sample moisture content tag, σ j To determine the jth coal training sample delta in the first k coal training samples j The weight of the distance of (a) is,
is the corresponding j-th Euclidean distance in the determined first k Euclidean distances.
2. The DW-KNN-based coal moisture content rapid nondestructive testing method as claimed in claim 1, wherein in step 1.2, during water spraying and stirring, the water spraying quality determined according to the moisture content label of the preset coal training sample for any experimental coal sample is as follows:
wherein M is add Is the mass of the water spray, M c&m Is the mass of the mixture of the experimental coal sample and water, MC before Is the moisture content, MC, of the experimental coal sample after Is a coal sample moisture content label of a coal training sample.
3. The DW-KNN-based coal moisture content rapid nondestructive testing method according to claim 1, wherein the preset distance threshold is 1.5 x dis _ m, wherein dis _ m is a distance median after linear fitting distance sorting of spectral data of a region where the current raw coal sample is located.
4. The DW-KNN-based coal moisture content rapid nondestructive testing method as claimed in claim 1, wherein in step 1.1, a 13mm sieve is used for sieving the raw coal sample, and for the coal sample providing the moisture content to be detected, a 13mm sieve is also used for sieving the coal sample, and for the sieved coal sample, the coal sample microwave S parameter spectrum information of the coal sample is obtained, wherein the coal sample is consistent with the coal training sample microwave S parameter spectrum information.
5. The DW-KNN-based coal moisture content rapid nondestructive testing method according to claim 1 or 3, characterized by further comprising a microwave testing system for microwave testing, wherein the microwave testing system comprises a vector network analyzer (2) and two horn antennas (3) matched with the vector network analyzer (2), a sample container (4) for containing a coal sample or a coal sample is arranged between the two horn antennas (3) which are arranged in a positive correspondence manner, and the vector network analyzer (2) is electrically connected with the testing main controller (1).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016888.2A CN114354654B (en) | 2022-01-07 | 2022-01-07 | DW-KNN-based rapid nondestructive testing method for coal moisture content |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210016888.2A CN114354654B (en) | 2022-01-07 | 2022-01-07 | DW-KNN-based rapid nondestructive testing method for coal moisture content |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114354654A CN114354654A (en) | 2022-04-15 |
| CN114354654B true CN114354654B (en) | 2023-03-21 |
Family
ID=81107026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210016888.2A Active CN114354654B (en) | 2022-01-07 | 2022-01-07 | DW-KNN-based rapid nondestructive testing method for coal moisture content |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114354654B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115575417B (en) * | 2022-08-03 | 2023-10-31 | 华能应城热电有限责任公司 | Coal moisture detection method and device |
| CN116070929B (en) * | 2022-10-12 | 2023-10-03 | 华院计算技术(上海)股份有限公司 | Self-adaptive local dynamic coke quality prediction method and system |
| CN116844658B (en) * | 2023-07-13 | 2024-01-23 | 中国矿业大学 | Method and system for rapidly measuring water content of coal based on convolutional neural network |
| CN117723514B (en) * | 2024-02-07 | 2024-05-17 | 山西品东智能控制有限公司 | Multi-index coal quality analyzer based on sodium light source and intelligent algorithm and detection method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112881306A (en) * | 2021-01-15 | 2021-06-01 | 吉林大学 | Hyperspectral image-based method for rapidly detecting ash content of coal |
| WO2021243848A1 (en) * | 2020-06-03 | 2021-12-09 | 枣庄学院 | Anomaly detection method for wireless sensor network |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH01282464A (en) * | 1988-05-10 | 1989-11-14 | Chubu Electric Power Co Inc | Moisture measuring device |
| US6137294A (en) * | 1999-01-21 | 2000-10-24 | Usx Corporation | Prediction of bulk density of particulates with a correlation based on moisture content |
| AU1285102A (en) * | 2000-10-11 | 2002-04-22 | Nz Forest Research Inst Ltd | Microwave moisture sensing via waveguide with slot array |
| CN104515749B (en) * | 2014-10-24 | 2017-11-03 | 中华人民共和国黄埔出入境检验检疫局 | The quick determination method of water in coal sample |
| CN204374103U (en) * | 2015-01-20 | 2015-06-03 | 西安阿尔特测控技术有限公司 | The online Coal ' moisture measuring system of microwave method |
| CN105277531B (en) * | 2015-09-25 | 2018-04-10 | 清华大学 | A kind of coal characteristic measuring method based on stepping |
| CN105842113A (en) * | 2016-06-15 | 2016-08-10 | 安徽有果信息技术服务有限公司 | On-line detection method of content of water in coal |
| CN107764796B (en) * | 2017-09-19 | 2019-07-26 | 华中科技大学 | A method of raw coal moisture, volatile matter are detected with baseline drift amount |
| CN110308166B (en) * | 2019-07-05 | 2020-06-05 | 中国矿业大学 | Experimental device and method for measuring moisture content and distribution in coal rock by utilizing X-ray |
| CN110879234B (en) * | 2019-12-27 | 2020-12-25 | 中国矿业大学 | Method for rapidly detecting total moisture of coal based on low-field nuclear magnetic resonance |
| CN111753877B (en) * | 2020-05-19 | 2024-03-05 | 海克斯康制造智能技术(青岛)有限公司 | Product quality detection method based on deep neural network migration learning |
| CN111754028A (en) * | 2020-06-08 | 2020-10-09 | 吉林大学 | Hyperspectrum-based coal ash content and moisture detection system and method |
-
2022
- 2022-01-07 CN CN202210016888.2A patent/CN114354654B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021243848A1 (en) * | 2020-06-03 | 2021-12-09 | 枣庄学院 | Anomaly detection method for wireless sensor network |
| CN112881306A (en) * | 2021-01-15 | 2021-06-01 | 吉林大学 | Hyperspectral image-based method for rapidly detecting ash content of coal |
Non-Patent Citations (1)
| Title |
|---|
| 张军.基于Isomap-BP的寒地老年住宅热舒适度模型.2019,(第03期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114354654A (en) | 2022-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114354654B (en) | DW-KNN-based rapid nondestructive testing method for coal moisture content | |
| CN108132226B (en) | Terahertz spectrum quantitative analysis method for rubber reinforcing agent carbon black | |
| CN111157698B (en) | Inversion method for obtaining total potassium content of black soil by using emissivity data | |
| CN108872129A (en) | A kind of insulating paper near infrared spectrum analytic method based on Partial Least Squares | |
| CN110879212B (en) | Method for monitoring fluidized bed drying process state based on near infrared spectrum | |
| CN113297797A (en) | XGboost-based electronic transformer measurement error state evaluation method and device | |
| CN101726451A (en) | Method for measuring viscosity index of internal combustion engine oil | |
| CN109324014B (en) | Self-adaptive near-infrared rapid prediction method for crude oil properties | |
| CN104198512A (en) | Support vector machine-based X-ray fluorescence spectrum analysis method and support vector machine-based X-ray fluorescence spectrum analysis device | |
| CN111428561A (en) | Broadband electromagnetic interference source identification method and system, storage medium and equipment | |
| CN100443883C (en) | Method for detecting hydrgenated tail-oil paraffin composition using near-infrared spectrum | |
| CN115510754A (en) | Method and system for predicting and controlling performance of cement concrete | |
| CN112651173B (en) | Agricultural product quality nondestructive testing method based on cross-domain spectral information and generalizable system | |
| CN115876661A (en) | Asphalt pavement void ratio evaluation method based on three-dimensional ground penetrating radar image | |
| CN115452650A (en) | System and method for detecting density of asphalt pavement | |
| CN115456357A (en) | Decision-making method for 5G base station antenna deployment of transformer substation | |
| CN114755302A (en) | Ultrasonic phased array-based crack characterization method and device | |
| CN113591363A (en) | Transformer oil dielectric loss regression prediction method based on multi-frequency ultrasonic detection | |
| Wang et al. | A comparison of second‐order derivative based models for time domain reflectometry waveform analysis | |
| Kalandarov et al. | Analysis of the humidity measurement infrared method of the grain materials | |
| CN1979132A (en) | Method for detecting hydrogenated tail-oil cyclanes and arene composition using near infrared spectrum | |
| CN116046714B (en) | Method for establishing detection model of resin and/or volatile component content in phthalazinone polyarylether thermoplastic prepreg tape and application | |
| CN112630241B (en) | Method for analyzing mineral substances in coal | |
| Liu et al. | A Normalized Site Attenuation Automatic Measurement Method for OATS | |
| CN113324937B (en) | Biomass fuel quality inspection method and system based on terahertz spectrum |
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 |