Coal gangue dielectric identification method based on frequency shift characteristic

Abstract

The invention relates to the technical field of dielectric measurement, in particular to a coal gangue dielectric identification method based on frequency shift characteristics. The method comprises the following steps: obtaining a coal gangue sample, and carrying out dielectric constant analysis and dielectric capacitance change amount accounting to obtain capacitance change amounts after coal gangue medium changes corresponding to different polar plate structures; the method comprises the following steps: connecting a multi-frequency signal generator and a resistor between two polar plates corresponding to a variable-medium capacitive sensor, carrying out excitation circuit abstraction processing, and carrying out coal gangue voltage effective value calculation to obtain voltage effective values at two ends of the resistor; collecting a coal gangue medium response signal and a coal gangue medium excitation frequency through a data recorder, and carrying out signal noise reduction and frequency loss correction to obtain a medium loss correction excitation frequency; and constructing a coal gangue medium frequency shift characteristic feature vector through the voltage effective value at the two ends of the resistor and the dielectric loss correction excitation frequency, and outputting a dielectric classification result corresponding to the coal and gangue samples. According to the method, the coal gangue identification accuracy can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of dielectric measurement, and particularly to a coal gangue dielectric identification method based on frequency shift characteristics. Background Art

[0002] Coal accounts for about 25% of the world's total energy supply and is the second largest energy source after oil. Coal gangue separation underground and direct filling of goafs with gangue is one of the most effective ways to cleanly utilize coal gangue, which is of great significance for increasing the production capacity of coal hoisted from the mine, saving transportation costs and reducing waste emissions. For example, coal gangue contains certain mineral components and elements. Through effective identification and sorting technologies, it can be converted into useful industrial raw materials or energy. In addition, the frequency shift characteristic refers to the variation law of the dielectric constant and dielectric loss of a material under the action of an external electromagnetic field as the frequency changes. By analyzing the responses of electromagnetic waves in different frequency bands, the unique dielectric characteristics of coal gangue can be extracted, thus achieving more accurate identification and classification. However, existing coal gangue identification methods mostly rely on traditional experimental means, which are divided into two categories: manual gangue selection and mechanical device gangue selection. The data acquisition and processing processes are relatively complex, and are limited by the experimental environment and the accuracy of equipment, resulting in a reduction in the dielectric identification effect. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a coal gangue dielectric identification method based on frequency shift characteristics to solve at least one of the above technical problems.

[0004] To achieve the above object, a coal gangue dielectric identification method based on frequency shift characteristics includes the following steps:

[0005] Step S1: Obtain coal gangue samples, including coal samples and gangue samples, and analyze the dielectric constant of the medium of the coal gangue samples to obtain the relative dielectric constant of the coal gangue medium; based on the relative dielectric constant of the coal gangue medium, calculate the change amount of the medium capacitance by adding the coal gangue samples between the two electrodes of a variable medium type capacitive sensor to obtain the change amount of the capacitance of the coal gangue medium after change corresponding to different electrode structures, where the different electrode structures include series type, parallel type, and series-parallel type;

[0006] Step S2: Connect a multi-frequency signal generator and a resistor between the corresponding two electrodes of the variable medium type capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for the coal gangue medium; calculate the effective value of the voltage of the coal gangue based on the change amount of the capacitance of the coal gangue medium after change for the dual-frequency excitation RC circuit of the coal gangue medium to obtain the effective value of the voltage across the resistor corresponding to the coal gangue sample;

[0007] Step S3: Collect the coal-gangue medium response signal corresponding to the variable-medium capacitive sensor in the coal-gangue medium dual-frequency excitation RC circuit and the coal-gangue medium excitation frequency corresponding to the coal-gangue sample through a data recorder, and perform VMD-Median signal denoising on the coal-gangue medium response signal to generate a coal-gangue medium denoised signal; obtain the dielectric loss value of the coal-gangue medium from the coal-gangue medium denoised signal, and perform frequency loss correction on the coal-gangue medium excitation frequency corresponding to the coal-gangue sample based on the dielectric loss value of the coal-gangue medium to obtain the dielectric loss correction excitation frequency corresponding to the coal-gangue sample;

[0008] Step S4: Construct a coal-gangue medium frequency shift characteristic feature vector through the effective value of the voltage across the resistor corresponding to the coal-gangue sample and the dielectric loss correction excitation frequency, and construct a coal-gangue classification and recognition model based on the coal-gangue medium frequency shift characteristic feature vector to output the dielectric classification results corresponding to the coal and gangue samples.

[0009] Further, step S1 includes the following steps:

[0010] Step S11: Obtain coal-gangue samples, including coal samples and gangue samples;

[0011] Step S12: Analyze the dielectric constant of the medium of the coal-gangue samples to obtain the relative dielectric constant ε of the coal-gangue medium jec ;

[0012] Step S13: Obtain the dielectric constant ε of the medium inside the electrode plate obj ; If it is determined that the corresponding medium structure between the two electrode plates in the variable-medium capacitive sensor is a series structure, then this variable-medium capacitive sensor is considered to be an up-and-down series structure between the medium corresponding to the coal-gangue sample and the medium between the electrode plates, so as to obtain the electrode plate area A, the distance d1 between the added coal-gangue medium, and the distance d0 between the media inside the electrode plate, and based on the relative dielectric constant ε of the coal-gangue medium jec Perform a series capacitance change calculation on the electrode plate area A, the distance d1 between the added coal-gangue medium, the distance d0 between the media inside the electrode plate, and the dielectric constant ε of the medium inside the electrode plate obj To obtain the capacitance change amount after the change of the coal-gangue medium corresponding to the series-type electrode plate structure;

[0013] Step S14: If it is determined that the corresponding medium structure between the two electrode plates in the variable-medium capacitive sensor is a parallel structure, then this variable-medium capacitive sensor is considered to be a left-and-right parallel structure between the medium corresponding to the coal-gangue sample and the medium between the electrode plates, so as to obtain the area A1 occupied by the electrode plate of the added coal-gangue medium, the area A2 occupied by the electrode plate of the medium inside the electrode plate, and the distance d between the two electrode plates, and based on the relative dielectric constant ε of the coal-gangue medium jec Perform calculations on the area A1 occupied by the electrode plate of the added coal-gangue medium, the area A2 occupied by the electrode plate of the medium inside the electrode plate, the distance d between the two electrode plates, and the dielectric constant ε of the medium inside the electrode plate objPerform parallel capacitance change calculations to obtain the capacitance change amount after the change of coal gangue medium corresponding to the parallel plate structure;

[0014] Step S15: If it is determined that the medium structure between the two plates in the variable medium capacitive sensor is a series-parallel structure, then when designing the capacitor, if conditions permit, the plate thickness is set to less than 0.5 mm to reduce the influence of the edge effect on the capacitance, calculate the actual value of the corrected capacitance, and perform series-parallel capacitance change calculations based on the actual value of the corrected capacitance to obtain the capacitance change amount after the change of coal gangue medium corresponding to the series-parallel plate structure.

[0015] Further, in step S13, based on the relative permittivity ε of the coal gangue medium jec Perform series capacitance change calculations on the plate area A, the distance d1 between the added coal gangue media, the distance d0 between the media inside the plate, and the permittivity ε of the media inside the plate, including the following steps: obj Perform upper medium capacitance calculations according to the relative permittivity ε of the coal gangue medium, the plate area A, the distance d1 between the added coal gangue media, and the permittivity ε of the media inside the plate to obtain the upper medium capacitance of the added coal gangue.

[0016] According to the relative permittivity ε of the coal gangue medium jec 、the plate area A, the distance d1 between the added coal gangue media, and the permittivity ε of the media inside the plate obj Perform lower medium capacitance calculations to obtain the lower medium capacitance inside the plate.

[0017] Perform capacitance summation calculations on the upper medium capacitance C1 of the added coal gangue and the lower medium capacitance C2 inside the plate according to the capacitance operation principle corresponding to the series circuit to obtain the total capacitance value of the series structure. obj Obtain the initial capacitance of the series structure corresponding to when no coal gangue medium is added.

[0018] Perform series capacitance change calculations on the total capacitance value C of the series structure based on the initial capacitance C of the series structure to obtain the capacitance change amount after the change of coal gangue medium corresponding to the series plate structure.

[0019] Obtain the initial capacitance of the series structure corresponding to when no coal gangue medium is added. And perform series capacitance change calculations on the total capacitance value C of the series structure based on the initial capacitance C of the series structure to obtain the capacitance change amount after the change of coal gangue medium corresponding to the series plate structure. c0 Perform series capacitance change calculations on the total capacitance value C of the series structure based on the initial capacitance C of the series structure to obtain the capacitance change amount after the change of coal gangue medium corresponding to the series plate structure. c Perform series capacitance change calculations on the total capacitance value C of the series structure based on the initial capacitance C of the series structure to obtain the capacitance change amount after the change of coal gangue medium corresponding to the series plate structure.

[0020] Further, in step S14, based on the relative permittivity ε of the coal gangue medium jec Perform parallel capacitance change calculations on the area A1 occupied by the added coal gangue medium plate, the area A2 occupied by the plate of the medium inside the plate, the distance d between the two plates, and the permittivity ε of the medium inside the plate, including the following steps: obj Perform parallel capacitance change calculations on the area A1 occupied by the added coal gangue medium plate, the area A2 occupied by the plate of the medium inside the plate, the distance d between the two plates, and the permittivity ε of the medium inside the plate, including the following steps:

[0021] According to the relative dielectric constant ε of the coal gangue medium jec 、the area A1 occupied by the coal gangue medium plates added, the distance d between the two plates, and the dielectric constant ε of the medium inside the plates obj Perform the left medium capacitance calculation to obtain the left medium capacitance of the added coal gangue

[0022] According to the area A2 occupied by the medium plates inside the plates, the distance d between the two plates, and the dielectric constant ε of the medium inside the plates obj Perform the right medium capacitance calculation to obtain the right medium capacitance inside the plates

[0023] According to the capacitance operation principle corresponding to the parallel circuit, perform the capacitance summation calculation on the left medium capacitance C′1 of the added coal gangue and the right medium capacitance C′2 inside the plates to obtain the total capacitance value of the parallel - type structure

[0024] Obtain the initial capacitance of the parallel - type structure corresponding to when no coal gangue medium is added And based on the initial capacitance C of the parallel - type structure b0 Perform the parallel capacitance change calculation on the total capacitance value C of the parallel - type structure b To obtain the capacitance change amount after the change of the coal gangue medium corresponding to the parallel - type plate structure

[0025] Furthermore, step S15 includes the following steps:

[0026] Step S151: If it is determined that the medium structure between the two plates of the variable - medium capacitive sensor is a series - parallel structure, then obtain the medium surface area S between the two plates and the medium distance D between the two plates corresponding to this variable - medium capacitive sensor;

[0027] Step S152: According to the medium surface area S between the two plates and the medium distance D between the two plates, perform the series - parallel initial capacitance calculation to obtain the initial capacitance of the series - parallel structure where ε f is the dielectric constant of the medium between the two plates, and k is the medium capacitance influence constant;

[0028] Step S153: When designing this capacitor, under the condition of permission, set the plate thickness to less than 0.5 mm to reduce the influence of the edge effect on the capacitance, and calculate that the edge effect influence coefficient ξ = 0.163, and according to the edge effect influence coefficient ξ and the initial capacitance C of the series - parallel structure h0 To calculate the actual value of the corrected capacitance

[0029] Step S154: Based on the initial capacitance C of the series - parallel structure h0 Perform the calculation on the actual value C of the corrected capacitance hPerform the calculation of the change in the combined capacitance to obtain the change in capacitance after the change in the coal gangue medium corresponding to the combined plate structure

[0030] Further, step S2 includes the following steps:

[0031] Step S21: Connect a multi-frequency signal generator and a resistor between the two electrodes corresponding to the variable dielectric type capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for coal gangue medium;

[0032] Step S22: Optionally select the change in capacitance after the change in the coal gangue medium corresponding to different plate structures as the change in capacitance ΔC after the change in the medium of the dual-frequency excitation RC circuit for coal gangue medium, and calculate the capacitance frequency shift amount between the two electrodes corresponding to the variable dielectric type capacitive sensor in the dual-frequency excitation RC circuit for coal gangue medium based on the change in capacitance ΔC after the change in the medium to obtain the capacitance frequency shift amount of the excitation RC circuit where f is the frequency of the alternating voltage;

[0033] Step S23: Apply a standard sinusoidal alternating voltage to the dual-frequency excitation RC circuit for coal gangue medium where B is the amplitude of the applied alternating voltage and t is time, is the initial phase angle, where The effective value phasor of the input voltage in the series circuit corresponding to the dual-frequency excitation RC circuit for coal gangue medium is where j is the imaginary unit of the phasor, and the standard sinusoidal alternating voltage U s in At this time, U′ = B, the impedance of the resistor R in the dual-frequency excitation RC circuit for coal gangue medium is Z R = R, and the impedance of the capacitor C is Z C = -jX c ;

[0034] Step S24: Calculate the effective value of the voltage according to the effective value phasor U′ of the input voltage, the impedance Z R of the resistor R, and the impedance Z C of the capacitor C to obtain the effective value of the voltage across the resistor corresponding to the coal gangue sample

[0035] Further, step S3 includes the following steps:

[0036] Step S31: Collect the coal gangue medium response signal corresponding to the variable dielectric type capacitive sensor in the dual-frequency excitation RC circuit for coal gangue medium and the coal gangue medium excitation frequency corresponding to the coal gangue sample through a data recorder;

[0037] Step S32: Perform VMD decomposition on the coal-gangue medium response signal using the variational mode decomposition (VMD) algorithm to obtain the spectrum of the VMD-decomposed signal of the coal-gangue medium;

[0038] Step S33: Reconstruct and denoise the spectrum of the VMD-decomposed signal of the coal-gangue medium using the Median signal to generate the denoised signal of the coal-gangue medium;

[0039] Step S34: Obtain the dielectric loss value of the coal-gangue medium from the denoised signal of the coal-gangue medium;

[0040] Step S35: Perform frequency loss correction on the excitation frequency of the coal-gangue medium corresponding to the coal-gangue sample based on the dielectric loss value of the coal-gangue medium to obtain the corrected excitation frequency of the dielectric loss corresponding to the coal-gangue sample.

[0041] Further, Step S34 includes the following steps:

[0042] Step S341: Obtain the corresponding amplitude change and phase change of the coal-gangue medium signal from the denoised signal of the coal-gangue medium;

[0043] Step S342: Calculate the signal change rate of the denoised signal of the coal-gangue medium based on the amplitude change and phase change of the coal-gangue medium signal to obtain the signal change rate of the coal-gangue medium;

[0044] Step S343: Perform statistical analysis on the peak value of the conductance frequency of the denoised signal of the coal-gangue medium to obtain the peak value of the conductance frequency of the coal-gangue medium signal;

[0045] Step S344: Perform loss quantization calculation on the denoised signal of the coal-gangue medium using the dielectric loss calculation formula based on the signal change rate of the coal-gangue medium and the peak value of the conductance frequency of the coal-gangue medium signal to obtain the dielectric loss value of the coal-gangue medium.

[0046] Further, the dielectric loss calculation formula described in Step S344 is specifically:

[0047]

[0048] In the formula, Δ∈ f is the dielectric loss value of the coal-gangue medium, f1 is the initial value of the dielectric loss frequency range, f2 is the end value of the dielectric loss frequency range, f(t) is the frequency of the coal-gangue medium signal at time t of the denoised signal of the coal-gangue medium, is the signal change rate of the coal-gangue medium, α is the weight factor affecting the signal change rate, σ(f) is the conductivity of the coal-gangue medium at frequency f, β is the weight factor affecting the conductivity, f m$\omega$ is the peak value of the conductance frequency of the coal-gangue medium signal, $\gamma$ is the attenuation factor of the conductance peak value, exp is the exponential function, $f_0$ is the reference frequency of the coal-gangue medium, $\Delta f$ is the bandwidth of the signal frequency distribution, $\delta$ is the attenuation coefficient of the signal frequency distribution, and $\eta$ is the correction coefficient of the dielectric loss value of the coal-gangue medium.

[0049] Further, step S4 includes the following steps:

[0050] Step S41: Construct the frequency shift characteristic vector X of the coal-gangue medium = [U R1 , f R1 , U R2 , f R2 through the effective value of the voltage across the resistor corresponding to the coal-gangue sample and the dielectric loss correction excitation frequency, where U R1 is the effective value of the voltage across the resistor corresponding to the coal sample, U R2 is the effective value of the voltage across the resistor corresponding to the gangue sample, f R1 is the dielectric loss correction excitation frequency corresponding to the coal sample, and f R2 is the dielectric loss correction excitation frequency corresponding to the gangue sample;

[0051] Step S42: Use the support vector machine, random forest, and convolutional neural network to construct a coal-gangue classification and recognition model based on the frequency shift characteristic vector X of the coal-gangue medium, and identify and analyze the change trend rate corresponding to the effective value of the voltage across the resistor corresponding to the coal-gangue sample at the dielectric loss correction excitation frequency. If the effective value of the voltage across the resistor corresponding to the coal-gangue sample decreases as the dielectric loss correction excitation frequency increases and the change trend rate is relatively severe, it is a gangue sample; otherwise, it is a coal sample, and the dielectric classification results corresponding to the coal and gangue samples are output.

[0052] The beneficial effects of the present invention:

[0053] The coal-gangue dielectric identification method based on the frequency shift characteristic proposed by the present invention, compared with the prior art, the beneficial effects of the present application are as follows: By obtaining coal samples and gangue samples and conducting dielectric analysis on these samples, the relative dielectric constants of the coal-gangue samples are obtained. This process can help researchers deeply understand the physical properties of coal and gangue by analyzing the dielectric characteristics of the coal-gangue samples, especially their responses under different environmental conditions (such as changes in humidity, temperature, etc.). This is crucial for the accurate characterization of coal and gangue and related applications (such as resource utilization, harmless treatment of coal and gangue, etc.). The relative dielectric constant of the coal-gangue sample is closely related to its composition, structure, porosity, mineral composition, etc. By accurately measuring these parameters, it can provide a basis for subsequent coal-gangue treatment technologies. For example, by measuring the dielectric constant of coal and gangue, its response in the electromagnetic field can be predicted, and then an efficient coal-gangue detection technology can be designed. When analyzing the relative dielectric constant of coal-gangue media using a variable dielectric type capacitive sensor, different plate structures (such as series type, parallel type, and series-parallel type) can provide accurate quantification of the changes in the dielectric characteristics of coal-gangue media, ensuring the accuracy of subsequent processing. Secondly, by establishing a dual-frequency excitation RC circuit based on the influence of coal-gangue media on capacitance and calculating the effective value of the voltage, more accurate numerical support for the electrical characteristics of coal and gangue can be provided based on the change of capacitance caused by coal-gangue media. The dielectric characteristics of coal and gangue directly affect the capacitance value. Therefore, through a variable dielectric type capacitive sensor, the influence of coal-gangue samples on capacitance can be clearly observed. By cooperating a multi-frequency signal generator with a resistor, the reaction of coal and gangue in the electric field can be accurately simulated, further improving the accuracy of coal-gangue analysis. For calculating the effective value of the voltage, it can reflect the electrical response of coal-gangue media under specific frequency excitation, thus providing important data support for subsequent frequency loss correction and classification model establishment. In addition, through the calculation of the effective value of the voltage, the electrical characteristics of coal-gangue samples, especially their responses under complex environmental conditions, can be deduced, providing strong support for coal-gangue detection and classification. Then, the coal-gangue medium response signal corresponding to the variable dielectric type capacitive sensor in the coal-gangue medium dual-frequency excitation RC circuit and the coal-gangue medium excitation frequency corresponding to the coal-gangue sample are collected by a data recorder, and the coal-gangue medium response signal is processed by VMD-Median signal denoising. Based on the denoised signal, the dielectric loss value of coal and gangue is calculated, and the frequency loss correction of the excitation frequency is carried out. The role of signal denoising cannot be ignored. It can effectively reduce the influence of environmental noise, instrument errors, and interference signals during the experiment, thus ensuring the accuracy of the experimental results.Through the VMD-Median noise reduction method, different frequency components can be finely separated, useful signals can be retained, while low-frequency noise and high-frequency interference can be eliminated. Further, by calculating the dielectric loss value from the noise-reduced signal, the energy loss situation of the coal-gangue sample at a specific frequency can be revealed. Frequency loss correction helps to improve the accuracy of the test results, making the experimental results more in line with the actual characteristics of coal-gangue, thereby effectively reducing the error caused by frequency loss, and thus improving the accuracy and reliability of subsequent coal-gangue dielectric identification and analysis. Finally, by combining the effective voltage value of the coal-gangue sample and the excitation frequency of frequency loss correction, a characteristic vector of the frequency shift characteristics of the coal-gangue medium is constructed, and then a coal-gangue classification and identification model is developed. By constructing the characteristic vector, the electrical characteristics of coal-gangue can be effectively characterized, ensuring that different types of coal and gangue can be accurately distinguished. The frequency shift characteristics of the coal-gangue sample are the key parameters for characterizing its dielectric characteristics, and through the combination of the effective voltage value and the corrected excitation frequency, coal-gangue can be comprehensively analyzed from multiple dimensions. The significance of establishing a classification and identification model lies in being able to automatically and accurately classify coal-gangue and coal samples, reducing the burden of manual processing, and thus improving the efficiency and accuracy of coal-gangue dielectric identification and classification. Brief Description of the Drawings

[0054] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments read in conjunction with the accompanying drawings:

[0055] Figure 1 It is a schematic flow chart of the steps of the coal-gangue dielectric identification method based on frequency shift characteristics of the present invention;

[0056] Figure 2 It is a schematic diagram of the dual-frequency excitation RC circuit of the coal-gangue medium of the present invention. Detailed Description of the Specific Embodiment

[0057] The technical method of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0058] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0059] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0060] To achieve the above object, please refer to Figures 1 to 2 , the present invention provides a coal gangue dielectric identification method based on frequency shift characteristics, and the method includes the following steps:

[0061] Step S1: Obtain coal gangue samples, including coal samples and gangue samples, and analyze the dielectric constant of the medium of the coal gangue samples to obtain the relative dielectric constant of the coal gangue medium; based on the relative dielectric constant of the coal gangue medium, use a variable dielectric capacitive sensor to calculate the change in capacitance of the medium when a coal gangue sample is added between the two electrodes to obtain the change in capacitance after the change of the coal gangue medium corresponding to different electrode structures, where different electrode structures include series type, parallel type, and series-parallel type;

[0062] Step S2: Connect a multi-frequency signal generator and a resistor between the corresponding two electrodes of the variable dielectric capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for the coal gangue medium; calculate the effective value of the voltage across the resistor corresponding to the coal gangue sample based on the change in capacitance after the change of the coal gangue medium for the dual-frequency excitation RC circuit of the coal gangue medium.

[0063] Step S3: Collect the response signal of the coal gangue medium corresponding to the variable dielectric capacitive sensor in the dual-frequency excitation RC circuit of the coal gangue medium and the excitation frequency of the coal gangue medium corresponding to the coal gangue sample through a data recorder, and perform VMD-Median signal denoising on the response signal of the coal gangue medium to generate a denoised signal of the coal gangue medium; obtain the dielectric loss value of the coal gangue medium through the denoised signal of the coal gangue medium, and perform frequency loss correction on the excitation frequency of the coal gangue medium corresponding to the coal gangue sample based on the dielectric loss value of the coal gangue medium to obtain the corrected excitation frequency of the dielectric loss corresponding to the coal gangue sample.

[0064] Step S4: Construct a characteristic vector of the frequency shift characteristics of the coal gangue medium through the effective value of the voltage across the resistor corresponding to the coal gangue sample and the corrected excitation frequency of the dielectric loss, and construct a classification and identification model for the coal gangue based on the characteristic vector of the frequency shift characteristics of the coal gangue medium to output the dielectric classification results corresponding to the coal and gangue samples.

[0065] In the embodiments of the present invention, please refer to Figure 1As shown in the figure, it is a schematic diagram of the step flow of the coal gangue dielectric identification method based on the frequency shift characteristic of the present invention. In this example, the coal gangue dielectric identification method based on the frequency shift characteristic includes the following steps:

[0066] Step S1: Obtain coal gangue samples, including coal samples and gangue samples, and analyze the dielectric constant of the medium of the coal gangue samples to obtain the relative dielectric constant of the coal gangue medium; based on the relative dielectric constant of the coal gangue medium, use a variable dielectric type capacitive sensor to calculate the change in capacitance of the medium when the coal gangue sample is added between the two electrodes to obtain the change in capacitance after the coal gangue medium changes corresponding to different electrode structures, where different electrode structures include series type, parallel type, and series-parallel type;

[0067] In the embodiment of the present invention, coal samples and gangue samples are obtained from coal mining areas or mines. These samples need to ensure representativeness and meet the requirements of subsequent experiments. Subsequently, a capacitance measurement device is used to analyze the dielectric constant of the medium of these coal gangue samples. The specific operation is to place the coal gangue samples between the electrodes of the capacitive sensor, and a measuring instrument is used to obtain the relative dielectric constant of the medium of the sample. The obtained relative dielectric constant of the medium reflects the electrical properties of the coal gangue samples. Next, using a variable dielectric type capacitive sensor, the coal gangue samples are respectively placed in capacitors with different electrode structures for measurement. These structures include series electrode structures, parallel electrode structures, and series-parallel electrode structures. For each electrode structure, record the change in capacitance after the sample medium changes, and calculate the change in capacitance under different structures through a formula. This change reflects the capacitance characteristics of the coal gangue samples, and finally obtain the change in capacitance after the coal gangue medium changes corresponding to different electrode structures, where different electrode structures include series type, parallel type, and series-parallel type.

[0068] Step S2: Connect a multi-frequency signal generator and a resistor between the corresponding two electrodes of the variable dielectric type capacitive sensor and perform abstraction processing on the excitation circuit to generate a coal gangue medium dual-frequency excitation RC circuit; based on the change in capacitance after the coal gangue medium changes, calculate the effective value of the coal gangue voltage of the coal gangue medium dual-frequency excitation RC circuit to obtain the effective value of the voltage across the resistor corresponding to the coal gangue sample;

[0069] In the embodiment of the present invention, by connecting a multi-frequency signal generator and a resistor R between the two plates of a variable-dielectric capacitive sensor, a complete coal-gangue dielectric dual-frequency excitation RC circuit is formed. An appropriate resistor is connected in series in the circuit to ensure the stability of the signal. The signal generated by the excitation circuit can make the coal-gangue sample form a dual-frequency excitation response in the capacitive sensor. For the design of the excitation circuit, according to the dielectric response characteristics of the coal-gangue sample, by adjusting the signal frequency and the resistance of the circuit, a stable voltage change is obtained. By calculating the changed capacitance, based on the change of the relative dielectric constant of the medium, the effective value of the voltage of the coal-gangue sample is calculated. Through this process, the obtained effective value of the voltage provides electrical characteristic data for subsequent analysis, further helping to determine the resistance characteristics of the coal-gangue sample, and finally obtaining the effective value of the voltage across the two ends of the resistor corresponding to the coal-gangue sample.

[0070] Step S3: Use a data recorder to collect the coal-gangue medium response signal corresponding to the variable-dielectric capacitive sensor in the coal-gangue dielectric dual-frequency excitation RC circuit and the coal-gangue medium excitation frequency corresponding to the coal-gangue sample, and perform VMD-Median signal denoising on the coal-gangue medium response signal to generate a coal-gangue medium denoised signal; obtain the coal-gangue medium dielectric loss value through the coal-gangue medium denoised signal, and perform frequency loss correction on the coal-gangue medium excitation frequency corresponding to the coal-gangue sample based on the coal-gangue medium dielectric loss value to obtain the medium loss correction excitation frequency corresponding to the coal-gangue sample;

[0071] In the embodiment of the present invention, by using a data recorder to collect the coal-gangue medium response signal and the excitation frequency, the data recorder should ensure that it can accurately record the voltage signal in the dual-frequency excitation RC circuit and further analyze its changes. For the collected signal, first, denoising is performed through the VMD (Variational Mode Decomposition) and Median Filtering processing methods. The VMD method can effectively separate each frequency band in the signal and decompose the signal into multiple Intrinsic Mode Functions (IMFs) to reduce the influence of high-frequency noise. Subsequently, median filtering is applied to each decomposed signal to further remove the remaining noise, thereby generating a coal-gangue medium denoised signal. The processed coal-gangue medium denoised signal is used to calculate the dielectric loss value of the coal-gangue sample. Based on the denoised signal and the dielectric loss value, the excitation frequency of the coal-gangue sample is corrected. The corrected frequency is the medium loss correction excitation frequency. Through this frequency correction, it can ensure that the electrical response of the coal-gangue sample in the experiment is more accurate, eliminate the error caused by frequency loss, and finally obtain the medium loss correction excitation frequency corresponding to the coal-gangue sample.

[0072] Step S4: Construct a characteristic vector of the frequency shift characteristics of coal gangue media based on the effective value of the voltage across the resistor corresponding to the coal gangue sample and the excitation frequency corrected by the dielectric loss, and construct a classification and recognition model for coal gangue based on the characteristic vector of the frequency shift characteristics of coal gangue media to output the dielectric classification results corresponding to the coal and gangue samples.

[0073] In the embodiment of the present invention, a characteristic vector of the frequency shift characteristics of coal gangue samples is constructed based on the effective value of the voltage across the resistor corresponding to the coal gangue samples and the excitation frequency corrected by the dielectric loss. This characteristic vector contains the voltage response of the coal gangue samples at specific frequencies and can reflect the dielectric properties of the samples. These characteristic vectors are input into a machine learning model for model training. Through learning the characteristics of a large number of samples, the model can identify the dielectric property differences between coal and gangue and assign a classification label to each sample. Through the output of the classification model, it can be accurately determined whether the sample is a coal sample or a gangue sample. This classification result helps to automatically identify coal and gangue in practical applications and finally outputs the dielectric classification results corresponding to the coal and gangue samples.

[0074] Further, step S1 includes the following steps:

[0075] Step S11: Obtain coal gangue samples, including coal samples and gangue samples;

[0076] In the embodiment of the present invention, coal samples and gangue samples are collected from the coal mine excavation site. During collection, it should be ensured that the samples are representative, that is, extracted from different coal mine levels and different locations to ensure the diversity of coal gangue samples. The coal samples should have a standard particle size range, generally between 2 mm and 5 mm, to avoid collecting overly fine or overly coarse coal blocks. The gangue samples need to be classified according to different mineral compositions to reflect the physical and chemical properties of coal gangue. After sampling, all samples need to be immediately stored in a dry and pollution-free environment to prevent external factors from changing the properties of the samples. After the samples are prepared, the next step can be entered, and finally, coal gangue samples are obtained.

[0077] Step S12: Analyze the dielectric constant of the medium of the coal gangue samples to obtain the relative dielectric constant ε of the coal gangue medium jec ;

[0078] In an embodiment of the present invention, a dielectric spectrometer or a dielectric constant measuring device is used to measure the dielectric constant of a corresponding coal gangue sample. By using a device with a frequency range of 100 Hz to 1 GHz, the dielectric constant of the coal gangue sample at different frequencies is measured. During analysis, the coal gangue sample is processed into a form suitable for testing (such as powder or flake), and necessary pre-treatment is carried out according to the actual situation of the material, such as drying or crushing. By measuring the capacitance response of the coal gangue sample at different frequencies, the frequency-dependent characteristics of its dielectric constant can be obtained, and the relative dielectric constant of the coal gangue can be calculated through a corresponding formula. For the coal sample, the same method is used for testing to obtain its dielectric constant, and finally the relative dielectric constant ε of the coal gangue medium is obtained. jec 。

[0079] Step S13: Obtain the dielectric constant ε of the medium inside the electrode plate obj ; If it is determined that the corresponding medium structure between the two electrode plates in the variable medium type capacitive sensor is a series structure, then it is considered that the variable medium type capacitive sensor is an upper and lower series structure between the corresponding medium of the coal gangue sample and the medium between the electrode plates, so as to obtain the electrode plate area A, the distance d1 of the added coal gangue medium, and the distance d0 of the medium inside the electrode plate, and based on the relative dielectric constant ε of the coal gangue medium jec perform a series capacitance change calculation on the electrode plate area A, the distance d1 of the added coal gangue medium, the distance d0 of the medium inside the electrode plate, and the dielectric constant ε of the medium inside the electrode plate obj to obtain the capacitance change amount after the change of the coal gangue medium corresponding to the series type electrode plate structure;

[0080] In an embodiment of the present invention, by measuring and obtaining the dielectric constant ε of the medium inside the electrode plate in the capacitive sensor obj , usually a precision capacitance measuring device is used to obtain the dielectric constant value of the medium. If it is determined that the medium structure between the two electrode plates in the variable medium type capacitive sensor is a series structure, the coal gangue medium is used as a part of the medium. At this time, the coal gangue medium and the medium between the electrode plates in the sensor will form an upper and lower series structure. In this structure, the change of the capacitance is mainly affected by the relative dielectric constant of the coal gangue medium, the electrode plate area A, the distance d1 of the coal gangue medium, and the distance d0 of the medium inside the electrode plate. By establishing a capacitance calculation model, according to the capacitance calculation formula of the series structure, based on the relative dielectric constant of the coal gangue medium and other known parameters, the capacitance change amount caused by the coal gangue medium is calculated. Specifically, according to the capacitance formula, the capacitance change between the coal gangue medium and the electrode plate is calculated, so as to obtain the capacitance change amount of the series type electrode plate structure, and finally the capacitance change amount after the change of the coal gangue medium corresponding to the series type electrode plate structure is obtained.

[0081] Step S14: If it is determined that the corresponding dielectric structure between the two plates in the variable-dielectric capacitive sensor is a parallel structure, then the variable-dielectric capacitive sensor is considered to be a left-right parallel structure between the corresponding dielectric of the coal gangue sample and the dielectric between the plates, so as to obtain the area A1 occupied by the coal gangue dielectric plate, the area A2 occupied by the dielectric plate inside the plate, and the distance d between the two plates, and based on the relative dielectric constant ε of the coal gangue dielectric jec Perform calculations on the area A1 occupied by the coal gangue dielectric plate, the area A2 occupied by the dielectric plate inside the plate, the distance d between the two plates, and the dielectric constant ε of the dielectric inside the plate obj to calculate the change in capacitance after the change of the coal gangue dielectric corresponding to the parallel plate structure;

[0082] In the embodiment of the present invention, if the dielectric structure inside the capacitive sensor is confirmed to be a parallel structure, then the dielectric corresponding to the coal gangue sample will form a parallel structure with the dielectric between the plates. In the parallel structure, the change in capacitance is mainly determined by the area A1 occupied by the coal gangue dielectric, the area A2 occupied by the dielectric inside the plate, the distance d between the two plates, and the dielectric constant ε of the dielectric inside the plate jec Firstly, obtain the specific values of each parameter, including the area of the plate occupied by the coal gangue dielectric, the area of the plate occupied by the dielectric inside the plate, and the distance between the two plates, etc. Then, based on the relative dielectric constant of the coal gangue dielectric, calculate the capacitance change under each parameter. Using the parallel capacitance formula and combining the known data, calculate the capacitance change caused by the coal gangue dielectric, further understand the influence of the coal gangue on the performance of the capacitive sensor, and finally obtain the change in capacitance after the change of the coal gangue dielectric corresponding to the parallel plate structure.

[0083] Step S15: If it is determined that the corresponding dielectric structure between the two plates in the variable-dielectric capacitive sensor is a series-parallel structure, then when designing the capacitor, under the condition that it is allowed, the plate thickness is set to less than 0.5 mm to reduce the influence of the edge effect on the capacitance, calculate the actual value of the corrected capacitance, and perform series-parallel capacitance change calculation based on the actual value of the corrected capacitance to obtain the change in capacitance after the change of the coal gangue dielectric corresponding to the series-parallel plate structure.

[0084] In an embodiment of the present invention, if the dielectric structure inside the variable dielectric capacitive sensor is a series-parallel structure, a suitable capacitor structure needs to be designed. During the design process, the thickness of the electrode plate needs to be minimized as much as possible to reduce the influence of edge effects on the capacitance measurement results. Edge effects will cause uneven electric field distribution, thereby affecting the accuracy of the capacitance. Therefore, by adopting a thin electrode plate design, this influence can be reduced and the accuracy of the capacitance measurement results can be ensured. The corrected capacitance value will be calculated based on the actual measurement results. During the calculation process, the relative dielectric constant of the dielectric and other parameters under the series-parallel structure are considered, and the capacitance change caused by the change of coal gangue medium is calculated through the formula of series-parallel capacitance, so as to evaluate the influence of coal gangue on the capacitive sensor, and finally the capacitance change after the change of coal gangue medium corresponding to the series-parallel electrode plate structure is obtained.

[0085] Further, in step S13, based on the relative dielectric constant ε of the coal gangue medium jec For the electrode plate area A, the distance d1 between the added coal gangue media, the distance d0 between the media inside the electrode plate, and the dielectric constant ε of the media inside the electrode plate obj The calculation of the series capacitance change includes the following steps:

[0086] According to the relative dielectric constant ε of the coal gangue medium jec 、the electrode plate area A, the distance d1 between the added coal gangue media, and the dielectric constant ε of the media inside the electrode plate obj Calculate the capacitance of the upper medium to obtain the capacitance of the upper medium of the added coal gangue

[0087] In an embodiment of the present invention, by determining the relative dielectric constant of the coal gangue medium and the area of the electrode plate, coal gangue, as a dielectric material, has a certain dielectric constant, which can be obtained through experiments or data search. The area of the electrode plate is usually determined by the actual capacitor design and is usually known at the design stage. The thickness or distance of the coal gangue medium is also a physical quantity set during the design process. When calculating the capacitance of the upper medium, the capacitance formula Is used for calculation to obtain the capacitance of the upper medium of the added coal gangue

[0088] Preferably, according to the electrode plate area A, the distance d0 between the media inside the electrode plate, and the dielectric constant ε of the media inside the electrode plate obj Calculate the capacitance of the lower medium to obtain the capacitance of the lower medium inside the electrode plate

[0089] In the embodiment of the present invention, by obtaining the relative permittivity of the medium inside the electrode plate, this information is usually obtained from the physical property table of the material. Common media inside the electrode plate such as air or other common materials, and their permittivities are usually known data. Next, obtain the area of the electrode plate and the thickness of the medium inside the electrode plate. These parameters are obtained from the specific design of the capacitor or actual measurement. When actually calculating, the capacitance formula is still used for calculation Finally, the capacitance of the lower medium inside the electrode plate is obtained

[0090] Preferably, according to the capacitance operation principle corresponding to the series circuit, the capacitance of the upper medium C1 added with coal gangue and the capacitance of the lower medium C2 inside the electrode plate are summarized and calculated to obtain the total capacitance value of the series structure

[0091] In the embodiment of the present invention, by adopting the principle of series capacitance operation, the upper medium capacitance and the lower medium capacitance are summarized. According to the capacitance calculation principle in the series circuit, when multiple capacitors are in series, the calculation formula for the total capacitance is Finally, the total capacitance value of the series structure is calculated

[0092] Preferably, obtain the initial capacitance of the series structure corresponding to when no coal gangue medium is added And based on the initial capacitance C of the series structure c0 For the total capacitance value C of the series structure c Perform series capacitance change calculation to obtain the capacitance change amount after the change of the coal gangue medium corresponding to the series electrode plate structure

[0093] In the embodiment of the present invention, by obtaining the initial capacitance value of the series structure when no coal gangue medium is added, this value is usually measured or calculated under the design conditions before the coal gangue medium is added. The initial capacitance can be obtained by modeling and calculating the initial electrode plate structure of the capacitor And after obtaining the initial capacitance value, use the previously calculated total series capacitance, compare the differences between them. The calculation of the capacitance change amount is based on the series capacitance calculation formula. Specifically, by comparing the capacitance values before and after adding the coal gangue medium, the influence of the coal gangue medium on the overall capacitance can be obtained. The magnitude of this change amount directly affects the working performance of the capacitor, especially in terms of frequency response and voltage regulation, and calculate the capacitance change amount after the change of the coal gangue medium Finally, the capacitance change amount ΔC after the change of the coal gangue medium corresponding to the series electrode plate structure is obtained c .

[0094] Furthermore, in step S14, based on the relative permittivity ε of the coal gangue medium jecFor the area A1 occupied by the added coal gangue medium plate, the area A2 occupied by the medium plate inside the plate, the distance d between the two plates, and the dielectric constant ε of the medium inside the plate obj The calculation of the change in the parallel capacitance includes the following steps:

[0095] According to the relative dielectric constant ε of the coal gangue medium jec , the area A1 occupied by the added coal gangue medium plate, the distance d between the two plates, and the dielectric constant ε of the medium inside the plate obj Perform the left medium capacitance calculation to obtain the added coal gangue left medium capacitance

[0096] In the embodiment of the present invention, by collecting the basic physical parameters of the coal gangue medium, including the relative dielectric constant ε of the coal gangue medium jec , the area A1 occupied by the plate, the distance d between the two plates, and the dielectric constant ε of the medium inside the plate obj , when calculating the coal gangue left medium capacitance, use the capacitance formula for calculation to obtain the added coal gangue left medium capacitance C′1.

[0097] Preferably, according to the area A2 occupied by the medium plate inside the plate, the distance d between the two plates, and the dielectric constant ε of the medium inside the plate obj Perform the right medium capacitance calculation to obtain the right medium capacitance inside the plate

[0098] In the embodiment of the present invention, by considering the dielectric material inside the plate, its calculation method is similar to the previous one. First, relevant parameters of the dielectric inside the plate need to be collected, including the dielectric constant ε of the dielectric inside the plate obj , the area A2 occupied by the plate, and the distance d between the two plates, and perform the calculation by using the corresponding capacitance formula Finally, obtain the right medium capacitance C′2 inside the plate.

[0099] Preferably, perform the capacitance summation calculation on the added coal gangue left medium capacitance C′1 and the right medium capacitance C′2 inside the plate according to the capacitance operation principle corresponding to the parallel circuit to obtain the total capacitance value of the parallel structure

[0100] In the embodiment of the present invention, by calculating the previously obtained left medium capacitance and right medium capacitance (calculated according to the capacitance operation principle of the parallel circuit, according to the parallel capacitance formula, the total capacitance of the two capacitors can be calculated by the following formula C′1C′2, and finally obtain the total capacitance value of the parallel structure

[0101] Preferably, obtain the initial capacitance of the parallel structure corresponding to when no coal gangue medium is added And based on the initial capacitance C of the parallel structure b0For the total capacitance value C of the parallel structure b Perform calculations on the change in parallel capacitance to obtain the change in capacitance after the change in the coal gangue medium corresponding to the parallel plate structure

[0102] In the embodiment of the present invention, by obtaining the initial capacitance of the parallel structure when no coal gangue medium is added Assuming that no coal gangue medium is added, only the capacitance of the medium inside the plate exists under the entire plate area, and the total capacitance at this time is Next, based on the total capacitance of the parallel structure obtained previously, calculate the initial capacitance of the corresponding parallel structure, and calculate the change in capacitance after adding the coal gangue medium Finally, obtain the change in capacitance after the change in the coal gangue medium corresponding to the parallel plate structure

[0103] Furthermore, step S15 includes the following steps

[0104] Step S151: If it is determined that the dielectric structure between the two plates of the variable dielectric capacitive sensor is a series-parallel structure, then obtain the dielectric surface area S between the two plates of the variable dielectric capacitive sensor and the dielectric spacing D between the two plates

[0105] In the embodiment of the present invention, when it is determined that the dielectric structure between the two plates of the variable dielectric capacitive sensor is a series-parallel structure, first, the dielectric surface area between the two plates and the dielectric spacing between the two plates need to be obtained. Specifically, first measure or calculate the shape and size of the two plates of the capacitive sensor to obtain the area of the plates. This is usually obtained by geometric calculation of the surface area of the plates. If the electrode plates are rectangular or circular, they can be calculated by formulas (for example, rectangular area = length × width, circular area = π × radius²). In addition, use instruments with micron-level precision such as laser rangefinders or precision thickness gauges to measure the spacing between the two plates to ensure the accuracy of the spacing. The two data obtained are the physical parameters necessary for subsequent calculation of the series-parallel initial capacitance, and finally obtain the corresponding dielectric surface area S between the two plates and the dielectric spacing D between the two plates

[0106] Step S152: Perform calculations on the series-parallel initial capacitance according to the dielectric surface area S between the two plates and the dielectric spacing D between the two plates to obtain the series-parallel structure initial capacitance where ε f is the dielectric constant between the two plates, and k is the dielectric capacitance influence constant

[0107] In the embodiment of the present invention, after obtaining the dielectric surface area and dielectric spacing between the two plates, it enters the calculation stage of the series-parallel initial capacitance. Specifically, according to the capacitance formula Calculations are carried out. Here, considering that the dielectric structure in the sensor is a series-parallel structure, it is necessary to calculate the capacitance contributions of different types of dielectrics in this structure separately. According to the specific dielectric types (such as the dielectric constant of coal gangue medium and the corresponding capacitance influence constant), the capacitance calculations of different dielectrics are processed in segments, and the initial capacitance of the corresponding dielectric is calculated using the above formula, and finally the initial capacitance of the series-parallel structure is obtained.

[0108] Step S153: When designing this capacitor, if conditions permit, the plate thickness is set to less than 0.5 mm to reduce the influence of edge effect on the capacitance, and the edge effect influence coefficient ξ = 0.163 is calculated, and based on the edge effect influence coefficient ξ and the initial capacitance C of the series-parallel structure h0 to calculate the actual value of the corrected capacitance

[0109] In the embodiment of the present invention, when designing the capacitor, in order to reduce the influence of the edge effect on the capacitance value, the thickness of the plate should be controlled within a relatively small range as much as possible. Through precision machining techniques (such as CNC milling or electroforming techniques), an electrode plate with a thickness as small as possible is manufactured, usually requiring the thickness to be controlled within 0.5 mm. The edge effect refers to the uneven distribution of the electric field at the edges of the plate, resulting in an error in the capacitance value. In order to quantify this influence, the edge effect influence coefficient calculation formula is used, and the known coefficient is 0.163. This coefficient is multiplied by the previously obtained series-parallel initial capacitance to calculate the actual value of the corrected capacitance. In this step, the actual correction process is completed through computer simulation (such as the finite element analysis method) or by using a specific capacitance correction model, and finally the actual value of the corrected capacitance is calculated.

[0110] Step S154: Based on the initial capacitance C of the series-parallel structure h0 for the actual value C of the corrected capacitance h carry out the calculation of the change in the series-parallel capacitance to obtain the change in capacitance of the coal gangue medium corresponding to the series-parallel plate structure after the change

[0111] In the embodiment of the present invention, once the actual value of the corrected capacitance is obtained, the calculation of the change in the series-parallel capacitance is carried out next. The change in the capacitance of the series-parallel structure needs to be calculated in detail based on the characteristics of the coal gangue medium and the working principle of the capacitance sensor. By comparing the corrected capacitance value with the dielectric characteristics of the coal gangue medium after the change, the change in capacitance is calculated. At this time, the formula for the capacitance change rate is used for calculation, and finally the change in capacitance ΔC of the coal gangue medium corresponding to the series-parallel plate structure after the change is obtained h .

[0112] Further, step S2 includes the following steps:

[0113] Step S21: Connect a multi-frequency signal generator and a resistor between the two electrodes corresponding to the variable dielectric capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for coal gangue medium;

[0114] In the embodiment of the present invention, by connecting the two electrode plates of the variable dielectric capacitive sensor to a multi-frequency signal generator, which can provide voltage signals of multiple frequencies, the sensor can be multi-frequency excited. Specifically, the multi-frequency signal generator can be set to output a series of sine wave signals with known frequencies (for example, 10 kHz, 20 kHz, 50 kHz, etc.). A resistor is connected in parallel with this signal source, and the value of this resistor needs to be preset according to the conductivity of the coal gangue sample and optimized considering the characteristics of the capacitive sensor. The excitation circuit formed in this way can effectively perform dual-frequency excitation on the coal gangue medium. During the abstraction process, it is assumed that the circuit can be simplified into an RC circuit model, where the capacitance is composed of the capacitance of the variable dielectric capacitive sensor, and the resistance is jointly affected by the external resistor and the resistance characteristics of the coal gangue medium, and finally a dual-frequency excitation RC circuit for coal gangue medium is generated (as Figure 2 shown).

[0115] Step S22: Optionally select the capacitance change amount after the change of the coal gangue medium corresponding to different electrode plate structures as the capacitance change amount ΔC after the change of the medium of the dual-frequency excitation RC circuit for coal gangue medium, and calculate the capacitance frequency shift amount between the two electrodes corresponding to the variable dielectric capacitive sensor in the dual-frequency excitation RC circuit for coal gangue medium based on the capacitance change amount ΔC after the change of the medium to obtain the capacitance frequency shift amount of the excitation RC circuit where f is the frequency of the AC voltage;

[0116] In the embodiment of the present invention, through experiments according to the selected capacitance electrode plate structure, by adjusting the shape, area of the capacitor plate and the distance between the electrodes, the capacitance change amount after the change of the coal gangue medium under different coal gangue media is obtained as the capacitance change amount ΔC after the change of the medium corresponding to the dual-frequency excitation RC circuit for coal gangue medium, and this change amount is used to describe the influence of the dielectric on the RC circuit. Then, based on this capacitance change amount, the capacitance frequency shift amount is further calculated. This process involves accurately calculating the capacitance frequency response of the variable dielectric capacitive sensor. By measuring the capacitance change corresponding to the input multi-frequency signal, the capacitance frequency shift amount of the dual-frequency excitation RC circuit of the coal gangue medium at a specific frequency can be deduced backwards. Finally, the capacitance frequency shift amount X of the excitation RC circuit is obtained c 。

[0117] Step S23: Apply a standard sinusoidal AC voltage to the dual-frequency excitation RC circuit for coal gangue medium where B is the amplitude of the applied AC voltage and t is the time, is the initial phase angle, where In the series circuit corresponding to the coal gangue medium dual-frequency excitation RC circuit, the effective value phasor corresponding to the input voltage is where j is the phasor imaginary unit, and the standard sinusoidal AC voltage U s in At this time, U′ = B, the impedance of the resistor R in the coal gangue medium dual-frequency excitation RC circuit is Z R = R, and the impedance of the capacitor C is Z C = -jX c ;

[0118] In the embodiment of the present invention, by inputting a standard sinusoidal AC voltage into the coal gangue medium dual-frequency excitation RC circuit, the amplitude and frequency of the voltage signal are preset according to experimental requirements, usually set to the standard amplitude, such as 1V, and the initial phase angle is set to zero or a specific value, that is In the circuit, the amplitude and phase angle of the input voltage determine the voltage source characteristics of the input signal. At this time, the circuit can be simplified to a series RC circuit including a resistor R and a capacitor C. In this series circuit, the effective value of the input voltage can be expressed as a phasor, and by calculating the effective value of the phasor, the effective value phasor corresponding to the input voltage is obtained as to study the response behavior of the circuit. According to the basic impedance theory of the circuit, the standard sinusoidal AC voltage U s in At this time, U′ = B, the impedance of the resistor R is R, and the impedance of the capacitor C is jX c , where j is the phasor imaginary unit. At this time, the overall response of the circuit can be solved by the synthesis formula of the series impedance to further analyze the influence of the capacitor and the resistor on the circuit.

[0119] Step S24: According to the effective value phasor U′ of the input voltage, the impedance Z of the resistor R R and the impedance Z of the capacitor C C perform the calculation of the voltage effective value to obtain the voltage effective value across the resistor corresponding to the coal gangue sample

[0120] In the embodiment of the present invention, based on the circuit setting of the previous step, combined with the effective value phasor U′ of the input voltage, the impedance Z of the resistor R R and the impedance Z of the capacitor C C , perform the calculation of the voltage effective value. First, it is necessary to calculate the phasor of the input voltage. The effective value of the voltage is proportional to the magnitude of the phasor. By knowing the impedance of the resistor and capacitor in the circuit and applying the impedance calculation formula of the series circuit, the combined impedance of the entire circuit is obtained, and then the voltage effective value across the resistor is calculated During this calculation process, the effective value of the voltage across the resistor not only depends on the amplitude of the input voltage, but is also affected by the capacitor, specifically manifested as the phase change caused by the capacitor. Therefore, the calculation result needs to be corrected according to the actual capacitor change in order to accurately reflect the effective value of the voltage across the resistor of the coal gangue sample, where When calculating the modulus of this complex number, use the modulus calculation formula of the complex number Finally, the effective value of the voltage across the resistor corresponding to the coal gangue sample is calculated

[0121] Furthermore, step S3 includes the following steps:

[0122] Step S31: Collect the coal gangue medium response signal corresponding to the variable dielectric type capacitive sensor in the coal gangue medium dual-frequency excitation RC circuit and the coal gangue medium excitation frequency corresponding to the coal gangue sample through a data recorder;

[0123] In the embodiment of the present invention, the coal gangue medium response signal and the excitation frequency corresponding to the coal gangue sample are collected in real time through a data recorder. The data recorder is connected to the variable dielectric type capacitive sensor, and the sensor is installed in the dual-frequency excitation RC circuit and is used to collect the medium response signal of the coal gangue sample. The capacitive sensor obtains the response characteristics of the coal gangue medium to the electric field by measuring the capacitance change of the circuit. The recorder samples the signal through a high-precision sampler and records the coal gangue medium excitation frequency corresponding to the coal gangue sample. These excitation frequencies are usually set within a specific working frequency range of the coal gangue medium to ensure that the signal accurately reflects the dielectric characteristics of the coal gangue sample, and finally the coal gangue medium response signal and the coal gangue medium excitation frequency corresponding to the coal gangue sample are obtained.

[0124] Step S32: Use the variational mode decomposition (VMD) algorithm to perform VMD decomposition processing on the coal gangue medium response signal to obtain the coal gangue medium VMD decomposition signal spectrum;

[0125] In the embodiment of the present invention, the coal gangue medium response signal is decomposed by applying the variational mode decomposition (VMD) algorithm. The VMD algorithm adaptively decomposes the signal into multiple intrinsic mode functions (IMFs) with different frequency bands, and each mode function corresponds to different frequency components of the signal. Specifically, during operation, first input the coal gangue medium response signal into the VMD algorithm, and the algorithm automatically decomposes the signal according to the frequency characteristics of the signal. The signal after VMD decomposition is divided into several frequency bands, and each frequency band reflects the change of the signal within a specific frequency range. By analyzing the spectra of these frequency bands, the characteristics of the coal gangue medium signal at different frequencies can be deeply understood, and finally the coal gangue medium VMD decomposition signal spectrum is obtained.

[0126] Step S33: Perform Median signal reconstruction and noise reduction on the coal gangue medium VMD decomposition signal spectrum to generate a coal gangue medium noise reduction signal;

[0127] In the embodiment of the present invention, by performing noise reduction processing on the spectrum of the coal gangue medium VMD decomposition signal, the Median signal reconstruction method is used to reconstruct and denoise the signal. Specifically, during the operation, first, apply a median filter to each band signal after VMD decomposition. This method can effectively remove the impulse noise and high-frequency interference in the signal. The median filter smooths the signal by calculating the median value within the signal window to replace the outliers in the original signal, thereby reducing the influence of noise on the signal. After median filtering, the reconstructed signal obtained is the coal gangue medium signal after noise reduction processing, that is, the coal gangue medium noise reduction signal. While retaining the main features of the original signal, the noise components have been greatly weakened, and finally, the coal gangue medium noise reduction signal is generated.

[0128] Step S34: Obtain the dielectric loss value of the coal gangue medium through the coal gangue medium noise reduction signal;

[0129] In the embodiment of the present invention, calculate the dielectric loss value of the coal gangue medium through the coal gangue medium noise reduction signal. The dielectric loss value reflects the energy loss characteristics of the coal gangue medium in the electric field. First, input the coal gangue medium noise reduction signal into the loss calculation formula. During the calculation process, utilize the amplitude and phase information of the coal gangue medium signal. At the same time, the conductance value is extracted through the frequency response and the amplitude change of the signal, and the dielectric constant is set according to the electrical characteristics of the material. Through this process, the dielectric loss value of the coal gangue medium can be obtained, which directly reflects the energy absorption and loss characteristics of the coal gangue medium at a specific excitation frequency, and finally, the dielectric loss value of the coal gangue medium is obtained.

[0130] Step S35: Perform frequency loss correction on the excitation frequency of the coal gangue medium corresponding to the coal gangue sample based on the dielectric loss value of the coal gangue medium to obtain the dielectric loss correction excitation frequency corresponding to the coal gangue sample.

[0131] In the embodiment of the present invention, by performing frequency loss correction on the excitation frequency corresponding to the coal gangue sample based on the dielectric loss value of the coal gangue medium. First, evaluate the loss characteristics at different excitation frequencies through the obtained dielectric loss value of the coal gangue medium. According to the dielectric loss characteristics of the coal gangue medium, use the frequency correction formula to correct the excitation frequency. The correction formula utilizes the dielectric loss data of the coal gangue medium at different frequencies to adjust the frequency of the excitation signal to compensate for the deviation caused by the dielectric loss. During the specific calculation process, use the interpolation method to accurately fit the relationship between frequency and loss to obtain the excitation frequency after loss correction. The excitation frequency after frequency loss correction provides a more accurate analysis basis for the coal gangue sample, and finally, the dielectric loss correction excitation frequency corresponding to the coal gangue sample is obtained.

[0132] Furthermore, step S34 includes the following steps:

[0133] Step S341: Obtain the corresponding amplitude change and phase change of the coal gangue medium signal from the coal gangue medium noise reduction signal;

[0134] In the embodiment of the present invention, the coal gangue medium noise reduction signal is collected and transmitted to the signal processing system. Through high-precision signal processing instruments (such as oscilloscopes, spectrum analyzers, etc.), the Fourier transform is performed on the noise-reduced signal to obtain its frequency domain characteristics. Next, the amplitude of the signal is extracted to calculate the change amplitude of the coal gangue medium signal, that is, the intensity of the signal changing with time or frequency. Through phase difference calculation, the change phase of the signal is extracted, and then the phase shift situation of the signal with time is analyzed. Finally, the amplitude change of the coal gangue medium signal and the phase change of the coal gangue medium signal are obtained.

[0135] Step S342: Calculate the signal change rate of the coal gangue medium noise reduction signal based on the amplitude change and phase change of the coal gangue medium signal to obtain the coal gangue medium signal change rate;

[0136] In the embodiment of the present invention, by calculating the amplitude change rate and phase change rate of the coal gangue medium signal, the overall change rate of the signal is obtained. Specifically, during operation, first calculate the change amount of the signal's change amplitude and change phase per unit time, that is, the change amplitude per second and the phase change rate. Through the formula: change rate = (current value - initial value) / time, the change rate of the coal gangue medium signal is obtained. This change rate reflects the response speed of the coal gangue medium signal at different time points and helps to identify the trend of signal change. Finally, the coal gangue medium signal change rate is obtained.

[0137] Step S343: Conduct statistical analysis on the conductance frequency peak of the coal gangue medium noise reduction signal to obtain the coal gangue medium signal conductance frequency peak;

[0138] In the embodiment of the present invention, through the use of high-frequency conductance measurement equipment to perform frequency response analysis on the coal gangue medium noise reduction signal, the signal is analyzed through a specific conductance measurement instrument (such as an LCR meter or a conductance analyzer), the conductance frequency response curve of the signal is calculated, the peak value of this curve is statistically analyzed, and the conductance frequency peak of the signal is extracted. The conductance frequency peak represents the conductivity change characteristics of the coal gangue medium within a specific frequency range. Finally, the coal gangue medium signal conductance frequency peak is obtained.

[0139] Step S344: Perform loss quantization calculation on the coal gangue medium noise reduction signal based on the coal gangue medium signal change rate and the coal gangue medium signal conductance frequency peak using the dielectric loss calculation formula to obtain the coal gangue medium dielectric loss value.

[0140] In the embodiment of the present invention, by combining the initial value of the dielectric loss frequency range, the end value of the dielectric loss frequency range, the coal gangue medium signal frequency, the coal gangue medium signal change rate, the signal change rate influence weight factor, the coal gangue medium conductivity, the conductivity influence weight factor, the coal gangue medium signal conductance frequency peak value, the conductance peak attenuation factor, the coal gangue medium reference frequency, the signal frequency distribution bandwidth and related parameters, a suitable dielectric loss calculation formula is constructed to perform loss quantization calculation on the coal gangue medium noise reduction signal, so as to quantitatively obtain the dielectric loss value of the coal gangue medium. This loss value reflects the energy loss characteristics of the coal gangue medium in the electric field, can effectively evaluate the physical properties of the coal gangue, and finally obtains the dielectric loss value of the coal gangue medium.

[0141] Further, the dielectric loss calculation formula described in step S344 is specifically:

[0142]

[0143] In the formula, Δ∈ f is the dielectric loss value of the coal gangue medium, f1 is the initial value of the dielectric loss frequency range, f2 is the end value of the dielectric loss frequency range, f(t) is the coal gangue medium signal frequency of the coal gangue medium noise reduction signal at time t, is the coal gangue medium signal change rate, α is the signal change rate influence weight factor, σ(f) is the coal gangue medium conductivity at frequency f, β is the conductivity influence weight factor, f m is the coal gangue medium signal conductance frequency peak value, γ is the conductance peak attenuation factor, exp is the exponential function, f0 is the coal gangue medium reference frequency, Δf is the signal frequency distribution bandwidth, δ is the attenuation coefficient of the signal frequency distribution, and η is the correction coefficient of the coal gangue medium dielectric loss value.

[0144] The present invention obtains a dielectric loss calculation formula through the use of a specific mathematical model and verification, which is used to quantitatively calculate the loss of the noise reduction signal of coal gangue medium. This dielectric loss calculation formula quantifies the dielectric loss value of coal gangue medium through the signal change rate of coal gangue medium and the conductance frequency peak. This approach combines the change trend of the signal with the conductance characteristics of the signal at different frequencies, making the understanding of the coal gangue medium signal more comprehensive. The introduction of the conductance frequency peak can reflect the conductance characteristics of coal gangue medium at a specific frequency, which is very important for describing the energy propagation and electrical characteristics of the signal in this frequency band. By introducing this value, the loss analysis of the coal gangue medium signal in different frequency bands can be improved, making the calculation results more accurate. The formula also uses the frequency range from f1 to f2, which enables the calculation of dielectric loss to be carried out within a specified frequency bandwidth, ignoring irrelevant frequency bands. Precise definition of the frequency range helps to reduce the influence of irrelevant frequencies on loss quantification, thereby improving the calculation efficiency and the reliability of the results. By introducing these influence weight factors, the contribution of the coal gangue medium signal to dielectric loss can be flexibly adjusted according to the change rate of the signal and the characteristics of conductivity. This adjustment helps to dynamically reflect the actual loss of coal gangue medium according to the changes in different signal characteristics. An exponential decay function is also introduced This exponential function takes into account the bandwidth of the frequency distribution and controls the width of the signal frequency through the decay factor. This part reasonably handles the influence of the signal frequency change, ensuring that the contribution of the frequency change to the loss is accurately reflected when calculating the dielectric loss. In addition, the introduction of a correction factor can compensate for the error caused by the difference between the theoretical model and the actual situation, making the loss quantification closer to the actual measured value and improving the applicability and reliability of the formula. In summary, this formula fully considers the dielectric loss value Δ∈ f , the initial value f1 of the dielectric loss frequency range, the end value f2 of the dielectric loss frequency range, the coal gangue medium signal frequency f(t) of the coal gangue medium noise reduction signal at time t, the signal change rate of the coal gangue medium The signal change rate influence weight factor α, the conductivity σ(f) of the coal gangue medium at frequency f, the conductivity influence weight factor β, the conductance frequency peak f of the coal gangue medium signal m , the conductance peak decay factor γ, the exponential function exp, the reference frequency f0 of the coal gangue medium, the signal frequency distribution bandwidth Δf, the attenuation coefficient δ of the signal frequency distribution, the correction factor η of the dielectric loss value of the coal gangue medium. According to the dielectric loss value Δ∈ f and the mutual correlation relationship between the above parameters constitutes a functional relationship:

[0145]

[0146] This formula can achieve the loss quantization calculation process of the noise reduction signal of the coal-gangue medium. At the same time, by introducing the correction coefficient η of the dielectric loss value of the coal-gangue medium, it can be adjusted according to the error situation that appears in the calculation process, so as to improve the accuracy and applicability of the dielectric loss calculation formula.

[0147] Further, step S4 includes the following steps:

[0148] Step S41: Construct the frequency shift characteristic vector X = [U R1 , f R1 , U R2 , f R2 of the coal-gangue medium through the effective value of the voltage across the resistor corresponding to the coal-gangue sample and the dielectric loss correction excitation frequency, where U R1 is the effective value of the voltage across the resistor corresponding to the coal sample, U R2 is the effective value of the voltage across the resistor corresponding to the gangue sample, f R1 is the dielectric loss correction excitation frequency corresponding to the coal sample, f R2 is the dielectric loss correction excitation frequency corresponding to the gangue sample;

[0149] In the embodiment of the present invention, by preparing coal-gangue samples, the resistance of the coal-gangue samples is measured through an experimental device, and the corresponding effective voltage values are recorded. At this time, the measurement needs to be carried out at multiple different excitation frequencies to obtain the resistance response of the coal-gangue samples at each frequency. Next, the dielectric loss correction method is applied to correct the measured effective voltage value data to eliminate the error caused by the dielectric loss. Then, the relationship between the effective value of the voltage across the resistor of the coal-gangue sample and the excitation frequency is constructed, and the frequency shift characteristics of the coal-gangue are extracted according to this relationship. The frequency shift characteristic vector includes the effective value of the voltage across the resistor corresponding to the coal sample and the dielectric loss correction excitation frequency, the effective value of the voltage across the resistor corresponding to the gangue sample and the dielectric loss correction excitation frequency. These characteristics reflect the response of the coal-gangue samples to different frequencies. The key point of this step is to use high-precision measuring instruments to obtain the resistance response data of the coal-gangue, and through precise correction and data processing means, ensure the accuracy of the characteristic vector, and finally construct the frequency shift characteristic vector X = [U R1 , f R1 , U R2 , f R2 of the coal-gangue medium.

[0150] Step S42: Use a support vector machine, random forest, and convolutional neural network, and construct a coal-gangue classification and recognition model based on the coal-gangue medium frequency-shift characteristic feature vector x to identify and analyze the change trend rate of the effective voltage across the resistor corresponding to the coal-gangue sample at the dielectric loss correction excitation frequency. If the effective voltage across the resistor corresponding to the coal-gangue sample decreases as the dielectric loss correction excitation frequency increases, and the change trend rate is relatively drastic, it is a gangue sample; otherwise, it is a coal sample, so as to output the dielectric classification results corresponding to the coal and gangue samples.

[0151] In the embodiment of the present invention, by using the previously obtained coal-gangue medium frequency-shift characteristic feature vector, a classification and recognition model is constructed. To classify coal samples and gangue samples, three main machine learning algorithms are used for modeling, including support vector machine (SVM), random forest (Random Forest), and convolutional neural network (CNN). Each model uses the same coal-gangue frequency-shift characteristic feature vector as input for training and optimization. The support vector machine uses a non-linear kernel function to map the feature space to ensure separability in the high-dimensional space. The random forest is integrated by constructing multiple decision trees and uses the majority voting mechanism for final classification. The convolutional neural network extracts features using convolutional layers and learns the complex relationships between samples through multiple layers of neural networks. During the training process, it is divided into a training set, a validation set, and a test set according to a certain ratio. The training set is used to train the classification model and adjust the model parameters, such as the penalty parameter C and kernel function parameter γ of SVM, the number of decision trees and maximum depth of RF, the number of network layers and neurons of CNN, etc. The validation set is used to monitor the training process, and cross-validation is used to adjust the hyperparameters of the model to ensure that the performance of each classifier reaches the optimal state. After the model training is completed, the effective voltage data across the resistor of the coal-gangue sample to be classified is input, and the coal-gangue and coal samples are discriminated according to the change trend rate of the excitation frequency. If the effective voltage value of the coal-gangue sample decreases rapidly as the excitation frequency increases, and the change rate is large, the model will identify the sample as a gangue; if the change rate is small or there is no obvious change, it is identified as a coal sample. This classification result is presented through an output mechanism, and finally the dielectric classification results corresponding to the coal and gangue samples are output.

[0152] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be included in the present invention.

[0153] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A coal gangue dielectric identification method based on frequency shift characteristics, characterized in that, It includes the following steps: Step S1: Obtain coal gangue samples, including coal samples and gangue samples, and perform dielectric constant analysis on the coal gangue samples to obtain the relative dielectric constant of the coal gangue medium; based on the relative dielectric constant of the coal gangue medium, use a variable dielectric type capacitive sensor to calculate the change in medium capacitance by adding coal gangue samples between the two plates to obtain the change in capacitance after the coal gangue medium changes corresponding to different plate structures, where different plate structures include series type, parallel type, and series-parallel type; Step S2: Connect a multi-frequency signal generator and a resistor between the corresponding two plates of the variable dielectric type capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for the coal gangue medium; Calculate the effective value of the voltage across the resistor corresponding to the coal gangue sample by calculating the effective value of the voltage of the coal gangue medium dual-frequency excitation RC circuit based on the change in capacitance after the coal gangue medium changes; Step S3: Collect the response signal of the coal gangue medium corresponding to the variable dielectric type capacitive sensor in the coal gangue medium dual-frequency excitation RC circuit and the excitation frequency of the coal gangue medium corresponding to the coal gangue sample through a data recorder, and perform VMD-Median signal denoising on the response signal of the coal gangue medium to generate a denoised signal of the coal gangue medium; obtain the dielectric loss value of the coal gangue medium through the denoised signal of the coal gangue medium, and perform frequency loss correction on the excitation frequency of the coal gangue medium corresponding to the coal gangue sample based on the dielectric loss value of the coal gangue medium to obtain the corrected excitation frequency of the dielectric loss of the coal gangue sample; Step S4: Construct a feature vector of the frequency shift characteristics of the coal gangue medium based on the effective value of the voltage across the resistor corresponding to the coal gangue sample and the corrected excitation frequency of the dielectric loss, and construct a classification and recognition model for the coal gangue based on the feature vector of the frequency shift characteristics of the coal gangue medium to output the dielectric classification results corresponding to the coal and gangue samples.

2. The coal gangue dielectric identification method based on the frequency shift characteristic according to claim 1, wherein Step S1 includes the following steps: Step S11: Obtain coal gangue samples, including coal samples and gangue samples; Step S12: Analyze the dielectric constant of the coal gangue sample to obtain the relative dielectric constant ε of the coal gangue medium jec ; Step S13: Obtain the dielectric constant ε of the medium inside the electrode plate obj ; If it is determined that the corresponding medium structure between the two electrode plates in the variable medium type capacitive sensor is a series structure, then it is considered that the variable medium type capacitive sensor is an upper and lower series structure between the medium corresponding to the coal gangue sample added and the medium between the electrode plates, so as to obtain the electrode plate area A, the distance d1 between the added coal gangue media, and the distance d0 between the media inside the electrode plate, and based on the relative dielectric constant ε jec of the electrode plate area A, the distance d1 between the added coal gangue media, the distance d0 between the media inside the electrode plate, and the dielectric constant ε of the medium inside the electrode plate obj perform a series capacitance change calculation to obtain the capacitance change amount after the change of the coal gangue medium corresponding to the series electrode plate structure; Step S14: If it is determined that the dielectric structure between the two plates in the variable-dielectric capacitive sensor is a parallel structure, then the variable-dielectric capacitive sensor is considered to be a left-right parallel structure between the corresponding dielectric of the added coal gangue sample and the dielectric between the plates, so as to obtain the area A1 occupied by the coal gangue medium plate, the area A2 occupied by the dielectric plate in the plate, and the distance d between the two plates, and based on the relative dielectric constant ε of the coal gangue medium jec For the area A1 occupied by the coal gangue medium plate added, the area A2 occupied by the dielectric plate in the plate, the distance d between the two plates, and the dielectric constant ε of the dielectric in the plate obj Perform the calculation of the change in parallel capacitance to obtain the change in capacitance after the change of the coal gangue medium corresponding to the parallel plate structure; Step S15: If it is determined that the corresponding medium structure between the two plates in the variable dielectric type capacitive sensor is a series-parallel type structure, then when designing the capacitor, if conditions permit, the plate thickness is set to less than 0.5 mm to reduce the influence of the edge effect on the capacitance, calculate the actual value of the corrected capacitance, and perform calculation of the change in the series-parallel capacitance based on the actual value of the corrected capacitance to obtain the change in capacitance after the coal gangue medium changes corresponding to the series-parallel type plate structure.

3. The coal gangue dielectric identification method based on frequency shift characteristics according to claim 2, characterized in that The relative permittivity ε of the coal gangue medium described in step S13 jec For the plate area A, the distance d1 between the applied coal gangue media, the distance d0 between the media in the plate, and the relative permittivity ε of the media in the plate obj The calculation of the series capacitance change includes the following steps: According to the relative permittivity ε of the coal gangue medium jec , the plate area A, the distance d1 between the applied coal gangue media, and the permittivity ε of the medium inside the plate obj calculate the capacitance of the upper medium to obtain the capacitance of the upper coal gangue medium According to the plate area A, the distance d0 between the media inside the plate, and the dielectric constant ε of the media inside the plate obj Perform the calculation of the capacitance of the lower medium to obtain the capacitance of the lower medium inside the plate According to the capacitance operation principle corresponding to the series circuit, the capacitance of the upper medium capacitor C1 on the added coal gangue and the lower medium capacitor C2 in the plate are summarized and calculated to obtain the total capacitance value of the series structure Obtain the initial capacitance of the series structure corresponding to the case without adding coal gangue medium And based on the initial capacitance C of the series structure c0 Calculate the change in the total capacitance value C of the series structure c To obtain the capacitance change amount after the change of the coal gangue medium corresponding to the series plate structure 4. The method for identifying coal gangue dielectric based on frequency shift characteristics according to claim 2, wherein The relative dielectric constant ε of the coal gangue medium described in step S14 jec For the area A1 occupied by the coal gangue medium plate added, the area A2 occupied by the medium plate inside the plate, the distance d between the two plates, and the dielectric constant ε of the medium inside the plate obj The calculation of the change in the parallel capacitance includes the following steps: According to the relative dielectric constant ε of the coal gangue medium jec 、the area A1 occupied by the coal gangue medium plates added, the distance d between the two plates, and the dielectric constant ε of the medium inside the plates obj calculate the capacitance of the left medium to obtain the capacitance of the left coal gangue medium added According to the area A2 of the dielectric plate in the electrode plate, the distance d between the two electrode plates, and the dielectric constant ε of the dielectric in the electrode plate obj Perform the calculation of the right dielectric capacitance to obtain the right dielectric capacitance in the electrode plate According to the capacitance operation principle corresponding to the parallel circuit, the capacitance summation calculation is carried out on the left medium capacitance C′1 of the added coal gangue and the right medium capacitance C′2 inside the electrode plate to obtain the total capacitance value of the parallel structure Obtain the initial capacitance of the parallel structure corresponding to the case without adding coal gangue medium And based on the initial capacitance C of the parallel structure b0 Calculate the change in the total capacitance value C of the parallel structure b To obtain the capacitance change amount after the change of coal gangue medium corresponding to the parallel plate structure 5. The coal gangue dielectric identification method based on the frequency shift characteristic according to claim 2, wherein Step S15 includes the following steps: Step S151: If it is determined that the corresponding medium structure between the two plates in the variable dielectric type capacitive sensor is a series-parallel type structure, then obtain the medium surface area s between the two corresponding plates of the variable dielectric type capacitive sensor and the medium spacing D between the two plates; Step S152: Calculate the initial capacitance of the series-parallel type according to the surface area S of the dielectric between the two plates and the distance D between the two plates, so as to obtain the initial capacitance of the series-parallel type structure where ε f is the dielectric constant of the dielectric between the two plates, and k is the dielectric capacitance influence constant; Step S153: When designing this capacitor, if conditions permit, the plate thickness is set to less than 0.5 mm to reduce the influence of the edge effect on the capacitance. The edge effect influence coefficient ξ = 0.163 is calculated, and based on the edge effect influence coefficient ξ and the initial capacitance C of the series-parallel structure h0 to calculate the actual value of the corrected capacitance Step S154: Based on the initial capacitance C of the series-parallel structure h0 for the actual value C of the corrected capacitance h perform series-parallel capacitance change calculations to obtain the capacitance change amount after the change in the coal gangue medium corresponding to the series-parallel plate structure 6. The method for identifying coal gangue dielectric based on frequency shift characteristics according to claim 1, characterized in that Step S2 includes the following steps: Step S21: Connect a multi-frequency signal generator and a resistor between the corresponding two plates of the variable dielectric type capacitive sensor and perform abstraction processing on the excitation circuit to generate a dual-frequency excitation RC circuit for the coal gangue medium; Step S22: Take the changed capacitance caused by arbitrarily selecting different coal gangue media corresponding to different plate structures as the changed capacitance ΔC of the coal gangue medium dual-frequency excitation RC circuit, and calculate the capacitance frequency shift amount between the two plates of the variable dielectric capacitive sensor in the coal gangue medium dual-frequency excitation RC circuit based on the changed capacitance ΔC to obtain the capacitance frequency shift amount of the excitation RC circuit where f is the frequency of the AC voltage; Step S23: Apply a standard sinusoidal AC voltage to the dual-frequency excitation RC circuit of coal gangue medium where B is the amplitude of the applied AC voltage, t is time, is the initial phase angle, where In the series circuit corresponding to the dual-frequency excitation RC circuit of coal gangue medium, the effective value phasor of the input voltage is where j is the imaginary unit of the phasor, and the standard sinusoidal AC voltage U s in At this time, U′ = B, and the impedance of the resistor R in the dual-frequency excitation RC circuit of coal gangue medium is Z R = R, and the impedance of the capacitor C is Z C = -jX c ; Step S24: Calculate the effective value of the voltage based on the effective value phasor U′ corresponding to the input voltage, the impedance Z of the resistor R R and the impedance Z of the capacitor C C to obtain the effective value of the voltage across the resistor corresponding to the coal gangue sample 7. The method for identifying coal gangue dielectric based on frequency shift characteristics according to claim 1, characterized in that Step S3 includes the following steps: Step S31: Collect the response signal of the coal gangue medium corresponding to the variable dielectric type capacitive sensor in the coal gangue medium dual-frequency excitation RC circuit and the excitation frequency of the coal gangue medium corresponding to the coal gangue sample through a data recorder; Step S32: Use the variational mode decomposition (VMD) algorithm to perform VMD decomposition on the coal-gangue medium response signal to obtain the VMD decomposition signal spectrum of the coal-gangue medium; Step S33: Reconstruct and denoise the VMD decomposition signal spectrum of the coal-gangue medium using the Median signal to generate the denoised signal of the coal-gangue medium; Step S34: Obtain the dielectric loss value of the coal-gangue medium from the denoised signal of the coal-gangue medium; Step S35: Perform frequency loss correction on the excitation frequency of the coal-gangue medium corresponding to the coal-gangue sample based on the dielectric loss value of the coal-gangue medium to obtain the corrected excitation frequency of the dielectric loss corresponding to the coal-gangue sample.

8. The method for identifying coal gangue dielectric based on frequency shift characteristics according to claim 7, characterized in that Step S34 includes the following steps: Step S341: Obtain the corresponding change amplitude and change phase of the coal-gangue medium signal from the denoised signal of the coal-gangue medium; Step S342: Calculate the signal change rate of the denoised signal of the coal-gangue medium based on the change amplitude and change phase of the coal-gangue medium signal to obtain the signal change rate of the coal-gangue medium; Step S343: Perform statistical analysis on the peak value of the conductance frequency of the denoised signal of the coal-gangue medium to obtain the peak value of the conductance frequency of the coal-gangue medium signal; Step S344: Perform loss quantization calculation on the denoised signal of the coal-gangue medium using the dielectric loss calculation formula based on the signal change rate and the peak value of the conductance frequency of the coal-gangue medium signal to obtain the dielectric loss value of the coal-gangue medium.

9. The coal gangue dielectric identification method based on the frequency shift characteristic according to claim 8, wherein The specific dielectric loss calculation formula described in Step S344 is: where Δ ∈ f is the dielectric loss value of the coal gangue medium, f1 is the initial value of the dielectric loss frequency range, f2 is the end value of the dielectric loss frequency range, f(t) is the frequency of the coal gangue medium signal at time t of the noise reduction signal of the coal gangue medium, is the rate of change of the coal gangue medium signal, α is the weight factor affecting the rate of change of the signal, σ(f) is the conductivity of the coal gangue medium at frequency f, β is the weight factor affecting the conductivity, f m is the peak value of the conductance frequency of the coal gangue medium signal, γ is the attenuation factor of the conductance peak value, exp is the exponential function, f0 is the reference frequency of the coal gangue medium, Δf is the bandwidth of the signal frequency distribution, δ is the attenuation coefficient of the signal frequency distribution, and η is the correction coefficient of the dielectric loss value of the coal gangue medium.

10. The coal gangue dielectric identification method based on the frequency shift characteristic according to claim 1, wherein Step S4 includes the following steps: Step S41: Construct the characteristic vector X of the frequency shift characteristics of coal gangue medium, X = [U R1 , f R1 , U R2 , f R2 , where U R1 is the effective value of the voltage across the resistor corresponding to the coal sample, U R2 is the effective value of the voltage across the resistor corresponding to the gangue sample, f R1 is the excitation frequency of dielectric loss correction corresponding to the coal sample, f R2 is the excitation frequency of dielectric loss correction corresponding to the gangue sample; Step S42: Use the support vector machine, random forest, and convolutional neural network, and construct a coal-gangue classification and recognition model according to the characteristic vector X of the frequency shift characteristics of the coal-gangue medium to identify and analyze the change trend rate of the effective value of the voltage across the resistor corresponding to the coal-gangue sample at the corrected excitation frequency of the dielectric loss. If the effective value of the voltage across the resistor corresponding to the coal-gangue sample decreases as the corrected excitation frequency of the dielectric loss increases and the change trend rate is relatively severe, it is a gangue sample; otherwise, it is a coal sample, and the dielectric classification results corresponding to the coal and gangue samples are output.

Images