CN104373153A - Coal and rock property identification method and system for underground coal mine full-mechanized caving face - Google Patents

Abstract

The invention relates to a coal rock property identification method that can be used in fully mechanized caving working faces in coal mines. By analyzing the vibration signal and sound pressure signal of the fully mechanized caving hydraulic support in time domain and wavelet envelope frequency band energy analysis, and extracting the grayscale and texture features of the image signal collected from the fully mechanized caving hydraulic support coal outlet, the The characteristic parameters of the unknown coal and rock properties under real working conditions are matched with the characteristic database of known coal and rocks through the feature matching identification module to identify the properties of unknown coal and rocks, and then the opening of the coal discharge port of the hydraulic support is controlled by the valve group or closed to control, so as to realize the automatic identification of coal and rock properties in the fully mechanized caving face, which provides a solution and technical approach for the automation and intelligence of the top coal caving mining process in the fully mechanized caving face, and at the same time provides a solution for the fully mechanized caving face Mining with fewer people or even unmanned mining has laid a theoretical and technical foundation.

Description

A method and system for identifying coal and rock properties in a fully mechanized caving face in a coal mine

technical field

The invention relates to a method and a system that can be used for identifying the properties of coal and rocks in coal mines, in particular, a method and a system that can be used for identifying the properties of coal and rocks in fully mechanized caving working faces. the

Background technique

Coal is the main source of primary energy in my country. The comprehensive mechanized top-coal caving mining method is one of the effective methods to realize high-yield and high-efficiency mining of thick coal seams (general coal seam thickness > 3.5m), and has been widely promoted at home and abroad. In a thick coal seam, a longwall working face with a mining height of 2-3 meters is arranged along the bottom of the coal seam (or section), and mining is carried out by conventional methods. After the coal is broken into bulk, it is discharged from the back (or above) of the support, and is transported out of the working face by the scraper conveyor. the

Although this method has doubled the output of the fully mechanized caving face, the degree of top coal caving, that is, the identification of coal rock properties, still depends entirely on the visual and auditory judgment of the coal caving workers on site, and then controls the caving of hydraulic supports. Closing timing of the coal mouth. However, the working environment of the fully mechanized caving face is harsh, with a large amount of dust, low visibility, and narrow working space. It is difficult to accurately judge the level of top-coal caving only by manual monitoring. Moreover, the randomness of the work quality is large, and it is difficult to achieve the ideal level. Control effect. Therefore, it inevitably leads to the "under-discharge" and "over-discharge" conditions in the caving process. "Under-discharge" will cause a decrease in coal recovery rate; The more serious is the occurrence of vicious accidents in fully mechanized caving face. Therefore, the development of unattended, to the greatest extent the realization of the fully mechanized caving face with fewer people, or even unmanned, will help to improve the safety of the production process, and it is also a direction for the development of coal mining machinery. Therefore, how to control the opening and closing time of the coal opening according to the degree of top coal caving is a major problem in the fully mechanized caving mining process. the

Therefore, people hope that some methods can be used to realize the automatic identification of coal and rock in the caving process so as to control coal caving.

The opening or closing of the mouth. In this way, the coal recovery rate can be effectively improved, the coal gangue content rate can be reduced, the coal quality can be improved, and the efficient production of top coal caving can be realized. The development of related equipment can be carried out on the basis of the identification theory of top coal caving coal rock properties, which can realize comprehensive Put fewer or no people on the working face, avoid unnecessary casualties, and create huge social and economic benefits. the

Contents of the invention

(1) Technical issues

The technical problem to be solved by the present invention is to provide an automatic identification method of coal and rock properties in the top coal caving process of fully mechanized caving face with high accuracy, safety, reliability and good effect in view of the deficiencies in existing coal rock identification methods ; The second is to realize the automatic control of the top coal caving of the hydraulic support of the fully mechanized caving face according to the method. the

(2) Technical plan

For this reason, the present invention provides a method and system for identifying coal and rock properties of fully mechanized caving working faces in coal mines based on vibration signals, sound pressure signals, and image signals. Including: vibration sensor, sound pressure sensor, image sensor, signal acquisition device, high-speed processor, controller, and control valve group, among which the vibration sensor and sound pressure sensor are installed at the rear tail beam of the hydraulic support of the fully mechanized caving face . The sensor is installed at the coal discharge port of the hydraulic support of the fully mechanized caving face.

The vibration sensor and the sound pressure sensor respectively pick up the vibration signal and sound pressure signal when the coal or roof rock collapses and impacts the tail beam of the hydraulic support during the roof coal drop process, converts it into an electrical signal, and sends it into the signal acquisition device; the signal acquisition device converts the input signal into a digital signal and sends it to the high-speed processor; the high-speed processor extracts its characteristic data after analysis and processing, and the characteristic matching and identification module compares the extracted characteristic data of unknown coal rocks with known coal rocks The characteristic database is matched to complete the identification of coal and rock properties, and the identification result is sent to the controller; the controller performs calculation and processing according to the received information, and then sends out a control command, and the control valve group controls the opening of the coal discharge port of the hydraulic support or off. the

The image sensor collects image signals of coal and rocks at the coal outlet during the top coal lowering process of the hydraulic support, and transmits the collected signals to a high-speed processor for coal and rock signal recognition processing and feature extraction. The feature matching identification module matches the extracted feature data of unknown coal rocks with the feature database of known coal rocks, completes the identification of coal rock properties, and sends the identification results to the controller; the controller performs calculation processing on the received information Finally, a control command is issued, and the control valve group controls the opening or closing of the coal discharge port of the hydraulic support.

(3) Beneficial effects

The technical scheme provided by the invention has the following advantages:

(1) There are many test parameters selected by the present invention, including vibration signal, sound pressure signal, and image signal, and the selected parameters are sensitive to coal and rock properties, so that errors caused by field equipment failures can be avoided, and when the high-speed processor analyzes information, it is ensured. accuracy;

(2) The equipment is easy to deploy , has strong adaptability, high recognition rate, and can perform real-time automatic recognition;

(3) The system responds quickly, and the test results are accurate and reliable;

(4) The method and system have wide applications and can be used on any working surface where the system can be installed. the

Description of drawings

Fig. 1 is a system structure diagram of the present invention

Fig. 2 is the system flowchart of the present invention

Figure 3 is a flow chart for extracting time-domain features of vibration and sound pressure signals

Figure 4 is a flow chart of the extraction of wavelet packet frequency band energy values of vibration and sound pressure signals

Figure 5 is a flow chart of image grayscale extraction

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are for illustration, but are not intended to limit the scope of the invention.

Fig. 1 is a structural diagram of the system of the present invention, including a vibration sensor, a sound pressure sensor, an image sensor, a signal acquisition device, a high-speed processor, a controller, and an electro-hydraulic control valve, wherein the vibration sensor and the sound pressure sensor are installed on the fully mechanized caving face At the rear tail beam of the hydraulic support, the image sensor is installed at the coal discharge port of the hydraulic support of the fully mechanized caving face.

The vibration sensor and the sound pressure sensor respectively pick up the vibration signal and the sound pressure signal generated when the coal or roof rock collapses and impacts the hydraulic support during the roof coal drop process, convert it into an electrical signal, and send it to the signal acquisition device; The signal acquisition device converts the input signal into a digital signal and sends it to the high-speed processor; the high-speed processor extracts its characteristic data after analysis and processing, and the feature matching and identification module compares the extracted characteristic data of unknown coal rocks with the characteristics of known coal rocks The database is matched to complete the identification of coal and rock properties, and the identification results are sent to the controller; after the controller performs calculations and processing according to the received information, it sends out control instructions, and the control valve group controls the opening or closing of the coal discharge port of the hydraulic support . the

The image sensor collects image signals of coal and rocks at the coal outlet during the top coal lowering process of the hydraulic support, and transmits the collected signals to a high-speed processor for coal and rock signal recognition processing and feature extraction. The matching and identification module matches the extracted feature data of unknown coal rocks with the feature database of known coal rocks, completes the identification of coal rock properties, and sends the identification results to the controller; the controller processes the received information , to issue a control command, and the control valve group controls the opening or closing of the coal discharge port of the hydraulic support.

Fig. 3 is a flow chart of the system of the present invention, and the specific steps include:

1. Under the same conditions (sensor type, installation method, working face environment, illumination, etc.), respectively collect the vibration, sound pressure, and image signals of known coal samples and known rock samples, and intercept a certain number of data points to carry out Domain feature analysis, wavelet packet frequency band energy analysis and gray level co-occurrence matrix extraction of image signals;

2. Calculate the time domain feature index and frequency band energy index based on the time domain feature analysis of the vibration signal and the sound pressure signal and the wavelet packet frequency band energy analysis, as well as the image feature value based on the image gray level co-occurrence matrix, in order to further improve the recognition accuracy , it is possible to collect multiple sets of data for known coal samples and known rock samples respectively, use the same processing method to extract multiple eigenvalues and calculate the average value, and finally use the eigenvalues as The fixed parameters of coal rock identification are retained;

3. Under the same conditions, the vibration, sound pressure, and image signals of unknown coal and rock samples during the caving process of the hydraulic support of the fully mechanized working face under real working conditions were collected, and the data were analyzed and processed by the same method as before to obtain characteristic indicators of the unknown coal rock sample;

4. Carry out numerical matching according to the relationship between eigenvalues to determine whether the unknown sample is coal or rock, and then achieve the purpose of coal and rock identification during the top coal caving process of fully mechanized caving face. the

Fig. 3 is a flow chart of time-domain feature extraction of vibration signal and sound pressure signal, and the specific steps include:

1. Time-domain feature analysis, preprocessing the selected vibration and sound pressure signals to remove trend items and gross errors, and then analyzing the time-domain features;

2. By analyzing the time-domain characteristics of the vibration and sound pressure signals, the time-domain statistical indicators and statistical values of the tail beam vibration and sound pressure signals of the fully mechanized caving hydraulic support under different working conditions are obtained;

3. In order to show more clearly the differences in the statistical values of the time-domain statistical indicators of the vibration and sound pressure signals of the tail beam of the fully mechanized caving hydraulic support under different working conditions, the The rate of change of the statistical value of the beam vibration and sound pressure signals in the time domain, and it is explained by the following formula

Among them, α is the statistical value of the time-domain feature index.

4. Find out the characteristic parameters sensitive to the working conditions by comparing the change rate of the statistical value of the tail beam vibration and sound pressure signal in the time domain of the fully mechanized caving hydraulic support under different working conditions;

5. Determine the parameters with a higher recognition rate among the characteristic parameters sensitive to working conditions through multiple data collection and analysis methods, and determine them as coal and rock identification parameters. the

Fig. 4 is the extraction flowchart of image grayscale, and specific steps include:

1. Convert the collected color images of coal rocks into grayscale images through relevant algorithms in MATLAB software, so as to achieve the purpose of maintaining the original characteristics of coal rock images and reducing the amount of calculation;

2. Process the signal through the medfilt algorithm in the MATLAB software, arrange all the pixel values in the window in order from small to large, take the value in the middle of these pixels to replace the pixel value in the center of the window, and then realize each pixel in turn, so that This is to weaken the noise generated in the process of image signal acquisition, transmission, conversion, etc., to ensure the quality of the image signal;

3. Extract the gray histogram of the signal through the imhist function in MATLAB software, and it can be observed from the gray distribution of the coal and rock images that there is a significant difference;

4. Extract the gray mean value, variance and other characteristic parameters of the image signal through the mean and other functions in the MATLAB software. Through observation and comparison, it is found that the gray mean value can be used as a coal rock identification index;

5. Compare the gray histogram of the image under different working conditions with the average gray value, find the difference and then distinguish coal or rock.

Fig. 5 is the flow chart of extracting the wavelet packet frequency band energy value of sound pressure and vibration signal, and the specific steps include:

1. Carry out two-layer wavelet packet decomposition on the sound pressure and vibration signals (the determination of the number of decomposition layers should be analyzed in detail according to the specific situation, here we take two layers as an example), and extract the four frequency components of the third layer from low frequency to high frequency respectively The signal features and its decomposition structure are shown in Table 1 .

Table 1 The wavelet packet decomposition level table of sound pressure and vibration signals

As shown in Table 1 , (i,j) represents the jth node of the i-th layer, where i=0,1,2; j=O,l,2,3, each node represents Certain signal characteristics. Among them, the (0,0) node represents the original signal S, and the (1,0) node represents the low-frequency coefficient X10 of the first layer of wavelet packet decomposition. , (1,1) node represents the high-frequency coefficient X11 of the first layer of wavelet packet decomposition, (2,0) node represents the coefficient of the Oth node of the third layer, and so on.

2. Reconstruct the wavelet packet decomposition coefficients, use S10 to represent the reconstructed signal of X 10, and so on, then all nodes in the second layer perform coefficient reconstruction to obtain a new signal function as

S ₂ =S ₂₀ +S ₂₁ +S ₂₂ +S ₂₃ (4.13)

3. Find the sum of the energy of each frequency band signal and the total energy of the signal. the

${\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; S</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\mathrm{<span\; class="notranslate">\; dt</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; jk</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4.14</span>}<span\; class="notranslate">\; )</span>$

Among them, X _jk represents the discrete point coefficient of the reconstructed signal S _2j .

total energy $\mathrm{E.}={\left(\underset{j=0}{\overset{3}{\mathrm{\&Sigma;}}}{\left|{\mathrm{E.}}_{2j}\right|}^2\right)}^{1/2}$

4. Construct the eigenvector of the energy signal and normalize the eigenvector

T=[E ₂₀ ,E ₂₁ ,E ₂₂ ,E ₂₃ ] (4.15)

The energy normalized eigenvector is

T'＝[E ₂₀ /E, E ₂₁ /E, E ₂₂ /E, E ₂₃ /E] (4.16)

In order to further improve the recognition accuracy, the same method can be used to collect multiple sets of data, process and analyze them, and calculate their average value to obtain the wavelet packet decomposition and reconstruction of the vibration signal and sound pressure signal of the rear tail beam of the hydraulic support of the fully mechanized caving face. The total energy value and energy normalized eigenvector are reserved as the characteristic parameters of coal rock identification;

5. Compare the total energy value of the coal and rock energy and the normalized eigenvector of the tail beam vibration and sound pressure signal of the hydraulic support under different working conditions, and find the difference to distinguish coal or rock. the

Claims (4)

1. the coal petrography proterties recognition methods of underground coal mine fully mechanized coal face and a system, putting coal equipment for controlling, realizing automation and the intellectuality of Sub-Level Caving;

It is characterized in that, comprise: vibrating sensor, sound pressure sensor, imageing sensor, signal pickup assembly, high speed processor, controller, control valve group, wherein vibrating sensor, sound pressure sensor are arranged on tail boom place after hydraulic support of top-coal saving face, and imageing sensor is arranged on hydraulic support of top-coal saving face coal discharge outlet place;

Described vibrating sensor, sound pressure sensor pick up vibration signal, sound pressure signal when to impact tail boom after hydraulic support when coal or balkstone are caving in the coal let-down process of top respectively, are converted into the signal of telecommunication, send into signal pickup assembly; The signal of input is converted to data signal and sends into high speed processor by signal pickup assembly; High speed processor extracts its characteristic after carrying out treatment and analysis, mates with the property data base of known coal petrography, completes the identification of coal petrography proterties, and recognition result is sent into controller; Controller sends control instruction after carrying out calculation process according to the information received, by opening or closing of control valve group hydraulic control support coal discharge outlet;

Described imageing sensor gathers the picture signal of coal and rock in the coal let-down process of hydraulic support top, and the signal collected is sent to high speed processor carries out the process of coal petrography Signal analysis and feature extraction, mate with the property data base of known coal petrography, complete the identification of coal petrography proterties, and recognition result is sent into controller; Controller sends control instruction, by opening or closing of control valve group hydraulic control support coal discharge outlet after the information received is carried out calculation process.

2. the coal petrography proterties recognition methods of underground coal mine fully mechanized coal face according to claim 1 and system, it is characterized in that, described high speed processor comprises signal analysis and processing module and characteristic matching identification module;

Wherein said signal analysis and processing module is used for carrying out time domain and Wavelet Packet Frequency Band Energy feature to the vibration signal of the unknown coal petrography proterties of collection in worksite, sound pressure signal and the picture signal of the unknown coal petrography proterties of collection in worksite being carried out to the extraction of gray scale and textural characteristics;

The property data base of the vibration signal that described characteristic matching identification module is used for signal analysis and processing module to obtain, sound pressure signal, picture signal characteristics index and known coal sample and known rock sample mates, and show that the unknown sample of collection in worksite is coal or rock.

3. the coal petrography proterties recognition methods of underground coal mine fully mechanized coal face according to claim 1 and system, it is characterized in that, described controller receives and deposits the order and parameter that high speed processor sends, and carries out decoding to received order and send control command.

4. the coal petrography proterties recognition methods of underground coal mine fully mechanized coal face according to claim 1 and system, is characterized in that, the instruction that described control valve group provides according to controller, and belonging to controlling, hydraulic support coal discharge outlet opens or closes.