A health assessment method and fault monitoring and identification method for bolter digging

Abstract

The present invention provides a health assessment method and a fault monitoring and identification method for an anchor mining machine. The health assessment method includes calculating a similarity matrix of each operation cycle of the anchor mining machine in the same tunnel, building an abnormality detection matrix, building a health assessment function, and determining a health assessment judgment index, so as to achieve an accurate health assessment effect on the wearing parts of the anchor mining machine. The fault monitoring and identification method determines the coefficient X by building a fault monitoring and identification function for the maximum value of each detection parameter of the wearing parts of the anchor mining machine. n , where X n represents any coefficient among X1, X2, X3, X4 and X5; secondly, the coefficient X is determined by using the fault monitoring identification index. n , and then judge whether the operating status of the anchor boring machine is normal or not, reducing construction safety risks.

Description

Technical Field

[0001] The present invention relates to the technical field of anchor diggers, and in particular to a health assessment method and a fault monitoring and identification method for anchor diggers. Background Art

[0002] The anchor digger is a large-scale excavation equipment. To ensure the normal operation of the anchor digger, it is necessary to conduct a health assessment of the anchor digger, especially the health assessment of the wearing parts in the anchor digger. At the same time, it is also very important to monitor and identify the faults of the wearing parts in the anchor digger. However, the current health assessment of the anchor digger only checks whether the operating parameters of the anchor digger are within the rated range, and it is impossible to calculate the health value of the wearing parts in real time according to the operating characteristics of the anchor digger. Summary of the invention

[0003] The purpose of the present invention is to provide a health assessment method and fault monitoring and identification method for an anchor miner, so as to achieve accurate health assessment and fault monitoring of vulnerable parts in the anchor miner and reduce construction safety risks. The specific technical solution is as follows:

[0004] In a first aspect, the present invention provides a health assessment method for a bolter miner, comprising the following steps:

[0005] Step A1: Calculate the similarity matrix of each operation cycle of the bolter miner in the same tunnel

[0006] Each operation cycle of the anchor digger in the same tunnel is regarded as a relatively independent distribution, and VisualStudio is used to read the K operation cycles of the anchor digger; the detection parameters of the wearing parts of the anchor digger in each operation cycle are collected by the sensor component and transmitted to the control terminal of the anchor digger, and the detection parameters include P detection quantities, which are the motor temperature, reducer vibration, boom measurement point deformation, sliding frame measurement point stress, cylinder pressure and cutting motor current in the wearing parts of the anchor digger; each detection quantity has a measurement value in the time series of each operation cycle; the total deviation of any operation cycle m in the K operation cycles and the latest operation cycle L of the anchor digger on the corresponding P detection quantities is the similarity matrix S(L, m) t , which is expressed by formula (1):

[0007]

[0008] In formula (1), m is the number of any one of the K operating cycles; L is the number of the latest operating cycle of the bolter; t = 1, 2, 3 ... a, represents any measurement value in the time series of each operating cycle; n = 1, 2, 3 ... P, represents any one of the P detection quantities of each operating cycle; x tnLrepresents the tth measured value of the bth detection quantity in the time series of the latest operation cycle L of the bolter; x tnm represents the tth measured value of the nth detection quantity in the time series of operation cycle m; δ n is the variance of the nth detection quantity within K operation cycles;

[0009] Step A2: Build anomaly detection matrix

[0010] Find the operation cycle F that is most similar to the latest operation cycle L among the K operation cycles, and transform the similarity matrix of the operation cycle F into the transformation matrix T tLF The mean of its time series is defined as the anomaly detection matrix T F , as shown in formula (2):

[0011]

[0012] Step A3: Build health assessment function and determine health assessment judgment indicators

[0013] The normal node proportion c is calculated by the anomaly detection matrix, which is the health degree, and is expressed by formula (3):

[0014]

[0015] In formula (3), J represents the similarity matrix S(L, m) t The number of eigenvectors; c≤1, when c=1, it indicates complete health; when 0.9≤c<1, it indicates health; when 0.7≤c<0.9, it indicates sub-health; when c<0.7, it indicates unhealthy.

[0016] Optionally, in step A2, the specific process of finding the operation cycle F that is most similar to the latest operation cycle L is as follows:

[0017] First, at the tth measurement value in the time series, the transformation matrix T is constructed according to the similarity matrix of the operation cycle m. tLm , which is expressed by formula (4):

[0018]

[0019] In formula (4), i = 1, 2, 3, ..., J; v tim Represents the similarity matrix S(L, m) t The eigenvector of Indicates v tim The transpose of tim Denotes the feature vector v tim The weight of is as shown in formula (5);

[0020]

[0021] Secondly, for the K transformation matrices T corresponding to the similarity matrices of the K operation cycles tLm In the above formula, find the transformation matrix that satisfies the conditional expression (6), which is the transformation matrix TtLF of the similarity matrix of the operation cycle F;

[0022]

[0023] In formula (6), <T tLm , S(L,m) t > F Denotes the transformation matrix T tLm And the similarity matrix S(L,m) t The inner product of <T tLm , T tLm > F Denotes the transformation matrix T tLm and the transformation matrix T tLm The inner product of <S(L,m) t , S(L,m) t >F represents the similarity matrix S(L,m) t And the similarity matrix S(L,m) t The inner product of .

[0024] Optionally, in step A1, the sensing component includes a temperature sensor for detecting the temperature of the motor, a vibration sensor for detecting the vibration of the reducer, a first strain gauge sensor for detecting the deformation of the boom measuring point, a second strain gauge sensor for detecting the stress of the sliding frame measuring point, a third strain gauge sensor for detecting the cylinder pressure and a current sensor for detecting the current of the cutting motor; the temperature sensor and the current sensor are respectively connected to the control terminal of the anchor digger through the circuit; the vibration sensor, the first strain gauge sensor, the second strain gauge sensor and the third strain gauge sensor are respectively connected to the control terminal of the anchor digger through the monitor.

[0025] Optionally, the K operating cycles are all normal operating cycles.

[0026] In a second aspect, the present invention provides a fault monitoring and identification method for an anchor miner, comprising the following steps:

[0027] Step B1, each operation cycle of the anchor digger in the same tunnel is regarded as a relatively independent distribution, and Visual Studio is used to read K operation cycles of the anchor digger; the detection parameters of the wearing parts of the anchor digger in each operation cycle are collected by the sensor component and transmitted to the control terminal of the anchor digger, and the detection parameters include P detection quantities, which are the motor temperature, reducer vibration, boom measurement point deformation, sliding frame measurement point stress and cutting motor current in the wearing parts of the anchor digger;

[0028] A fault monitoring and identification function is constructed according to the maximum value of the detection parameters, wherein the maximum value of each detection parameter includes the maximum value of the motor temperature T max , Maximum value of reducer vibration d max , the maximum deformation of the arm measuring point W max , Maximum stress value σ at the sliding frame measuring point max and the maximum current of the cutting motor max The fault monitoring identification function is shown in formula (7);

[0029]

[0030] In formula (7), X1 is the coefficient of the maximum value of the motor temperature; X2 is the coefficient of the maximum value of the reducer vibration; X3 is the coefficient of the maximum value of the arm measurement point deformation; X4 is the coefficient of the maximum value of the sliding frame measurement point stress; X5 is the coefficient of the maximum value of the cutting motor current;

[0031] Step B2: Determine the fault monitoring and identification judgment index of the anchor digger

[0032] A fault monitoring indicator light is arranged on the anchor mining machine, and the fault monitoring indicator light is connected to the control terminal of the anchor mining machine through a circuit. n <0.8, the fault monitoring indicator light turns green, indicating that the bolter is operating normally; when 0.8<X n <0.95, the fault monitoring indicator light turns yellow, indicating an operating status warning for the bolter; n ≥0.95, the fault monitoring indicator light turns red, indicating that the bolter digging machine is in high-risk operation; n Represents any coefficient of X1, X2, X3, X4, and X5.

[0033] The application of the technical solution of the present invention has the following beneficial effects:

[0034] (1) The health assessment method for the anchor mining machine described in the present invention collects the detection parameters of the wearing parts in the anchor mining machine in each operation cycle through the sensor component to construct a similarity matrix, uses the similarity matrix to find the operation cycle F that is most similar to the latest operation cycle L, thereby building an abnormal detection matrix, and constructs a health assessment function through the abnormal detection matrix, and finally uses the health assessment index to achieve an accurate health assessment effect of the wearing parts in the anchor mining machine. More specifically, the health value obtained by the health assessment method of the present invention can show the health status of the wearing parts in the anchor mining machine. The health status is quantified by numerical indicators, and the results are clear and reliable. Compared with the conventional health judgment or simple early warning health assessment, the present invention more intuitively shows the health of the anchor mining machine.

[0035] (2) The fault monitoring and identification method for the anchor mining machine described in the present invention first constructs a fault monitoring and identification function by taking the maximum values ​​of the detection parameters of the wearing parts of the anchor mining machine, thereby determining the coefficient X n , where X n represents any coefficient among X1, X2, X3, X4 and X5; secondly, the coefficient X is determined by using the fault monitoring identification index. n , and then judge whether the operating status of the anchor drill is normal or not, reducing construction safety risks.

[0036] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail below with reference to specific embodiments. DETAILED DESCRIPTION

[0037] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0038] Embodiment 1:

[0039] A health assessment method for a bolter miner comprises the following steps:

[0040] Step A1: Calculate the similarity matrix of each operation cycle of the bolter miner in the same tunnel

[0041] Each operation cycle of the anchor digger in the same tunnel is regarded as a relatively independent distribution, and VisualStudio is used to read the K operation cycles of the anchor digger; the detection parameters of the wearing parts of the anchor digger in each operation cycle are collected by the sensor component and transmitted to the control terminal of the anchor digger, and the detection parameters include P (P is specifically 6) detection quantities, which are the motor temperature, reducer vibration, boom measurement point deformation, sliding frame measurement point stress, cylinder pressure and cutting motor current in the wearing parts of the anchor digger; each detection quantity has a measurement value in the time series of each operation cycle; the total deviation of any operation cycle m in the K operation cycles and the latest operation cycle L of the anchor digger on the corresponding P detection quantities is the similarity matrix S(L, m) t , which is expressed by formula (1):

[0042]

[0043] In formula (1), m is the number of any one of the K operating cycles; L is the number of the latest operating cycle of the bolter; t = 1, 2, 3...a (a can be 3600 or 60), representing any measurement value in the time series of each operating cycle; n = 1, 2, 3...households, representing any one of the P detection quantities of each operating cycle; x tnL represents the tth measured value of the nth detection quantity in the time series of the latest operation cycle L of the bolter; x tnm represents the tth measured value of the nth detection quantity in the time series of operation cycle m; δ n is the variance of the nth detection quantity within K operation cycles;

[0044] Step A2: Build anomaly detection matrix

[0045] Find the operation cycle F that is most similar to the latest operation cycle L among the K operation cycles, and transform the similarity matrix of the operation cycle F into the transformation matrix T tLF The mean of its time series is defined as the anomaly detection matrix T F , as shown in formula (2):

[0046]

[0047] Step A3: Build health assessment function and determine health assessment judgment indicators

[0048] The normal node proportion c is calculated by the anomaly detection matrix, which is the health degree, and is expressed by formula (3):

[0049]

[0050] In formula (3), J represents the similarity matrix S(L, m)t The number of eigenvectors; c≤1, when c=1, it indicates complete health; when 0.9≤c<1, it indicates health; when 0.7≤c<0.9, it indicates sub-health; when c<0.7, it indicates unhealthy.

[0051] In step A2, the specific process of finding the operation cycle F that is most similar to the latest operation cycle L is as follows:

[0052] First, at the tth measurement value in the time series, the transformation matrix T is constructed according to the similarity matrix of the operation cycle m. tLm , which is expressed by formula (4):

[0053]

[0054] In formula (4), i = 1, 2, 3, ..., J; v tim Represents the similarity matrix S(L, m) t The eigenvector of Indicates v tim The transpose of tim Denotes the feature vector v tim The weight of is as shown in formula (5);

[0055]

[0056] Secondly, for the K transformation matrices T corresponding to the similarity matrices of the K operation cycles tLm In the above equation, we find the transformation matrix that satisfies the conditional expression (6), which is the transformation matrix T of the similarity matrix of the operation cycle F. tLF ;

[0057]

[0058] In formula (6), <T tLm , S(L,m) t > F Denotes the transformation matrix T tLm And the similarity matrix S(L,m) t The inner product of <T tLm , T tLm > F Denotes the transformation matrix T tLm and the transformation matrix T tLm The inner product of <S(L,m) t , S(L,m) t > F Represents the similarity matrix S(L, m) t And the similarity matrix S(L,m) t The inner product of .

[0059] In step A1, the sensing component includes a temperature sensor for detecting the temperature of the motor, a vibration sensor for detecting the vibration of the reducer, a first strain gauge sensor for detecting the deformation of the boom measuring point, a second strain gauge sensor for detecting the stress of the sliding frame measuring point, a third strain gauge sensor for detecting the cylinder pressure and a current sensor for detecting the current of the cutting motor; the temperature sensor and the current sensor are respectively connected to the control terminal of the anchor digger through circuits; the vibration sensor, the first strain gauge sensor, the second strain gauge sensor and the third strain gauge sensor are respectively connected to the control terminal of the anchor digger through monitors.

[0060] The K operating cycles are all normal operating cycles.

[0061] When the health assessment method for the anchor digger is applied, the detection parameters of the wearing parts in the anchor digger in each operation cycle are collected by the sensor component to construct a similarity matrix, and the similarity matrix is ​​used to find the operation cycle F that is most similar to the latest operation cycle L, so as to build an anomaly detection matrix, and a health assessment function is built through the anomaly detection matrix. Finally, the health assessment index is used to achieve an accurate health assessment effect of the wearing parts in the anchor digger.

[0062] A fault monitoring and identification method for an anchor mining machine comprises the following steps:

[0063] Step B1, each operation cycle of the anchor digger in the same tunnel is regarded as a relatively independent distribution, and Visual Studio is used to read K operation cycles of the anchor digger; the detection parameters of the wearing parts of the anchor digger in each operation cycle are collected by the sensor component and transmitted to the control terminal of the anchor digger, and the detection parameters include P detection quantities, which are the motor temperature, reducer vibration, boom measurement point deformation, sliding frame measurement point stress and cutting motor current in the wearing parts of the anchor digger;

[0064] A fault monitoring and identification function is constructed based on the maximum values ​​of the detection parameters, wherein the maximum values ​​of the detection parameters (the specific values ​​of the maximum values ​​are related to the model of the wearing parts) include the maximum value of the motor temperature T max , Maximum value of reducer vibration d max , the maximum deformation of the arm measuring point W max , Maximum stress value σ at the sliding frame measuring point max and the maximum current of the cutting motor I max The fault monitoring identification function is shown in formula (7);

[0065]

[0066] In formula (7), X1 is the coefficient of the maximum value of the motor temperature; X2 is the coefficient of the maximum value of the reducer vibration; X3 is the coefficient of the maximum value of the arm measurement point deformation; X4 is the coefficient of the maximum value of the sliding frame measurement point stress; X5 is the coefficient of the maximum value of the cutting motor current;

[0067] Step B2: Determine the fault monitoring and identification judgment index of the anchor digger

[0068] A fault monitoring indicator light is arranged on the anchor mining machine, and the fault monitoring indicator light is connected to the control terminal of the anchor mining machine through a circuit. n <0.8, the fault monitoring indicator light turns green, indicating that the bolter is operating normally; when 0.8<X n <0.95, the fault monitoring indicator light turns yellow, indicating an operating status warning for the bolter; n ≥0.95, the fault monitoring indicator light turns red, indicating that the bolter digging machine is in high-risk operation; n Represents any coefficient of X1, X2, X3, X4, and X5.

[0069] When the fault monitoring and identification method for the anchor mining machine is applied, firstly, a fault monitoring and identification function is constructed by taking the maximum values ​​of the detection parameters of the wearing parts in the anchor mining machine, thereby determining the coefficient X n , where X n represents any coefficient among X1, X2, X3, X4 and X5; secondly, the coefficient X is determined by using the fault monitoring identification index. n , and then judge whether the operating status of the anchor drill is normal or not, reducing construction safety risks.

[0070] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A health assessment method for bolter miners, characterized in that: The following steps are involved: Step A1: Calculate the similarity matrix of each operation cycle of the bolter miner in the same tunnel Each operation cycle of the anchor digger in the same tunnel is regarded as a relatively independent distribution, and VisualStudio is used to read the K operation cycles of the anchor digger; the detection parameters of the wearing parts of the anchor digger in each operation cycle are collected by the sensor component and transmitted to the control terminal of the anchor digger, and the detection parameters include P detection quantities, which are the motor temperature, reducer vibration, boom measurement point deformation, sliding frame measurement point stress, cylinder pressure and cutting motor current in the wearing parts of the anchor digger; each detection quantity has a measurement value in the time series of each operation cycle; the total deviation of any operation cycle m in the K operation cycles and the latest operation cycle L of the anchor digger on the corresponding P detection quantities is the similarity matrix S(L, m) t , which is expressed by formula (1): In formula (1), m is the number of any one of the K operating cycles; L is the number of the latest operating cycle of the bolter; t = 1, 2, 3...a, represents any measurement value in the time series of each operating cycle; n = 1, 2, 3...P, represents any one of the P detection quantities of each operating cycle; x tnL represents the tth measured value of the nth detection quantity in the time series of the latest operation cycle L of the bolter; x tnm represents the tth measured value of the nth detection quantity in the time series of operation cycle m; δ n is the variance of the nth detection quantity within K operation cycles; Step A2: Build anomaly detection matrix Find the operation cycle F that is most similar to the latest operation cycle L among the K operation cycles, and transform the similarity matrix of the operation cycle F into the transformation matrix T tLF The mean of its time series is defined as the anomaly detection matrix T F , as shown in formula (2): Step A3: Build health assessment function and determine health assessment judgment indicators The normal node proportion c is calculated by the anomaly detection matrix, which is the health degree, and is expressed by formula (3): In formula (3), J represents the similarity matrix S(L, m) t The number of eigenvectors; c≤1, when c=1, it indicates complete health; when 0.9≤c<1, it indicates health; when 0.7≤c<0.9, it indicates sub-health; when c<0.7, it indicates unhealthy; In step A2, the specific process of finding the operation cycle F that is most similar to the latest operation cycle L is as follows: First, at the tth measurement value in the time series, the transformation matrix T is constructed according to the similarity matrix of the operation cycle m. tLm , which is expressed by formula (4): In formula (4), i = 1, 2, 3, ... J; v tim Represents the similarity matrix S(L, m) t The eigenvector of Indicates v tim The transpose of tim Denotes the feature vector v tim The weight of is as shown in formula (5); Secondly, for the K transformation matrices T corresponding to the similarity matrices of the K operation cycles tLm In the above equation, we find the transformation matrix that satisfies the conditional expression (6), which is the transformation matrix T of the similarity matrix of the operation cycle F. tLF ; In formula (6), <T tLm , S(L,m) t > F Denotes the transformation matrix T tLm And the similarity matrix S(L,m) t The inner product of <T tLm , T tLm > F Denotes the transformation matrix T tLm and the transformation matrix T tLm The inner product of <S(L,m) t , S(L,m) t > F Represents the similarity matrix S(L, m) t And the similarity matrix S(L,m) t The inner product of .

2. The health assessment method for bolter miner according to claim 1, characterized in that: In step A1, the sensing component includes a temperature sensor for detecting the temperature of the motor, a vibration sensor for detecting the vibration of the reducer, a first strain gauge sensor for detecting the deformation of the boom measuring point, a second strain gauge sensor for detecting the stress of the sliding frame measuring point, a third strain gauge sensor for detecting the cylinder pressure and a current sensor for detecting the current of the cutting motor; the temperature sensor and the current sensor are respectively connected to the control terminal of the anchor digger through circuits; the vibration sensor, the first strain gauge sensor, the second strain gauge sensor and the third strain gauge sensor are respectively connected to the control terminal of the anchor digger through monitors.

3. The health assessment method for bolter miner according to claim 1, characterized in that: The K operating cycles are all normal operating cycles.

Images