Rocker arm inclination angle detection method based on MEMS accelerometer

Abstract

The application relates to a MEMS accelerometer-based rocker arm inclination angle detection method, three three-axis MEMS accelerometers a, b and c are respectively installed in an electric control box and left and right rocker arms of a coal mining machine, the y detection axes of the accelerometers b and c are respectively parallel to swing axes of the left and right rocker arms, the z detection axes of the accelerometers b and c are vertically upward, the x detection axes of the accelerometers b and c are horizontally extended and far away from the swing axes of the respective rocker arms when the electric control box and the left and right rocker arms are in a horizontal state, the corresponding detection axes of the accelerometers a and c are parallel, then original acceleration vectors output by the MEMS accelerometers a, b and c are respectively acquired, the original acceleration vectors are subjected to filtering processing to obtain filtered acceleration vectors, inclination angle values of the filtered acceleration vectors relative to a horizontal plane are respectively calculated, and the included angles of the left and right rocker arms relative to the electric control box are respectively calculated. Compared with a traditional method, the application has the advantages of small sensor volume, simple installation and low cost.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of coal winning machine rocker arm inclination detection method, the core sensor used is MEMS accelerometer. BACKGROUND

[0002] In order to control the safe mining height of coal winning machine in underground mining condition, the inclination detection of coal winning machine rocker arm is needed. At present, there is almost no inclination sensor that can be directly used for inclination detection of coal winning machine rocker arm. The traditional rocker arm inclination detection method includes cylinder stroke displacement method and coded potentiometer rotation distance measurement method. The cylinder stroke displacement method uses displacement sensor, which is mainly installed in the internal of height-adjusting cylinder. It needs to establish corresponding mathematical model according to different models of coal winning machine mechanical structure, and the relationship curve between displacement distance and drum height is calculated by the displacement distance of connecting piece. However, the calculation process of obtaining drum height by this method is repeated and complicated, and the efficiency is low. Since the sensor is installed in the internal of height-adjusting cylinder, the replacement cost is high once it is damaged. The coded potentiometer rotation distance measurement method mainly uses angle sensor, which is installed at the rocker pin of coal winning machine. The drum height value is obtained by directly obtaining the swing angle value of coal winning machine rocker arm by angle sensor. However, this method is affected by the underground environment condition, and the failure rate is high. At the same time, the connecting rod mechanism at the rocker pin is easy to wear, which will reduce the reliability of the obtained drum height value and cause measurement dead zone. SUMMARY

[0003] The present application provides a rocker arm inclination detection method based on MEMS accelerometer. Compared with the traditional method, the sensor volume is small, the installation is simple and easy to disassemble, the cost is low, the installation position is less affected by the environment, and the method is not prone to failure, and has strong practicability.

[0004] The main technical scheme of the present application is as follows:

[0005] A rocker arm inclination detection method based on MEMS accelerometer, three three-axis MEMS accelerometers a, b and c are installed in the electric control box of coal winning machine and on the left and right rockers respectively, and protective covers are arranged outside the MEMS accelerometers b and c. The y detection axes of the MEMS accelerometers b and c are parallel to the swing axes of the left and right rockers respectively. When the electric control box and the left and right rockers are in a horizontal state, the z detection axes of the MEMS accelerometers b and c are vertically upward, and the x detection axes are horizontally extended and away from the swing axes of the respective rockers. The detection axes of the MEMS accelerometer a are parallel to the corresponding detection axes of the MEMS accelerometer c. The following measurement steps are performed:

[0006] S1. Obtain the original acceleration vectors A a , A b , A c output by the MEMS accelerometers a, b and c respectively.

[0007] In coordinate representation, A a =(A xa , A ya , A za ), A b =(A xb , A yb , A zb ), A c =(A xc , A yc , A zc ), x, y, z represent three detection axes of the MEMS accelerometer, A xi , A yi , A zi are components of the acceleration vector A i on the x, y, z axes, and i represents a, b, c;

[0008] S2. Filtering the original acceleration vector to filter out high-frequency high-amplitude vibration noise and impact noise to obtain a filtered acceleration vector A′ a , A′ b , A′ c :

[0009] A′ a =(A′ xa , A′ ya , A′ za ), A′ b =(A′ xb , A′ yb , A′ zb ), A′ c =(A′ xc , A′ yc , A′ zc ),

[0010] A′ xi , A′ yi , A′ zi are components of the acceleration vector A i ′ on the x, y, z axes, and i represents a, b, c;

[0011] S3. Calculate the inclination angle values θ a , θ b , θ c of the filtered acceleration vectors A′ a , A′ b , A′ c relative to the horizontal plane, respectively:

[0012]

[0013] S4. Calculate the included angles θL , θ R :

[0014] θ L = θ b + θ a , θ R = θ c - θ a , relative to the horizontal plane passing through the swing axis of the corresponding swing arm, θ L , θ R are positive when the left and right swing arms are in the upper swing state, and θ L , θ R are negative when the left and right swing arms are in the lower swing state. θ a is positive when the right end of the electric control box is higher than the left end, and vice versa.

[0015] In step S2, preferably, a critical damping low-pass filter is used for filtering processing. In the frequency domain, the amplitude square function expression of the critical damping low-pass filter is:

[0016]

[0017] In the formula, n is the order, ω is the angular frequency, ω c is the low-pass filter cutoff angular frequency, and η is a real constant,

[0018] In step S2, an eight-order critical damping low-pass filter is used for filtering processing. The transfer function of the filter is:

[0019]

[0020] In the formula, ω ω c is the low-pass filter cutoff angular frequency, ω c = 2π·f c , f c is the low-pass filter cutoff frequency, and f c = 0.5 Hz.

[0021] The sampling frequency of the MEMS accelerometer can be 26.667 KHz.

[0022] Before the filtering processing in step S2, the length of each original acceleration vector is also calculated, and all original acceleration vectors A a , A b , and A c taken at the same sampling time as the original acceleration vector with a length of zero are removed.

[0023] In step S4, the maximum and minimum values of the inclination angles of the left and right swing arms per second are calculated, and the absolute value of the difference between the maximum and minimum values, i.e., the range value R, is calculated max When the range value R max is greater than the maximum swing angle of the corresponding swing arm per second, all the data collected in the second is discarded.

[0024] The threshold value of the maximum swing angle of the swing arm per second is set to 1°.

[0025] Further, in step S3, the mean values of the filtered and calculated inclination angles θ a , θ b , and θ c in the m groups in the set sampling period are calculated.

[0026] According to , the average inclination angle value θ of a set sampling period is calculated, and in step S4, the average angles θ of the left and right swing arms relative to the shearer body in a set sampling period are calculated using the formula as the actual output left and right swing arm inclination detection values.

[0027] The set sampling period can be between 0.2 and 1 second.

[0028] The present application has the following advantages:

[0029] Using MEMS accelerometers as core sensors, the sensing part is small, flexible, and low in cost. In the case of not affecting the overall shearer structural strength, it can be installed in the appropriate position in the shearer shell through screw fastening, and the sensor itself has high integration, and is not prone to water immersion, hardware damage, and other faults, so the failure rate is low, and it is easy to disassemble, replace, and maintain, with low maintenance cost.

[0030] Since the critical damping low-pass filter is used to filter the original data signal output by the MEMS accelerometer, the influence of high-frequency high-amplitude vibration noise, impact noise, and other interference generated during the shearer coal cutting process on the shearer swing arm inclination detection is effectively suppressed, the problem of accelerometer measurement angle accuracy is solved, the inclination measurement precision is improved, and the reliability of the detection result is ensured.

[0031] For the critical damping low-pass filter used in the present application, since it only has a pole on the negative real axis, the impulse response cannot oscillate, and the step response cannot vibrate or overshoot. Compared with other low-pass filters, it can reach the steady state value with faster response speed and without overshoot, ensuring that the shearer does not cause abnormal conditions such as cutting the top during automatic operation, and ensuring the safety of equipment operation.

[0032] The inclination measurement error of the method is 0.3°, which is much lower than the inclination measurement error of the mine inclination sensor based on the MEMS accelerometer principle in the current market. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 A flow chart of the rocker arm inclination detection method of the application;

[0034] Figure 2 A schematic diagram of the installation direction of the three-axis MEMS accelerometer on the coal mining machine in the reference state;

[0035] Figure 3 A principle diagram of inclination calculation of the rocker arm inclination detection method of the application;

[0036] Figure 4 A magnitude-frequency characteristic diagram of the eight-order critical damping low-pass filter;

[0037] Figure 5 A phase-frequency characteristic diagram of the eight-order critical damping low-pass filter;

[0038] Figure 6 A step response diagram of the eight-order critical damping low-pass filter. DETAILED DESCRIPTION

[0039] As shown in Figure 1 , 2 , the application discloses a rocker arm inclination detection method based on a MEMS accelerometer, which comprises the following steps: three three-axis MEMS accelerometers a, b and c are respectively installed in an electric control box and on left and right rocker arms of a coal mining machine, and are fixed relative to the electric control box shell and the left and right rocker arm shells. The accelerometers on the rocker arms can be directly fixed on the coal wall side end face of the rocker arm shell, and protective covers are arranged outside the MEMS accelerometers b and c to protect the sensors and the wiring from the working environment. The y detection axes of the MEMS accelerometers b and c are respectively parallel to the swing axes of the left and right rocker arms, and the swing axes of the left and right rocker arms are parallel, so the y detection axes of the MEMS accelerometers b and c are also parallel, Figure 2 , 3 The extension directions of the swing axes of the left and right rocker arms are perpendicular to the paper surface direction in the visual angle. When the electric control box (as a part of the machine body, which can also be called the machine body, the same below) and the left and right rocker arms are in a horizontal state (this state can be called a reference state, and the left and right rocker arms are in a horizontal state, which means that the inclination of the left and right rocker arms relative to the horizontal plane is zero, and the swing axes of the left and right rocker arms are at the same height, which is externally manifested as that the left and right ends of the electric control box are at the same height), the z detection axes of the MEMS accelerometers b and c are vertically upward, the x detection axes are horizontally extended and away from the swing axes of the respective rocker arms, and the detection axes of the MEMS accelerometer a are parallel to the corresponding detection axes of the MEMS accelerometer c.

[0040] In summary, according to the right-hand rule, under the reference state, the z-axis of MEMS accelerometers a, b, and c are all upward, the x-axis are to the right, left, and right respectively, and the y-axis is... Figure 2 From the perspective of the paper, the directions are perpendicular to the paper and pointing inward, outward, and inward, respectively, which means pointing to the coal wall side, the goaf side, and the coal wall side of the coal mining machine.

[0041] The installation positions and orientations of MEMS accelerometers a, b, and c were determined during the design of the coal mining machine rocker arm and electrical control box. The corresponding positioning and connection structures on the rocker arm and electrical control box were manufactured together with the rocker arm and electrical control box body, ensuring the accurate installation position of the MEMS accelerometers.

[0042] Perform the following measurement steps:

[0043] S1. MEMS accelerometers a, b, and c are operational, acquiring the raw acceleration vector A output by MEMS accelerometers a, b, and c respectively. a A b A c :

[0044] Represented using coordinates, A a =(A xa A ya A za ), A b =(A xb A yb A zb ), A c =(A xc A yc A zc ), where x, y, and z represent the three sensing axes of the MEMS accelerometer. A xi A yi A zi These are acceleration vectors A and B, respectively. i The components on the x, y, and z axes, where i represents a, b, and c;

[0045] S2. The original acceleration vector is filtered to remove high-frequency, high-amplitude vibration noise and impact noise, and to eliminate the significant fluctuations in the accelerometer output angle value caused by the severe vibration transmitted from the rocker arm to the accelerometer, thus obtaining the filtered acceleration vector A′. a A′ b A′ c :

[0046] A′ a =(A′) xa A′ ya A′ za ), A′ b =(A′)xb A′ yb A′ zb ), A′ c =(A′) xc A′ yc A′ zc ),

[0047] A′ xi A′ yi A′ zi These are acceleration vectors A and B, respectively. i The components of ' on the x, y, and z axes, where i represents a, b, and c;

[0048] S3. Calculate the filtered acceleration vector A′ respectively. a A′ b A′ c The angle of inclination θ relative to the horizontal plane a θ b θ c :

[0049]

[0050] S4. For example Figure 3 As shown, calculate the angles θ between the left and right rocker arms and the electrical control box (representing the angles between the left and right rocker arms and the coal mining machine body). L θ R :

[0051] θ L =θ b +θ a θ R =θ c -θ a When the left and right rockers are in an upward swing state, θ is relative to the horizontal plane passing through the swing axis of the corresponding rocker arm. L θ R When θ is positive and in the downward swing state L θ R The value is negative. When the right end of the electrical control box is higher than the left end, θ is recorded as θ. a A positive value indicates a positive value, and vice versa.

[0052] It should be noted that the installation orientation of MEMS accelerometers a, b, and c is not unique. For example, some detection axes can be set in the opposite direction to the aforementioned orientations. In this case, the sign of the corresponding angle values ​​needs to be changed when substituting them into the formula for calculation. For another example, MEMS accelerometer c can also have its x-axis vertically upward and its y-axis horizontally to the right.

[0053] Further, the filtering process in step S2 is preferably performed using a critically damped low-pass filter, and the amplitude squared function of the critically damped low-pass filter in the frequency domain is:

[0054]

[0055] where n is the order, ω is the angular frequency, ω c is the low-pass filter cutoff angular frequency, and η is a real constant,

[0056] When n = 1, the critically damped low-pass filter is essentially a first-order inertial link, which only has a pole on the negative real axis. Thus, it can be concluded that a high-order critically damped low-pass filter is obtained by cascading and decoupling multiple first-order low-pass filters, or is equivalent to a cascaded circuit composed of multiple real double-pole second-order low-pass filters. After the signal passes through the filter, a group time delay will be generated, and the length of the time delay will determine the time of returning to the receiver terminal.

[0057] The critically damped low-pass filter proposed in the present application can enhance the selectivity of the frequency response as the order of the filter increases. The present application preferably uses an eight-order critically damped low-pass filter, i.e., n = 8. This order of filter can provide a steeper transition between the passband and the stopband, so that it can more effectively suppress signals above the cutoff frequency while keeping signals below the cutoff frequency almost unaffected, and compared with higher-order filters, it has higher data processing efficiency, lower delay, and lower requirement for hardware computing capacity.

[0058] The transfer function of the eight-order critically damped low-pass filter is as follows:

[0059]

[0060] where ω c = 2π·f c , f c is the low-pass filter cutoff frequency, and f c = 0.5 Hz.

[0061] Figures 4-6 The amplitude-frequency characteristic diagram, phase-frequency characteristic diagram, and step response diagram of the eight-order critically damped low-pass filter are shown, and the corresponding cutoff frequency is 0.5 Hz. According to the amplitude-frequency characteristic diagram, it can be seen that the eight-order critically damped low-pass filter can more effectively suppress signals above the cutoff frequency while keeping signals below the cutoff frequency almost unaffected, which is very important for applications that need to strictly filter out high-frequency noise or interference. As can be seen from the step response diagram, the cutoff frequency of 0.5 Hz has a good response speed, which has a low delay while ensuring the filtering performance.

[0062] The sampling frequency of the MEMS accelerometer can be 26.667KHz. Of course, other settings can be made according to actual needs (for example, adapting to the processing speed of different processors, meeting the needs of different scenes) to improve the flexibility of the design. In theory, the faster the sampling frequency, the more accurate the tilt angle value obtained after the application.

[0063] Further, the rocker tilt angle detection method of the application can be optimized in one or more of the following aspects:

[0064] 1. In step S2, preferably, the module of each original acceleration vector is calculated before filtering, and all original acceleration vectors A collected at the same sampling time as the original acceleration vector with a module of zero are removed. a b c For example, the modules of the original acceleration vectors A at bt ct , L at , L bt , L ct are calculated at a certain sampling time t. It is found that the module L bt is equal to zero, indicating that the MEMS accelerometer b is not working properly and no reliable data is collected. Therefore, A at , A bt , A ct are removed to ensure the accuracy and reliability of subsequent rocker tilt angle calculation.

[0065] 2. In step S4, preferably, the maximum and minimum values of the left and right rocker tilt angles per second are calculated, as well as the absolute value of the difference between the maximum and minimum values, i.e. the range value R max When the range value R max is greater than the maximum swing angle of the corresponding rocker per second, it is considered that the filter used cannot completely filter out the external impact on the accelerometer (for example, large pieces of coal falling on the coal machine during the process of cutting coal seam), resulting in errors in the collected data and the errors are large. Therefore, all data collected in that second are discarded, so that the data collected by the accelerometer can be more accurate, which is beneficial to subsequent improvement of the accuracy of calculating the tilt angle.

[0066] Taking the sampling frequency of the accelerometer as 26.667KHz as an example, 26667 groups of data can be collected per second, and the angle can be calculated to obtain 26667 groups of tilt angle data per second. The first group of tilt angle data can be represented by θ1, including θ L1 , θ R1 , the second group of tilt angle data can be represented by θ2, including θ L2 , θ​​​​R2 , until the 26667th set of inclination data θ 26667 , including θ L26667 , θ R26667 If the range value R Lmax , θ Rmax of one of the left and right swing arms is greater than the maximum swing angle per second of the corresponding swing arm, then all the data collected in that second is discarded.

[0067] In practical applications, the setting of the maximum swing angle per second threshold value of the swing arm needs to be determined according to different types of coal mining machines. In this embodiment, according to product requirements, the error value of the inclination detection of the swing arm needs to be less than 0.3°. According to actual tests, the maximum swing speed of the swing arm of a certain type of coal mining machine is about 0.7° per second. That is, assuming that the swing arm swings at the maximum speed, considering the inclination detection error, the range value R max is at most 1°, so the maximum swing angle per second threshold value of the swing arm can be set to 1°, and data greater than the threshold value can be determined as invalid data and discarded.

[0068] 3. In step S3, the m sets of filtered and calculated inclination values θ a , θ b , θ c in a set sampling period can also be subjected to mean value processing.

[0069] Specifically, an average inclination value θ of a set sampling period is calculated according to and in step S4, the average included angle θ of the left and right swing arms relative to the machine body in a set sampling period is calculated using the formula as the actual output left and right swing arm inclination detection value.

[0070] The set sampling period is determined according to actual needs, for example, it can be valued between 0.2-1 seconds.

[0071] In this way, by calculating the mean value of multiple sets of data, the data collected by the accelerometer can be optimized, which can reduce the random errors existing when the accelerometer collects data, and is conducive to subsequent improvement of the accuracy of the cutting height control.

[0072] The embodiments of the present application are described in detail above, but the content described is only the preferred embodiments of the present application and cannot be considered as limiting the scope of the implementation of the present application. Any equivalent changes and improvements made within the scope of the present application should still be attributed to the scope of the present patent.

Claims

1. A MEMS accelerometer-based method of detecting the tilt angle of a swing arm, characterized by: Three three-axis MEMS accelerometers a, b, c are respectively installed in an electric control box and left and right swing arms of the coal mining machine, and the MEMS accelerometers b, c are respectively provided with protective covers, the y detection axes of the MEMS accelerometers b, c are respectively parallel to swing axes of the left and right swing arms, the z detection axes of the MEMS accelerometers b, c are vertically upward, the x detection axes of the MEMS accelerometers b, c are horizontally extended and away from the swing axes of the respective swing arms when the electric control box and the left and right swing arms are in a horizontal state, the detection axes of the MEMS accelerometer a are parallel to the respective corresponding detection axes of the MEMS accelerometer c, and the following measurement steps are performed: S1. Obtain raw acceleration vectors A outputted by MEMS accelerometers a, b, c respectively a , A b , A c : In coordinate representation, A a = (A xa , A ya , A za ), A b = (A xb , A yb , A zb ), A c = (A xc , A yc , A zc ), x, y, z represent three detection axes of the MEMS accelerometer, A xi , A yi , A zi are components of the acceleration vector A i on the x, y, z axes, and i represents a, b, c; S2. Filtering the original acceleration vector to filter out high-frequency high-amplitude vibration noise, impact noise, to obtain a filtered acceleration vector A' a , A' b , A' c : A' a = (A' xa , A' ya , A' za ), A' b = (A' xb , A' yb , A' zb ), A' c = (A' xc , A' yc , A' zc ), A′ xi , A′ yi , A′ zi are the components of the acceleration vector A i ′ on the x, y, z axes, i representing a, b, c; S3. Calculate the filtered acceleration vector A' respectively a , A' b , A' c the inclination value θ of the relative horizontal plane a , θ b , θ c : S4. Calculate the angle θ of the left and right rocker arms relative to the control box L , θ R : θ L =θ b +θ a θ R =θ c -θ a When the left and right rockers are in an upward swing state, θ is relative to the horizontal plane passing through the swing axis of the corresponding rocker arm. L θ R When θ is positive and in the downward swing state L θ R When the value is negative, θ is recorded as the right end of the electrical control box being higher than the left end. a A positive value indicates a positive value, and vice versa.

2. The MEMS accelerometer-based rocker tilt angle detection method of claim 1, wherein: The critical damping low-pass filter is used for filtering in step S2, and the amplitude square function expression of the critical damping low-pass filter in the frequency domain is: where n is the order, ω is the angular frequency, ω c is a low-pass filter cut-off angular frequency, and η is a real normal number, 3. The MEMS accelerometer-based rocker tilt angle detection method of claim 1, wherein: The eight-order critical damping low-pass filter is used for filtering in step S2, and the transfer function of the filter is: wherein ω c = 2π · f c , f c is the low-pass filter cut-off frequency, f c = 0.5 Hz.

4. The MEMS accelerometer-based rocker tilt angle detection method of claim 3, wherein: The sampling frequency of the MEMS accelerometer is 26.667KHz.

5. The MEMS accelerometer-based rocker tilt angle detection method of claim 1, 2, 3, or 4, wherein: The module length of each original acceleration vector is calculated before the filtering process in step S2, and all original acceleration vectors A taken at the same sampling time as the original acceleration vector with a module length of zero are eliminated a , A b , A c .

6. The MEMS accelerometer-based rocker tilt angle detection method of claim 1, 2, 3, or 4, wherein: In step S4, the maximum and minimum values of the inclination angle of each of the left and right swing arms per second are calculated, as well as the absolute value of the difference between the maximum and minimum values of the inclination angle, i.e. the range value R max When the range value R max is greater than the maximum swing angle of the corresponding swing arm per second, all the data collected in that second is discarded.

7. The MEMS accelerometer-based rocker tilt angle detection method of claim 6, wherein: The threshold of the maximum swing angle of the swing arm per second is set to 1°.

8. The MEMS accelerometer-based rocker tilt angle detection method of claim 1, 2, 3, or 4, wherein: In step S3, the m sets of filtered calculated inclination values θ a , θ b , θ c are subjected to mean value processing; According to The average inclination value of a set sampling period is calculated And in step S4, the formula The average angle of the left and right rockers relative to the shearer body in a set sampling period is calculated As the actual output of the left and right rocker inclination detection value.

9. The MEMS accelerometer-based rocker tilt angle detection method of claim 8, wherein: The set sampling period is valued between 0.2-1 seconds. The set sampling period is valued between 0.2-1 seconds.

Images