Bracket sensor data processing method for optimizing simulation effect of fully mechanized coal mining face based on spline interpolation method

Abstract

The invention discloses a bracket sensor data processing method for optimizing a simulation effect of a fully mechanized coal mining face based on a spline interpolation method, which comprises the following steps of: performing multi-step processing and optimization on original data of a bracket sensor, eliminating wrong data and abnormal values, making up for data missing, and improving the simulation effect of the fully mechanized coal mining face. The processed sensor data can more truly reflect the actual attitude of the working face bracket, so that high-quality basic data is provided for simulation, and the accuracy and reliability of fully mechanized coal mining face simulation are improved; the reasonable range of sensor values is set according to the working face mining data, and the original data is adjusted according to the reasonable range, so that the goodness of fit between the data processing process and the actual mining condition is ensured, and the field applicability of simulation is improved.

Description

Technical Field

[0001] The present invention belongs to the field of working face support sensor data processing, and specifically relates to a support sensor data processing method for optimizing the simulation effect of a fully mechanized mining working face based on a spline interpolation method. Background Art

[0002] With the development of my country's coal industry, the automation degree of coal mining equipment in coal mine fully mechanized working faces is getting higher and higher. Objectively, various sensors installed on the automated coal mining working faces are required to collect more accurate information such as displacement and pressure of coal mining equipment in order to provide basic data for the automated operation of coal mine equipment. At present, the commonly used simulation is to obtain the posture characteristics of the support through sensors on special parts of the support body in order to restore the posture of the hydraulic support. However, the working environment underground in coal mines is harsh, usually with high humidity and strong electrical interference. At the same time, there are often many faults in a large number of support sensors, which leads to distortion of some sensor data, and then to strange displacement of the support and obvious distortion of the working face simulation. Even if the sensor is fault-free, the real-time and accuracy of its data are difficult to guarantee. For example, when the support push rod is pushed from left to right in sequence, since the change time of the sensor data of each support may not be delivered instantly, it presents a disordered push in the simulation process, resulting in chaotic scraper push and slide, and the simulation effect is very poor. For example, the accuracy of the support sensor is affected by many factors, and there will be a certain error. It is not easy to see a single support, but it will be more obvious when multiple supports are compared. When the bottom is not lifted, the sensor data of the working surface bracket base should be a row of smooth curves, but the corresponding data obtained from the working surface fluctuates to a certain extent, which will also affect the simulation effect of the bracket.

[0003] Existing data processing methods for mining sensors, such as Kalman filtering and mean filtering, are suitable for data processing of sensors such as temperature, humidity and gas concentration, but are less applicable to tilt sensors and length sensors on brackets, because the values ​​of these sensors usually change irregularly, so special data processing methods can only be adopted. The time and trend of sensor value changes cannot be predicted without considering the combined process flow, and the data can only be optimized through special methods. Summary of the invention

[0004] In view of the defects and problems existing in the existing data processing methods of sensors on the supports of comprehensive mining working faces, the present invention provides a data processing method of sensors on supports for optimizing the simulation effect of comprehensive mining working faces based on the spline interpolation method, thereby solving the problem that the data processing methods of existing mining sensors have low applicability to the posture sensors on the supports, and improving the accuracy and reliability of the simulation of comprehensive mining working faces.

[0005] The solution adopted by the present invention to solve the technical problem is: a support sensor data processing method for optimizing the simulation effect of a fully mechanized mining working face based on a spline interpolation method, comprising the following steps: Step 1: Set the maximum exclusion number and minimum exclusion number of each posture sensor of the working face support according to the actual mining data of the working face; use the maximum exclusion number and minimum exclusion number of each posture sensor of the working face support to eliminate the error value in the data of each sensor of the working face support; Step 2: confirm the highest and lowest values ​​of the posture sensors of the working face supports according to the actual mining data of the working face; Step 3: According to the maximum exclusion number and the minimum exclusion number of each posture sensor of the working surface support and the highest value and the lowest value of each posture sensor of the working surface support, adjust the values ​​sent by each corresponding posture sensor of the working surface support; Step 4: For the erroneous data that have been excluded by the posture sensors of the working surface support, the cubic spline interpolation method is used to complete the data. When the excluded erroneous data is the end data, the valid data that has not been excluded at the other end is taken as the data of the point; Step 5: For the posture sensor data that need to be smoothly transitioned between the brackets in the working surface, such data are arranged in order from left to right of the brackets into an array, and each element in the array represents the sensor value of the corresponding bracket; Step 6: Remove abnormal data from the array that needs to be smoothed. The abnormal data removal includes two situations: no real-time geological data and real-time geological data. Step seven, use the spline interpolation method to interpolate and complete the eliminated data, and use cubic spline interpolation based on the non-eliminated data on both sides of the eliminated data point to generate a new bracket sensor data sequence.

[0006] In the process of adjusting the values ​​sent by the posture sensors of the working surface support in the step three, when the values ​​sent by the posture sensors of the working surface support are between the minimum exclusion number and the minimum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is adjusted to the minimum value of the posture sensor corresponding to the working surface support; when the values ​​sent by the posture sensors of the working surface support are between the maximum exclusion number and the maximum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is adjusted to the maximum value of the sensor corresponding to the working surface support; when the values ​​sent by the posture sensors of the working surface support are between the maximum value and the minimum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is not adjusted.

[0007] In the step 6, when there is no real-time geological data, the data is eliminated by using a data difference elimination method, which includes the following steps: S1, according to the type of sensor, set the maximum angle difference or height difference ΔM of the bracket as the basis for difference judgment; S2, determine whether the difference Δ of adjacent data is greater than ΔM, if it is greater, the data is discarded; S3, after elimination, continue to use the last uneliminated data as a reference to judge the next data; S3, the judgment standard is ΔM multiplied by the number of data points, until the data meets the conditions.

[0008] In the step 6, when there is real-time geological data, removing data includes the following steps: S1, obtaining the top and bottom heights of the corresponding support position points according to geological data; S2, reversely calculating the ideal posture of the bracket according to the top and bottom heights of the corresponding bracket position points; S3, correcting the support posture sensor data according to the posture sensor data of each position in the ideal posture of the support.

[0009] In the step seven, after generating the new bracket sensor data sequence, the new bracket sensor data sequence is fitted.

[0010] The method also includes step eight, which sends out a sensor fault alarm of a corresponding level according to the data discharge and rejection situation, and reminds the staff to check and repair the faulty sensor.

[0011] The fault alarm includes a low-level alarm and a high-level alarm. A high-level alarm is issued for the posture sensor that excludes the error value in the first step, and a low-level alarm is issued for the posture sensor that excludes the data in the remaining steps. The posture sensor includes an inclination sensor and a length sensor. The posture sensor includes a top beam sensor, a guardrail sensor, a front beam sensor, a connecting rod sensor and a base sensor. The length sensor includes a mining height sensor, a push rod sensor and a telescopic beam sensor.

[0012] Beneficial effects of the invention: The present invention provides a support sensor data processing method for optimizing the simulation effect of a fully mechanized mining working face based on a spline interpolation method, which has the following beneficial effects: 1. Through multi-step processing and optimization of the original data of the bracket sensor, erroneous data and outliers are eliminated, and data missing is compensated, so that the processed sensor data can more realistically reflect the actual posture of the working face bracket, thereby providing high-quality basic data for simulation and improving the accuracy and reliability of comprehensive mining working face simulation; the reasonable range of sensor values ​​is set through the working face mining data, and the original data is adjusted accordingly, ensuring the consistency between the data processing process and the actual mining situation, and improving the on-site applicability of the simulation.

[0013] 2. It integrates real-time geological data, obtains the top and bottom plate heights of the support position, inverts the ideal posture of the support, and corrects the posture data accordingly, so that the simulation process can dynamically adapt to changes in geological conditions of the working face, with stronger real-time and adaptability.

[0014] 3. A variety of mathematical methods are used to optimize data quality, and spline interpolation, data smoothing, outlier removal and other technologies are comprehensively used to improve the continuity, smoothness and accuracy of sensor data from multiple angles, laying a data foundation for high-simulation visual presentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic diagram of the process of eliminating erroneous data. DETAILED DESCRIPTION

[0016] The present invention is further described below in conjunction with the accompanying drawings and embodiments. Example

[0017] This embodiment provides a support sensor data processing method for optimizing the simulation effect of a fully mechanized mining working face based on a spline interpolation method, comprising the following steps: Step one, according to the actual mining data of the working face, set the maximum exclusion number and the minimum exclusion number of each posture sensor of the working face support; use the maximum exclusion number and the minimum exclusion number of each posture sensor of the working face support to eliminate the erroneous values ​​in the data of each sensor of the working face support; the posture sensor includes an inclination sensor and a length sensor, and the posture sensor includes a top beam sensor, a guardrail sensor, a front beam sensor, a connecting rod sensor, a base sensor and other sensors, and the length sensor includes a mining height sensor, a push rod sensor, a telescopic beam sensor and other sensors.

[0018] In this embodiment, the error value of the working face support posture sensor is defined as a value that is significantly different from the actual value and the error is already unacceptable. Setting the maximum exclusion number and the minimum exclusion number can effectively exclude most of the support sensor error data. For example: when the working face mining height is 4-5 meters, the value sent by the support's mining height sensor is less than 2 meters, it is regarded as error data (this minimum value can be flexibly adjusted) and should be excluded, because the height of such supports must be at least higher than 2 meters; similarly, the maximum exclusion number is set to 6m, and the value sent by the support mining height sensor is greater than 6 meters, it is regarded as error data (this maximum value can be flexibly adjusted).

[0019] Step 2: confirm the highest and lowest values ​​of the posture sensors of the working face supports according to the actual mining data of the working face; Step three, according to the maximum exclusion number and the minimum exclusion number of each posture sensor of the working surface support and the highest value and the lowest value of each posture sensor of the working surface support, adjust the values ​​sent by the corresponding posture sensors of the working surface support; when the values ​​sent by each posture sensor of the working surface support are between the minimum exclusion number and the minimum value of the corresponding posture sensor of the working surface support, adjust the value sent by the posture sensor to the minimum value of the corresponding posture sensor of the working surface support; when the values ​​sent by each posture sensor of the working surface support are between the maximum exclusion number and the maximum value of the corresponding posture sensor of the working surface support, adjust the value sent by the posture sensor to the maximum value of the sensor corresponding to the working surface support; when the values ​​sent by each posture sensor of the working surface support are between the maximum value and the minimum value of the corresponding posture sensor of the working surface support, the value sent by the posture sensor is not adjusted.

[0020] Step 4, use the interpolation method to complete the erroneous data that have been excluded by the posture sensors of the working surface support. When the excluded erroneous data is end data, take the valid data that has not been excluded at the other end as the data of this point; in this embodiment, the interpolation method uses one of the cubic spline interpolation method and the polynomial interpolation method to complete the data.

[0021] Step 5: For the posture sensor data that need to be smoothly transitioned between the brackets in the working surface, such data are arranged in order from left to right of the brackets into an array, and each element in the array represents the sensor value of the corresponding bracket; Step 6: Remove abnormal data from the array that needs to be smoothed. The removal of abnormal data includes two situations: no real-time geological data and real-time geological data. When there is no real-time address data, the data is removed by using the data difference elimination method, including the following steps: S1, according to the type of sensor, set the maximum angle difference or height difference ΔM of the bracket as the basis for difference judgment; S2, determine whether the difference Δ of adjacent data is greater than ΔM, if it is greater, the data is discarded; S3, after elimination, continue to use the last uneliminated data as a reference to judge the next data; S3, the judgment standard is ΔM multiplied by the number of data points, until the data meets the conditions.

[0022] Assume that the height difference ΔM is 0.1m. If the difference Δ between the value of the next array n+1 and the value of the current array n is greater than ΔM, the value of the n+1 array is considered unreasonable and the data is discarded. Figure 1 The Δ of the n+1th bit is greater than 0.1, so the n+1th bit is eliminated. Continue to judge n+2. When judging the n+2 data, n should be used as the standard to judge, based on the limit angle multiplied by the number of separated data, until the data meets the judgment conditions. Figure 1The Δ of the n+2 position is less than or equal to 0.1×2, which meets the requirements and the data is retained. The n+3 position is based on the last retained data, that is, n+2. Obviously, Figure 1 In this example, n+3 bits are reserved.

[0023] When real-time geological data is available, data removal includes the following steps: S1. Obtain the top and bottom heights of the corresponding support position points according to the geological data. The geological data should include information such as coal seam thickness, inclination, top and bottom plate elevation at different positions of the working face. S2, reversely calculate the ideal posture of the bracket according to the top and bottom heights of the corresponding bracket position points obtained, S3, correcting the support posture sensor data according to the posture sensor data of each position in the ideal posture of the support.

[0024] Step seven, use the spline interpolation method to interpolate and complete the eliminated data, and use cubic spline interpolation based on the non-eliminated data on both sides of the eliminated data point to generate a new bracket sensor data sequence. Example

[0025] The difference between Example 2 and Example 1 is that it also includes step eight, according to the data discharge and elimination situation, a sensor fault alarm of a corresponding level is issued to remind the staff to check and repair the faulty sensor; the fault alarm includes a low-level alarm and a high-level alarm, a high-level alarm is issued for the posture sensor that eliminates the error value in the first step, and a low-level alarm is issued for the posture sensor whose data is eliminated in the remaining steps. Example

[0026] The difference between Example 3 and Example 2 is that, after generating a new bracket sensor data sequence, step 7 fits the new bracket sensor data sequence to generate an optimized bracket sensor data sequence, thereby improving the authenticity and visualization effect of the comprehensive mining face simulation. There are many ways of fitting, for example, using one of the fitting methods such as piecewise low-order interpolation, piecewise cubic Hermite interpolation, polynomial fitting, least squares method, power function, logarithmic function and exponential function.

[0027] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

Claims

1. A support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method, characterized in that: The following steps are involved: Step 1: Set the maximum exclusion number and minimum exclusion number of each posture sensor of the working face support according to the actual mining data of the working face; use the maximum exclusion number and minimum exclusion number of each posture sensor of the working face support to eliminate the error value in the data of each sensor of the working face support; Step 2: confirm the highest and lowest values ​​of the posture sensors of the working face supports according to the actual mining data of the working face; Step 3: According to the maximum exclusion number and the minimum exclusion number of each posture sensor of the working surface support and the highest value and the lowest value of each posture sensor of the working surface support, adjust the values ​​sent by each corresponding posture sensor of the working surface support; Step 4: For the erroneous data that have been excluded by the sensors of each position of the work surface support, the cubic spline interpolation method is used to complete the data. When the erroneous data that has been excluded is the end data, the valid data that has not been excluded at the other end is taken as the data of this point; Step 5: For the posture sensor data that need to be smoothly transitioned between the brackets in the working surface, such data are arranged in order from left to right of the brackets into an array, and each element in the array represents the sensor value of the corresponding bracket; Step 6: Remove abnormal data from the array that needs to be smoothed. The abnormal data removal includes two situations: no real-time geological data and real-time geological data. Step seven, use the spline interpolation method to interpolate and complete the eliminated data, and use cubic spline interpolation based on the non-eliminated data on both sides of the eliminated data point to generate a new bracket sensor data sequence.

2. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: In the process of adjusting the values ​​sent by the posture sensors of the working surface support in the step three, when the values ​​sent by the posture sensors of the working surface support are between the minimum exclusion number and the minimum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is adjusted to the minimum value of the posture sensor corresponding to the working surface support; when the values ​​sent by the posture sensors of the working surface support are between the maximum exclusion number and the maximum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is adjusted to the maximum value of the sensor corresponding to the working surface support; when the values ​​sent by the posture sensors of the working surface support are between the maximum value and the minimum value of the posture sensor corresponding to the working surface support, the value sent by the posture sensor is not adjusted.

3. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: In the step 6, when there is no real-time geological data, the data is eliminated by using a data difference elimination method, which includes the following steps: S1, according to the type of sensor, set the maximum angle difference or height difference ΔM of the bracket as the basis for difference judgment; S2, determine whether the difference Δ of adjacent data is greater than ΔM, if it is greater, the data is discarded; S3, after elimination, continue to use the last uneliminated data as a reference to judge the next data; S3, the judgment standard is ΔM multiplied by the number of data points, until the data meets the conditions.

4. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: In the step 6, when there is real-time geological data, removing data includes the following steps: S1, obtaining the top and bottom heights of the corresponding support position points according to geological data; S2, reversely calculating the ideal posture of the bracket according to the top and bottom heights of the corresponding bracket position points; S3, correcting the support posture sensor data according to the posture sensor data of each position in the ideal posture of the support.

5. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: In the step seven, after generating the new bracket sensor data sequence, the new bracket sensor data sequence is fitted.

6. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: The method also includes step eight, which sends out a sensor fault alarm of a corresponding level according to the data discharge and rejection situation, and reminds the staff to check and repair the faulty sensor.

7. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 6 is characterized in that: The fault alarm includes a low-level alarm and a high-level alarm. A high-level alarm is issued for the posture sensor that eliminates the error value in the first step, and a low-level alarm is issued for the posture sensor that eliminates the data in the remaining steps.

8. The support sensor data processing method for optimizing the simulation effect of fully mechanized mining working face based on spline interpolation method according to claim 1 is characterized in that: The posture sensor includes an inclination sensor and a length sensor. The posture sensor includes a top beam sensor, a guardrail sensor, a front beam sensor, a connecting rod sensor and a base sensor. The length sensor includes a mining height sensor, a push rod sensor and a telescopic beam sensor.

Images