A coal mining machine positioning system and method integrating odometer and Hall proximity sensor

Abstract

The present invention discloses a coal machine positioning system and method that integrates an odometer and a Hall proximity sensor, which is applied to the positioning field of coal mine machinery. The system addresses the problems of complex calculations in the prior art and the difficulty in overcoming the influence of the complex underground environment on positioning accuracy. The system installs a high-precision odometer device inside the coal machine, installs a signal transmission module of the Hall proximity sensor at the center position of the coal machine close to the side of the scraper conveyor, and a data processing substation connected to each Hall proximity sensor is installed on the corresponding bracket. Optical fiber communication is adopted between the odometer on the coal machine and the chute computer. The coal machine is operated and the odometer value is recorded each time the Hall proximity sensor is triggered. The odometer data each time the Hall proximity sensor is triggered is recorded, and the Holt exponential smoothing filter algorithm is used to predict the distance difference between the next bracket position and the current bracket position. The prediction result is used for subsequent judgment of the position of the coal machine and updating of the operation trajectory of the coal machine.

Description

Technical Field

[0001] The invention belongs to the field of safe production, and in particular relates to a positioning technology for coal mine machinery. Background Art

[0002] With the advancement of modern technology, intelligent coal mining has become a key tool for improving coal mine safety and mining efficiency. The operational efficiency and safety of intelligent coal mining machinery are directly related to mine production efficiency and miner safety. In fully mechanized mining faces, precise positioning of coal mining machinery is crucial for improving coal mining efficiency and ensuring operational safety. However, traditional coal mining machinery positioning methods often have limitations and cannot meet the requirements for high precision and reliability.

[0003] Currently, the positioning of coal mining machinery primarily relies on sensor technology and data processing techniques. Common sensors include infrared sensors, proximity switches, and encoders, which are used to measure the operating distance and position of coal mining machinery. However, due to the unique characteristics of the coal mining environment, such as coal dust, water mist, and electromagnetic interference, these sensors are susceptible to interference, resulting in large positioning errors. For example, relying on encoders for positioning can easily result in cumulative errors, infrared sensors have high installation requirements, and the coal dust environment can obstruct infrared signals, making them difficult to trigger. Alternatively, some methods combine multiple positioning sensors, such as infrared sensors with inertial navigation systems. Based on the coal mining machinery positions obtained by each sensor, data processing algorithms are used to calculate the confidence level of the coal mining machinery position. While these methods offer some improvement in positioning accuracy compared to traditional single-sensor positioning methods, the algorithms involved are relatively complex and cannot effectively reduce the impact of the complex underground environment on positioning accuracy.

[0004] In fully mechanized mining faces, precise positioning of coal mining machinery is crucial for improving coal mining efficiency and ensuring operational safety. However, existing coal mining machinery positioning methods have certain limitations and cannot meet the requirements of high precision and high reliability. The specific issues are as follows:

[0005] 1. Traditional sensors are easily affected by factors such as coal dust, water mist, and electromagnetic interference in the coal mine environment, resulting in large positioning errors.

[0006] 2. Traditional data processing methods are often unable to effectively fuse data from multiple sensors to further improve positioning accuracy.

[0007] 3. During actual operation, due to the influence of various factors, the sensor may not be triggered normally, resulting in failure of the positioning system. Summary of the Invention

[0008] In order to solve the above technical problems, the present invention proposes a coal machine positioning system and method that integrates an odometer and a Hall proximity sensor, which improves the accuracy of the coal machine positioning system by effectively integrating data from multiple sensors.

[0009] One of the technical solutions adopted by the present invention is: a coal mining machine positioning system integrating an odometer and a Hall proximity sensor, comprising: an odometer, a Hall proximity sensor, a signal processing substation and a data centralized processing unit;

[0010] The odometer is fixed inside the coal mining machine body; the odometer and the data centralized processing unit are connected to the network through the optical fiber in the coal mining machine power cable to ensure the stability and transmission efficiency of the communication line;

[0011] The Hall proximity sensor includes a signal receiving unit and a signal generating unit, wherein the signal generating unit is fixed to the side of the coal mining machine body close to the scraper conveyor baffle, and the signal generating unit is used to provide a magnetic field signal; the signal receiving unit is installed at the scraper conveyor baffle position corresponding to the horizontal center of each bracket;

[0012] Each Hall proximity sensor is connected to a corresponding signal processing substation; the specific signal processing substation is connected to the signal receiving unit of the Hall proximity sensor and is used to receive the signal of the Hall proximity sensor. Each signal processing substation is connected to form a data bus. When a Hall proximity sensor is triggered, the trigger signal is received by the signal processing substation, and the signal processing substation sends the sensor status data to the data concentration processing unit through the data bus;

[0013] The data center processing unit records the mileage value each time the Hall proximity sensor is triggered, and establishes a gap prediction model between two adjacent supports based on the mileage value when the Hall proximity sensor is triggered multiple times; and confirms the positioning of the coal machine based on the gap prediction value.

[0014] The gaps between multiple adjacent brackets are obtained based on the mileage values ​​when the Hall proximity sensor is triggered multiple times and the width of the bracket itself; these gap data are averaged to obtain the average gap value;

[0015] If the absolute value of the difference between the predicted position and the odometer data when the Hall proximity sensor is triggered is less than or equal to the gap average value, the coal machine positioning is considered accurate; otherwise, the position corresponding to the current bracket is calculated based on the odometer data corresponding to the previous bracket, the gap average value and the width of the bracket itself.

[0016] The second technical solution adopted by the present invention is: a coal mining machine positioning method integrating an odometer and a Hall proximity sensor, comprising:

[0017] S1. Use a high-precision odometer to measure the displacement of the shearer in the direction of the working face;

[0018] S2. Use the trigger signal of the Hall proximity sensor to detect the bracket number corresponding to the location of the coal machine;

[0019] S3. Record the mileage value each time the Hall proximity sensor is triggered;

[0020] S4. Establish a gap prediction model between two adjacent brackets based on the odometer value when the Hall proximity sensor is triggered multiple times;

[0021] S5. Confirm the positioning of the coal machine based on the gap prediction value.

[0022] Beneficial effects of the present invention: By combining the data of the odometer and the Hall proximity sensor, the present invention can provide the precise position of the coal machine in the comprehensive mining working face. The odometer provides absolute position information, while the Hall proximity sensor provides relative position information. The combination of the two improves the overall accuracy of the positioning system. For example, the commonly used positioning methods on the market currently include infrared trigger positioning, which uses the coal machine to transmit infrared signals, and each bracket is installed with an infrared receiving sensor. During normal operation, the position of the coal machine can be located to a specific bracket number. The installation requirements are high, and the anti-interference ability is weak. When the working face is in production, coal dust interference may cause the infrared receiving sensor to fail to be triggered; there are also higher-precision solutions, such as using an encoder to calculate the relative position data of the coal machine operation. This solution is more accurate than the infrared sensor, and the sensor accuracy can reach the millimeter level. However, during the forward movement of the underground working face, this solution will produce position drift, which is difficult to calibrate the position. Moreover, due to the inconsistent bracket spacing, there may be large errors when converting the relative displacement into the bracket number corresponding to the coal machine. This solution uses two sensor fusions, and the positioning accuracy can be stabilized to the centimeter level. The present invention also has the following advantages:

[0023] 1. Anti-interference capability: The odometer is installed inside the coal mining machine and will not come into direct contact with the production environment during the operation of the coal mining machine. The design of the Hall proximity sensor enables it to operate stably in the coal mining environment, reducing the impact of factors such as coal dust, water mist, and electromagnetic interference on positioning accuracy.

[0024] 2. Easy to implement: The positioning system of the present invention has a simple structure, is easy to install and maintain, does not require complex hardware or software support, and is easy to promote and apply in actual coal mines.

[0025] 3. Robustness: Even if the Hall proximity sensor fails to trigger normally in some cases, the present invention can still accurately determine the position of the coal machine through the odometer data and prediction model, thereby improving the robustness of the system.

[0026] 4. Cost-effectiveness: Compared with existing complex and high-cost positioning systems, the present invention provides a cost-effective positioning solution that can improve positioning accuracy while reducing the operating costs of coal mining enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a structural diagram of the coal mining machine positioning system that integrates an odometer and a Hall proximity sensor according to the present invention;

[0028] Figure 2 This is a flow chart of the coal mining machine positioning method that integrates an odometer and a Hall proximity sensor according to the present invention. DETAILED DESCRIPTION

[0029] To facilitate those skilled in the art to understand the technical content of the present invention, the present invention is further explained below with reference to the accompanying drawings.

[0030] This embodiment provides a coal mining machine positioning system that integrates an odometer and a Hall proximity sensor. Figure 1 As shown, it includes: a high-precision odometer, a Hall proximity sensor, a signal processing substation and a data centralized processing unit.

[0031] The high-precision odometer is fixed inside the shearer. Its measurement accuracy must meet the positioning requirements of the fully mechanized mining face and typically incorporates a high-precision encoder and gyroscope. The odometer's installation position ensures it can stably measure the shearer's travel distance during operation.

[0032] The Hall proximity sensor consists of a signal receiving unit and a signal generating unit. The signal transmitting unit is installed at the center of the coal machine, near the scraper conveyor. The signal receiving unit is installed on the conveyor baffle corresponding to the center of each bracket. The signal generating unit provides a stable magnetic field signal. The signal receiving unit is installed on the coal machine, near the conveyor baffle. When the bracket passes the signal transmitting unit installed at the center of the coal machine, near the scraper conveyor, a trigger signal is generated. The selection of the Hall proximity sensor should take into account its anti-interference performance and trigger sensitivity in the coal mine environment, such as bipolar Hall sensors.

[0033] The signal processing substation is connected to the Hall proximity sensor signal receiving unit. The signal processing substation has a built-in MCU processing unit for receiving the signal from the connected Hall proximity sensor and analyzing the received data to determine whether the sensor is triggered normally. Each substation is connected to form a data bus, which transmits the Hall proximity sensor signal status of the entire working face; the data processing substation connected to each Hall proximity sensor is installed on the corresponding bracket, and each substation has a corresponding number. When a sensor captures the coal machine proximity signal, the signal is uploaded to the corresponding signal processing substation. The signal processing substation analyzes the signal. When it is confirmed that the coal machine triggered the sensor, it sends a trigger signal to the centralized processing unit via the Modbus bus. The data centralized processing unit uses a chute computer.

[0034] The data concentration processing unit receives mileage data from the odometer and bus data from the signal processing substation, analyzes and integrates the data, and ultimately determines the accurate location of the coal mining machine. Fiber optic communication is used between the odometer on the coal mining machine and the chute computer, transmitting mileage data in real time. The chute computer records and analyzes the received sensor status data along with the odometer data.

[0035] like Figure 2 As shown, the coal machine positioning process in this embodiment is as follows:

[0036] A1. Analyze data to obtain bracket gap

[0037] Run the coal machine and record the mileage value each time the Hall proximity sensor is triggered. During the operation of the coal machine, the data recording and analysis module monitors the mileage value and the triggering of the Hall proximity sensor in real time. When the coal machine approaches a bracket, the Hall proximity sensor is triggered and the mileage value at that moment is recorded, which is X1, X2, X3...X n The width of the bracket itself is a fixed value m. By analyzing the difference in odometer values ​​between two adjacent triggers, it can be clearly found that the difference in odometer values ​​between the two adjacent triggers is inconsistent with the bracket width. This is because the underground geological conditions are complex and there will be a certain gap between the brackets during operation. Based on the odometer data and the width of the bracket itself, the gap between two adjacent brackets can be calculated as:

[0038] d n =X n -X n-1 –m

[0039] A2. Filtering and predicting the next spacing

[0040] The Holt exponential smoothing filter algorithm is used to predict the next bracket gap, and the predicted value is expressed as Y n The calculation method of the exponential smoothing filter algorithm is as follows:

[0041] A21. Update the forecast level term Y n :

[0042] Y n =αd n-1 +(1-α)(Y n-1 +T n-1 +Φ n-1 )

[0043] Among them, Y n is the horizontal item of the support position n predicted based on the position n-1, that is, the gap between the support position n and the support position n predicted based on the position n-1; d n-1 is the gap between two adjacent brackets calculated at position n-1; Y n-1is the predicted bracket gap at the previous position, initially Y1=d1; T n-1 is the trend value at position n-1, T n-1 The initial value of can be determined based on the distribution of supports under initial conditions:

[0044] If the spacing between brackets increases gradually, T n The initial value of is set to a positive number;

[0045] If the spacing between brackets is gradually reduced, the T n The initial value of is set to a negative number;

[0046] At the initial position, since there is no historical data, T n The initial value of is set to 0;

[0047] Φ n is a nonlinear correction term, calculated as follows:

[0048] Φ n =γΦ n-1 +(1-γ)(X n -X n-1 ) 2

[0049] Where γ is the smoothing coefficient of the nonlinear correction term. Usually, the value of γ is small, such as 0.01 to 0.1. In this case, γ is set to 0.01. Then, based on the deviation between the predicted results and the actual data, the value of γ is gradually adjusted until the optimal value is found.

[0050] Φ n-1 is the nonlinear correction term of the previous position;

[0051] (X n -X n-1 ) 2 It is the square of the odometry difference between the current position and the previous position, and is used to capture the nonlinear change of the bracket gap.

[0052] α is a smoothing coefficient used to balance the importance of current observations and historical forecasts, and its value range is 0 to 1. A22. Update the trend term T of the forecast value (n+1) :

[0053] T n+1 =β(X n -X n-1 )+(1-β)(T n +Ψ n )

[0054] Among them, T n+1is the trend term of the distance between the predicted next support position n+1 and the current support position; β is the trend coefficient, which determines the degree of influence of the new observation value on the trend term update, and its value range is 0 to 1; X n is the odometer difference value triggered by the Hall proximity sensor at the current position n; X n-1 is the odometer difference triggered by the Hall proximity sensor at the previous position n-1; T n is the trend value of the current position n; n is the autoregressive correction term, calculated as follows:

[0055] Ψ n =δΨ n-1 +(1-δ)(T n -T n-1 )

[0056] Among them, δ is the smoothing coefficient of the autoregressive correction term, ranging from 0 to 1; Ψ n-1 is the autoregressive correction term of the previous position; (T n -T n-1 ) is the difference between the trend values ​​of the current position and the previous position, which is used to capture the time series autocorrelation of the trend term.

[0057] Those skilled in the art will recognize that when using the Holt exponential smoothing filter algorithm to predict the next stent gap, at least two gap data points must be calculated using the equation "d = Xn - Xn - 1 - m." The first data point is used to initialize the horizontal term, and the second data point (along with the first data point) is used to estimate the initial trend term. As more data is collected, the algorithm can continuously update the horizontal and trend terms, improving prediction accuracy.

[0058] A23. Calculate the predicted displacement value △X between the next support position triggered by the coal machine and the current support position

[0059] △X n =Y n +T n +m

[0060] In the present invention, the distance △X that the coal machine needs to move from the current Hall proximity sensor trigger position to the next position trigger can be obtained by the Holter smoothing filter algorithm. n During the actual operation of the coal machine, the actual mileage s=|Xn-Xn-1| from the current Hall proximity sensor trigger position to the next position trigger can be obtained, and the historical s and △X can be recorded. n Data, calculate the current position of the coal machine.

[0061] A3. Calculate the current coal machine position

[0062]

[0063] Where λ is the offset parameter:

[0064]

[0065] m is the width of the bracket;

[0066] To calculate the average value of the bracket gap

[0067]

[0068] A4. Update prediction parameters

[0069] According to the change in the predicted position and the actual position En, and the error change △E caused by multiple measurements n , the α and β values ​​in the Holt smoothing filter algorithm can be dynamically adjusted to make the algorithm's prediction results closer to the real data. The calculation is as follows:

[0070] A41. Calculate the prediction error:

[0071] En=△X n -△X n-1

[0072] where is the actual observed value.

[0073] A42. Calculate the error change:

[0074] △En=En-En-1

[0075] If △En>0, the surface prediction value is lower than the actual value;

[0076] If △En<0, the surface prediction value is higher than the actual value;

[0077] A43. Update parameters α and β:

[0078] If △En>0:

[0079] anew=αold+kΔα

[0080] βnew=βold-kΔβ

[0081] If △En<0:

[0082] anew=αold-kΔα

[0083] βnew=βold+kΔβ

[0084] Where k is an adjustment factor. The value of k should be chosen to ensure that the updated α and β values ​​are within a preset range, for example, 0 < α < 1 and 0 < β < 1. The initial value of α (the horizontal smoothing coefficient) is typically 0.2 to 0.3. This range is intended to balance the impact of current observations and historical forecasts, allowing the model to adapt to short-term data fluctuations.

[0085] The initial value of β (trend term smoothing coefficient) is usually taken as 0.1 to 0.2. This range is to ensure that the trend estimate is not overly sensitive to a single observation, while being able to capture long-term trends in the data.

[0086] Therefore, initially, α and β can be set to α = 0.2 and β = 0.1, respectively, and αnew and βnew are updated to the Holt smoothing filter algorithm.

[0087] A5. Monitor the triggering of the Hall proximity sensor

[0088] During the operation of the coal machine, the chute computer monitors the triggering status of the Hall proximity sensor in real time. If the sensor triggers normally, the coal machine and the predicted position can be verified based on the difference between the odometer data at the time of triggering and the odometer data at the last triggering. The verification method is as follows:

[0089] Absolute error between the coal machine's current calculated position and the odometer recorded position:

[0090] w=|X n -x n |

[0091] when When the calculated error is lower than the average gap of the bracket, the calibration is passed and the sensor position is accurate. n As the coal machine position at this time; When , it means that the calculation error is greater than the average gap of the bracket. At this time, the verification fails, the predicted value is discarded, and the offset parameter λ is set to 1. At this time, the coal machine position is expressed as

[0092]

[0093] Those skilled in the art will appreciate that the embodiments described herein are intended to aid the reader in understanding the principles of the present invention, and it should be understood that the scope of the present invention is not limited to such specific descriptions and embodiments. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A coal mining machine positioning system integrating an odometer and a Hall proximity sensor, characterized in that: include: Odometer, Hall proximity sensor, signal processing substation and data centralized processing unit; The odometer is fixed inside the coal mining machine body; The odometer is connected to the data concentration processing unit; The Hall proximity sensor includes a signal receiving unit and a signal generating unit, wherein the signal generating unit is fixed to the side of the coal mining machine body close to the scraper conveyor baffle, and the signal generating unit is used to provide a magnetic field signal; the signal receiving unit is installed at the scraper conveyor baffle position corresponding to the horizontal center of each bracket; Each Hall proximity sensor is connected to a corresponding signal processing substation; the specific signal processing substation is connected to the signal receiving unit of the Hall proximity sensor and is used to receive the signal of the Hall proximity sensor. Each signal processing substation is connected to form a data bus. When a Hall proximity sensor is triggered, the trigger signal is received by the signal processing substation, and the signal processing substation sends the sensor status data to the data concentration processing unit through the data bus; The data collection processing unit records the mileage value each time the Hall proximity sensor is triggered, and establishes a gap prediction model between two adjacent brackets based on the mileage values ​​when the Hall proximity sensor is triggered multiple times; The coal machine positioning is confirmed based on the gap prediction value.

2. The coal mining machine positioning system integrating odometer and Hall proximity sensor according to claim 1 is characterized in that: The gaps between the multiple adjacent brackets are obtained based on the mileage values ​​when the Hall proximity sensor is triggered multiple times and the width of the bracket itself; These gap data are averaged to obtain the gap average value; If the absolute value of the difference between the predicted position and the odometer data when the Hall proximity sensor is triggered is less than or equal to the gap average value, the coal machine positioning is considered accurate; otherwise, the position corresponding to the current bracket is calculated based on the odometer data corresponding to the previous bracket, the gap average value and the width of the bracket itself.

3. A coal mining machine positioning method integrating odometer and Hall proximity sensor, characterized in that: Based on the coal machine positioning system integrating an odometer and a Hall proximity sensor according to claim 1 or 2, the positioning method specifically includes the following steps: S1. Use a high-precision odometer to measure the displacement of the shearer in the direction of the working face; S2. Use the trigger signal of the Hall proximity sensor to detect the bracket number corresponding to the location of the coal machine; S3. Record the mileage value each time the Hall proximity sensor is triggered; S4. Establish a gap prediction model between two adjacent brackets based on the odometer value when the Hall proximity sensor is triggered multiple times; S5. Confirm the positioning of the coal machine based on the gap prediction value.

4. The coal mining machine positioning method integrating odometer and Hall proximity sensor according to claim 3 is characterized in that: In step S4, a gap prediction model between two adjacent brackets is established. The specific process is as follows: Obtaining distances between a plurality of adjacent brackets based on the plurality of odometer data recorded in step S3; Based on the fixed width of the bracket itself, the gap between multiple adjacent brackets is calculated; By performing Holt exponential smoothing filtering on the gaps between these multiple adjacent brackets, the horizontal term of bracket n is predicted, which is recorded as Y n ; The expression is: Y n =αd n-1 +(1-α)(Y n-1 +T n-1 +Φ n-1 ) Among them, α is the smoothing coefficient, d n-1 is the calculated actual gap between bracket n-1 and bracket n-2, Y n-1 is the gap between scaffold n-2 and scaffold n-1 predicted, T n-1 The trend term representing the gap between the predicted support n and support n-1, T n-1 =β(X n-2 -X n-3 )+(1-β)(T n-2 +Ψ n-2 ), β represents the trend coefficient, X n-2 Indicates the odometer data triggered by the Hall proximity sensor corresponding to bracket n-2, X n-3 Indicates the odometer data triggered by the Hall proximity sensor corresponding to bracket n-3, T n-2 The trend term for the predicted spacing between bracket n-1 and bracket n-2, Ψ n-2 Indicates T n-2 The corresponding autoregressive correction term, Ψ n-2 =δΨ n-3 +(1-δ)(T n-2 -T n-3 ), T n-3 The trend term for the predicted distance between bracket n-2 and bracket n-3, Ψ n-3 Indicates T n-3 The corresponding autoregressive correction term, δ is the smoothing coefficient of the autoregressive correction term, Φ n-1 Indicates Y n-1 The nonlinear correction term, Φ n-1 =γΦ n-2 +(1-γ)(X n-1 -X n-2 ) 2 , γ is the smoothing coefficient of the nonlinear correction term, Φ n-2 Indicates Y n-2 The nonlinear correction term, Y n-2 represents the predicted gap between scaffold n-2 and scaffold n-1.

5. The coal mining machine positioning method integrating odometer and Hall proximity sensor according to claim 4 is characterized in that: It also includes updating α and β. The updating process is as follows: First, based on the predicted support gap, calculate the displacement prediction value between the coal machine triggering the next support position and the current support position: △X n =Y n +T n+1 +m Among them, △X n T represents the distance the coal machine needs to move from support n to support n+1. n+1 represents the trend term of the predicted distance between bracket n+1 and bracket n, and m represents the width of the bracket itself; Next, calculate the prediction error value: In=△X n -△X n-1 Among them, △X n-1 Indicates the distance the coal machine needs to move from support n-1 to support n; The current α and β are recorded as αold and βold respectively; the updated α and β are recorded as αnew and βnew respectively; If △En>0: αnew=αold+kΔα βnew=βold-kΔβ If △En<0: αnew=αold-kΔα βnew=βold+kΔβ Where k is the adjustment factor.

6. The coal mining machine positioning method integrating odometer and Hall proximity sensor according to claim 5 is characterized in that: The implementation process of step S5 is: Calculate the absolute error between the coal machine's current calculated position and the odometer recorded position: w=|X n -x n | when When , it means that the calculation error is lower than the average clearance of the bracket The verification is passed, and the sensor position is accurate. n As the coal machine position at this time; When , it means that the calculation error is greater than the average gap of the bracket. At this time, the verification fails and the predicted value is discarded. At this time, the coal machine position is expressed as λ is the offset parameter.

7. The coal mining machine positioning method integrating odometer and Hall proximity sensor according to claim 6 is characterized in that: The calculation formula of λ is:

Images