An intelligent speed regulation method for a long-distance two-way belt conveyor used in mines

Abstract

An intelligent speed regulation method for a long-distance two-way belt conveyor in coal mines. During operation, monitoring signals are collected in real time, amplified and preliminarily denoised by a signal preprocessing module, and then filtered by a Kalman filter module to output high-precision monitoring data. The load is divided into three levels: low load, medium load, and high load. According to the actual on-site operation situation, a standard fuzzy logic rule base is established by using the empirical induction method. The state credibility is introduced, and according to the size of the diagonal elements of the error covariance matrix, the state credibility is divided into low, medium, and high to assist in judging the reliability of the monitoring data and adjusting the fuzzy control rules to obtain an enhanced fuzzy rule base. The center-of-gravity method is used to convert the output of the fuzzy controller into an accurate speed regulation instruction, and the conveying speed is precisely controlled by using the speed regulation instruction. This method has high control accuracy, short response time, and can significantly improve the robustness of the control system.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent speed regulation, and particularly relates to an intelligent speed regulation method for a long-distance two-way belt conveyor for mining use. Background Technique

[0002] As a key material transportation device, the belt conveyor plays a crucial role in the coal industry. In recent years, with the continuous growth of coal demand, in order to meet the requirements of output and efficiency, the underground coal mining volume needs to be improved urgently. The transportation distance of the belt conveyor for mining use is often up to thousands of meters. When there is insufficient coal volume or the conveyor runs empty during the transportation process, it not only reduces the transportation efficiency, but also causes a large amount of electric energy waste. At the same time, the long-term operation will accelerate the wear of equipment components (such as idlers, drums, etc.), thus shortening the service life of the equipment and increasing the maintenance and replacement costs.

[0003] At present, the long-distance transportation for mining mainly relies on the DSJ100 type quick-disassembly two-way belt conveyor. This device adopts a double-layer structure design. The upper layer is used for transporting coal, and the lower layer transports production auxiliary materials. However, such devices have significant limitations in the scenario of a kilometer-level transportation distance: frequent disassembly and assembly lead to an increase in downtime and maintenance costs; although the double-layer structure improves the space utilization rate, it exacerbates the mechanical complexity and poses higher requirements for real-time speed regulation control.

[0004] In the prior art, the fuzzy logic control method can dynamically adjust the conveying speed through sensor data, but there are some defects. Due to the influence of environmental noise and signal interference, the data collected by the sensor usually has large errors. Especially in the long-distance transportation scenario, the sensor is vulnerable to more electromagnetic interference, mechanical vibration and environmental temperature and humidity influence. The noise amplitude of the collected original data can reach 20% - 30% of the effective signal, which will further lead to a significant increase in the misjudgment rate of the controller. At the same time, the traditional filtering algorithm depends on a fixed window or threshold, and it is difficult to adapt to dynamic working conditions such as sudden load changes and sudden slope changes. The contradiction between noise reduction and real-time performance is prominent. Therefore, there is an urgent need to provide an intelligent speed regulation method applicable to the long-distance two-way belt conveyor for mining use. Summary of the Invention

[0005] Aiming at the problems existing in the above-mentioned prior art, the present invention provides an intelligent speed regulation method for a long-distance two-way belt conveyor for mining use. This method can effectively reduce the interference of noise, has high control precision and short response time, can significantly improve the robustness of the control system, and can thus significantly improve the transportation efficiency of the belt conveyor. It is applicable to the energy-saving optimization control of the long-distance two-way belt conveyor in the complex working conditions of coal mines.

[0006] To achieve the above object, the present invention provides an intelligent speed regulation method for a long-distance two-way belt conveyor for mining use, including the following steps:

[0007] Step 1: Establish a sensor monitoring network;

[0008] Install multiple sensors at multiple key monitoring nodes on the long-distance two-way belt conveyor for mining, and form a sensor monitoring network through multiple sensors;

[0009] Step 2: Collect operation data and perform filtering processing;

[0010] During the operation of the long-distance two-way belt conveyor for mining, collect monitoring signals in real time through the sensor monitoring network, and use the signal preprocessing module for amplification and preliminary denoising processing, then perform the fusion of multi-modal data, and use the distributed Kalman filtering architecture for filtering processing to output high-precision monitoring data;

[0011] Step 3: Fuzzify the key parameters in the high-precision monitoring data;

[0012] Divide the load into three levels, namely low load, medium load and high load; divide the slope into three levels, namely downhill, relatively flat and uphill; divide the conveying speed into three levels, namely deceleration, normal working speed and acceleration;

[0013] Step 4: Establish a standard fuzzy logic rule base according to the actual on-site operation conditions by using the empirical induction method;

[0014] IF load = low load AND slope = uphill THEN conveying speed = increase by one gear speed;

[0015] IF load = low load AND slope = relatively flat THEN conveying speed = increase by two gear speeds;

[0016] IF load = low load AND slope = downhill THEN conveying speed = decrease by one gear speed;

[0017] IF load = medium load AND slope = uphill THEN conveying speed = normal working speed;

[0018] IF load = medium load AND slope = relatively flat THEN conveying speed = increase by one gear speed;

[0019] IF load = medium load AND slope = downhill THEN conveying speed = decrease by one gear speed;

[0020] IF load = high load AND slope = uphill THEN conveying speed = decrease by one gear speed;

[0021] IF the load = high load AND the slope = relatively flat THEN the conveying speed = normal operating speed;

[0022] IF the load = high load AND the slope = downhill THEN the conveying speed = reduced second gear speed;

[0023] Step Five: Expand the fuzzy logic control rules;

[0024] Introduce the state credibility. According to the diagonal element size of the error covariance matrix, the state credibility is divided into low, medium, and high to assist in judging the reliability of the monitoring data and adjusting the fuzzy control rules;

[0025] Step Six: Output the speed regulation command to perform intelligent control on the long-distance two-way belt conveyor for mining;

[0026] Adopt the centroid method to convert the output of the fuzzy controller into an accurate speed regulation command, and send the speed regulation command to the motor drive control system of the long-distance two-way belt conveyor for mining to achieve precise control of the speed of the long-distance two-way belt conveyor for mining using the speed regulation command.

[0027] As an optimization, in Step One, the multiple sensors include a load sensor, a speed sensor, a current sensor, a slope sensor, a temperature sensor, and a torque sensor.

[0028] Furthermore, in order to effectively remove the interference of noise and effectively ensure the accuracy of the monitoring data, in Step Two, the process of filtering using the distributed Kalman filter architecture is as follows:

[0029] S21: During the operation of the long-distance two-way belt conveyor for mining, the monitoring signals are collected in real time through the sensor monitoring network, and the signal preprocessing module is used for amplification and preliminary denoising processing;

[0030] S22: Define the state vector according to formula (1) , establish the state equation based on the motion model of the long-distance two-way belt conveyor for mining according to formula (2), and establish the observation equation based on the motion model of the long-distance two-way belt conveyor for mining according to formula (3);

[0031] (1);

[0032] (2);

[0033] (3);

[0034] In the formula, is the conveyor belt The speed at a moment, is the conveyor belt The slope at a moment, Conveyor belt The load at a moment, is the transpose operation; is the state transition matrix based on the motion model of the long-distance two-way belt conveyor for mine use, , is the sampling time interval, is the acceleration due to gravity, is the conveyor belt The sine value of the slope at a moment, is the friction coefficient of the conveyor belt, is the conveyor belt at The load at a moment, is The state at a moment, is the control input matrix, , is the equivalent mass of the conveyor belt, is the motor torque control amount, is the process noise; is The sensor measurement value at a moment; is the observation matrix, , is the calibration coefficient of the load sensor; is the observation noise; is The state at a moment;

[0035] S23: Continuously correct the state equation through the observation equation, and use the Kalman gain K K Recursively update the state estimate value and the covariance matrix, and output high-precision monitoring data.

[0036] Furthermore, in order to more accurately control the long-distance two-way belt conveyor for mine use, in step three, the specific process of fuzzy processing is as follows:

[0037] For the load : The load in the range of 0 ≤ < 150 kg / m is regarded as a low load, the load in the range of 150 kg / m ≤ < 300 kg / m is regarded as a medium load, and the load in the range of 300 kg / m ≤ ≤ 600 kg / m is regarded as a high load;

[0038] For the slope : The The slope within the range of < -5° is regarded as a downhill slope, and -5° ≤ The slope within the range of < 5° is regarded as a relatively flat slope, and 5° ≤ The slope within the range is regarded as an uphill slope;

[0039] For the speed : Keeping the speed unchanged is regarded as the normal working speed, increasing the speed is regarded as acceleration, and decreasing the speed is regarded as deceleration.

[0040] Furthermore, in order to effectively ensure stable and precise speed regulation control of the conveyor under dynamic working conditions such as sudden load changes or transient slope changes, in step five, the state credibility is calculated according to formula (4) ; When < 0.3, external PID is used for compensation so that the output of the fuzzy controller is multiplied by a decay coefficient of 0.5. When 0.3 ≤ < 0.6, the standard fuzzy rule base is used for control. When 0.6 ≤ , the enhanced fuzzy rule base is enabled to reduce the response speed;

[0041] (4);

[0042] In the formula, is the trace of the error covariance matrix, the sum of the diagonal elements, is the error covariance matrix of the Kalman filter, is the set maximum acceptable error matrix.

[0043] Furthermore, in order to effectively monitor the data and status at key monitoring nodes, in step one, the load sensor is installed on the idler support in the middle section of the long-distance two-way belt conveyor for mining, and is used to collect the actual load signal in real time; the speed sensor is installed at the driving wheel of the long-distance two-way belt conveyor for mining, and is used to collect the conveyor speed signal in real time; the current sensor is installed on the driving motor of the long-distance two-way belt conveyor for mining, and is used to collect the current signal in real time; multiple slope sensors are installed on the frame of the long-distance two-way belt conveyor for mining at intervals of a set distance in sequence, and are used to collect the belt slope signal in the area where they are located in real time; the temperature sensor is installed on the driving motor and controller of the long-distance two-way belt conveyor for mining, and is used to collect the temperature signal in real time; the torque sensor is installed on the driving motor of the long-distance two-way belt conveyor for mining, and is used to collect the torque signal in real time.

[0044] Furthermore, in order to improve the control accuracy, in step five, the construction process of the enhanced fuzzy rule base is as follows:

[0045] S51: Enhance the fuzzification of key parameters in high-precision monitoring data;

[0046] Divide the load into five levels: low load, slightly low load, medium load, slightly high load, and high load; divide the slope into five levels: steep downhill, downhill, flat, uphill, and steep uphill;

[0047] S52: Establish an enhanced fuzzy logic rule base using the empirical induction method according to the actual on-site operation situation;

[0048] IF load = low load AND slope = steep uphill THEN conveyor speed = increase by 2 gear speeds;

[0049] IF load = low load AND slope = uphill THEN conveyor speed = increase by 3 gear speeds;

[0050] IF load = low load AND slope = flat THEN conveyor speed = increase by 4 gear speeds;

[0051] IF load = low load AND slope = downhill THEN conveyor speed = increase by 2 gear speeds;

[0052] IF load = low load AND slope = steep downhill THEN conveyor speed = normal operating speed;

[0053] IF load = slightly low load AND slope = steep uphill THEN conveyor speed = normal operating speed;

[0054] IF load = slightly low load AND slope = uphill THEN conveyor speed = increase by 1 gear speed;

[0055] IF load = slightly low load AND slope = flat THEN conveyor speed = increase by 2 gear speeds;

[0056] IF load = slightly low load AND slope = downhill THEN conveyor speed = increase by 1 gear speed;

[0057] IF load = slightly low load AND slope = steep downhill THEN conveyor speed = decrease by 1 gear speed;

[0058] IF load = medium load AND slope = steep uphill THEN conveyor speed = decrease by 2 gear speeds;

[0059] IF Load = Medium Load AND Gradient = Uphill THEN Conveyor Speed = Normal Operating Speed;

[0060] IF Load = Medium Load AND Gradient = Flat THEN Conveyor Speed = Normal Operating Speed;

[0061] IF Load = Medium Load AND Gradient = Downhill THEN Conveyor Speed = Normal Operating Speed;

[0062] IF Load = Medium Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduced by 1 Gear Speed;

[0063] IF Load = Slightly Higher Load AND Gradient = Steep Uphill THEN Conveyor Speed = Reduced by 3 Gear Speeds;

[0064] IF Load = Slightly Higher Load AND Gradient = Uphill THEN Conveyor Speed = Reduced by 2 Gear Speeds;

[0065] IF Load = Slightly Higher Load AND Gradient = Flat THEN Conveyor Speed = Reduced by 1 Gear Speed;

[0066] IF Load = Slightly Higher Load AND Gradient = Downhill THEN Conveyor Speed = Normal Operating Speed;

[0067] IF Load = Slightly Higher Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduced by 2 Gear Speeds;

[0068] IF Load = High Load AND Gradient = Steep Uphill THEN Conveyor Speed = Reduced by 4 Gear Speeds;

[0069] IF Load = High Load AND Gradient = Uphill THEN Conveyor Speed = Reduced by 3 Gear Speeds;

[0070] IF Load = High Load AND Gradient = Flat THEN Conveyor Speed = Reduced by 2 Gear Speeds;

[0071] IF Load = High Load AND Gradient = Downhill THEN Conveyor Speed = Reduced by 3 Gear Speeds;

[0072] IF Load = High Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduced by 4 Gear Speeds.

[0073] Furthermore, in order to improve the control accuracy, in S51, the specific process of enhancing the fuzzification process is as follows:

[0074] For the load : The load within the range of 0 ≤ < 150 kg / m is regarded as a low load, the load within the range of 150 kg / m ≤ < 225 kg / m is regarded as a slightly low load, the load within the range of 225 kg / m ≤ < 300 kg / m is regarded as a medium load, the load within the range of 300 kg / m ≤ < 350 kg / m is regarded as a slightly high load, and the load within the range of 350 kg / m ≤ ≤ 600 kg / m is regarded as a high load;

[0075] For the slope : The slope within the range of < -8° is regarded as a steep downhill, the slope within the range of -8° ≤ < -4° is regarded as a downhill, the slope within the range of -4° ≤ < 4° is regarded as flat, the slope within the range of 4° ≤ < 8° is regarded as an uphill, and the slope within the range of 8° ≤ is regarded as a steep uphill.

[0076] In the present invention, the running state of the conveyor is continuously monitored by using a sensor monitoring network, and the sensor data is optimized by Kalman filtering, which can remove the noise and other interference data in the monitoring data, effectively ensuring the accuracy of the monitoring data. Then, the key parameters in the monitoring data are fuzzified, and fuzzy inference is carried out according to a preset fuzzy rule base. At the same time, the centroid method is used to convert the fuzzy output into an accurate speed adjustment command, and then the speed adjustment command is fed back to the motor drive control system to realize the closed-loop control of accelerating, maintaining, and decelerating the motor. This method can ensure that the optimal control command is obtained efficiently and accurately, thereby significantly improving the control accuracy and greatly shortening the control response time. By introducing the state credibility, the filtering and fuzzy control parameters can be updated in real time according to the actual running state during the operation process, effectively ensuring that the conveyor can still maintain stable and accurate speed regulation performance under dynamic working conditions such as sudden load changes or sudden slope changes. The present invention adopts a distributed Kalman filtering architecture to optimize the sensor data, combines a fuzzy logic control strategy and a state credibility evaluation, effectively reducing the sensor noise interference and improving the accuracy of state estimation. Compared with the traditional speed regulation method, the present invention is more adaptable to sudden load changes and environmental changes in a long-distance transportation environment, has higher steady-state accuracy and stronger robustness, and significantly improves the transportation efficiency and equipment service life of the conveyor.

[0077] This method can effectively reduce the interference of noise, with high control precision and short response time, which can significantly improve the robustness of the control system, and then can significantly improve the transportation efficiency of the belt conveyor, and is applicable to the energy-saving optimization control of long-distance two-way belt conveyors under complex working conditions in coal mines. Description of the Drawings

[0078] Figure 1 is the control principle block diagram of the present invention. Detailed Embodiment

[0079] The present invention will be further described below with reference to the drawings.

[0080] As Figure 1 shown, the present invention provides an intelligent speed regulation method for a long-distance two-way belt conveyor for coal mine use, including the following steps:

[0081] Step 1: Establish a sensor monitoring network;

[0082] Install a variety of sensors at multiple key monitoring nodes on the long-distance two-way belt conveyor for coal mine use, and form a sensor monitoring network through the variety of sensors to achieve comprehensive monitoring of the conveyor state;

[0083] Step 2: Collect operation data and perform filtering processing;

[0084] During the operation of the long-distance two-way belt conveyor for coal mine use, collect monitoring signals in real time through the sensor monitoring network, and use the signal preprocessing module to perform amplification and preliminary noise reduction processing, then perform the fusion of multi-modal data, and use the distributed Kalman filter architecture to perform filtering processing;

[0085] Step 3: Fuzzify the key parameters in the high-precision monitoring data;

[0086] Divide the load into three levels, namely low load, medium load and high load; divide the slope into three levels, namely downhill, relatively flat and uphill; divide the conveying speed into three levels, namely deceleration, normal working speed and acceleration;

[0087] Step 4: According to the actual on-site operation situation, establish a standard fuzzy logic rule base by using the empirical induction method, as shown in Table 1;

[0088] Table 1: Standard Fuzzy Logic Rule Base

[0089]

[0090] In Table 1, 0 represents the normal working speed, the negative sign represents deceleration, the specific numerical value of the negative sign corresponds to the speed gear, the positive number represents acceleration, and the specific numerical value corresponds to the speed gear;

[0091] The standard fuzzy logic rule base is represented in the form of "IF THEN" as follows:

[0092] IF the load = low load AND the slope = uphill THEN the conveying speed = increase the speed by one gear;

[0093] IF the load = low load AND the slope = relatively flat THEN the conveying speed = increase the speed by two gears;

[0094] IF the load = low load AND the slope = downhill THEN the conveying speed = decrease the speed by one gear;

[0095] IF the load = medium load AND the slope = uphill THEN the conveying speed = normal operating speed;

[0096] IF the load = medium load AND the slope = relatively flat THEN the conveying speed = increase the speed by one gear;

[0097] IF the load = medium load AND the slope = downhill THEN the conveying speed = decrease the speed by one gear;

[0098] IF the load = high load AND the slope = uphill THEN the conveying speed = decrease the speed by one gear;

[0099] IF the load = high load AND the slope = relatively flat THEN the conveying speed = normal operating speed;

[0100] IF the load = high load AND the slope = downhill THEN the conveying speed = decrease the speed by two gears;

[0101] Step Five: Expand the fuzzy logic control rules;

[0102] Introduce the state credibility. According to the diagonal elements of the error covariance matrix, the state credibility is divided into low, medium, and high to assist in judging the reliability of the monitoring data and adjust the fuzzy control rules;

[0103] Step Six: Output the speed regulation command to perform intelligent control on the long-distance two-way belt conveyor for mining;

[0104] Use the centroid method to convert the output of the fuzzy controller into an accurate speed regulation command, and send the speed regulation command to the motor drive control system of the long-distance two-way belt conveyor for mining to achieve precise control of the speed of the long-distance two-way belt conveyor for mining using the speed regulation command.

[0105] As an optimization, in step one, the multiple sensors include a load sensor, a speed sensor, a current sensor, a slope sensor, a temperature sensor, and a torque sensor.

[0106] In order to effectively remove the interference of noise and effectively ensure the accuracy of the monitoring data, in step two, the process of filtering using a distributed Kalman filter architecture is as follows:

[0107] S21: During the operation of the long-distance two-way belt conveyor for mining, the monitoring signals are collected in real time through the sensor monitoring network, and are amplified and preliminarily denoised using the signal preprocessing module;

[0108] S22: Combining the kinematic characteristics of the long-distance two-way belt conveyor for mining and the sensor layout method, define the state vector according to formula (1) ; establish the state equation based on the motion model of the long-distance two-way belt conveyor for mining according to formula (2), so that each node can estimate the current state based on the previous state; establish the observation equation based on the motion model of the long-distance two-way belt conveyor for mining according to formula (3) for the fusion of the observation data;

[0109] (1);

[0110] (2);

[0111] (3);

[0112] In the formula, is the speed of the conveyor belt at time is the slope of the conveyor belt at time is the load of the conveyor belt at time is the transpose operation, is the transpose of; is the state transition matrix based on the motion model of the long-distance two-way belt conveyor for mining. When the slope changes slowly (constant in a short time), the load change is linearly related to the transportation volume, and the speed change is affected by the motor torque, slope resistance, and load. The discretized state equation , the speed change in the first row is affected by the slope resistance and belt friction together. The short-term changes in slope and load in the second and third rows are ignored and assumed to be constant values. is the sampling time interval, is the acceleration due to gravity, is the sine value of the slope of the conveyor belt at time is the friction coefficient of the conveyor belt, is the load of the conveyor belt at time, is the state at time, is the control input matrix, , is the equivalent mass of the conveyor belt, is the motor torque control amount, is the process noise; is the sensor measurement value at time; is the observation matrix, and the pressure is linearly related to the load, , is the calibration coefficient of the load sensor; is the observation noise; is the state at time;

[0113] S23: Continuously correct the state equation through the observation equation, and use the Kalman gain K K to recursively update the state estimate and the covariance matrix, and output high-precision monitoring data.

[0114] In order to perform more precise control on the long-distance bidirectional belt conveyor, in step three, the specific process of fuzzy processing is as follows:

[0115] For the load : Consider the load in the range of 0 ≤ < 150 kg / m as a low load, the load in the range of 150 kg / m ≤ < 300 kg / m as a medium load, and the load in the range of 300 kg / m ≤ ≤ 600 kg / m as a high load;

[0116] For the slope : Consider the slope in the range of < -5° as a downhill slope, the slope in the range of -5° ≤ < 5° as a relatively flat slope, and the slope in the range of 5° ≤ as an uphill slope;

[0117] For the speed : Consider maintaining the speed unchanged as the normal operating speed, increasing the speed as acceleration, and decreasing the speed as deceleration.

[0118] In order to effectively ensure stable and precise speed control of the conveyor under dynamic conditions such as sudden load changes or transient slopes, in step five, calculate the state credibility according to formula (4) ; When When it is less than 0.3, external PID is used for compensation so that the output of the fuzzy controller is multiplied by a decay coefficient of 0.5. When 0.3 ≤ < 0.6, the standard fuzzy rule base is used for control. When 0.6 ≤ , the enhanced fuzzy rule base is enabled to reduce the response speed;

[0119] (4);

[0120] In the formula, is the trace of the error covariance matrix, the sum of the diagonal elements, is the error covariance matrix of the Kalman filter, is the set maximum acceptable error matrix.

[0121] In order to effectively monitor the data and status at the key monitoring nodes, in step one, the load sensor is installed on the idler support in the middle section of the long-distance two-way belt conveyor for mines, preferably at the key nodes of the loading section, for real-time acquisition of the actual load signal; the speed sensor is installed at the driving wheel of the long-distance two-way belt conveyor for mines, for real-time acquisition of the conveying speed signal; the current sensor is installed on the driving motor of the long-distance two-way belt conveyor for mines, for real-time acquisition of the current signal; multiple slope sensors are installed on the frame of the long-distance two-way belt conveyor for mines at intervals of a set distance, for real-time acquisition of the belt slope signal in the area; the temperature sensor is installed on the driving motor and controller of the long-distance two-way belt conveyor for mines, for real-time acquisition of the temperature signal to effectively prevent the driving motor and controller from overheating; the torque sensor is installed on the driving motor of the long-distance two-way belt conveyor for mines, for real-time acquisition of the torque signal.

[0122] In order to improve the control accuracy, in step five, the construction process of the enhanced fuzzy rule base is as follows:

[0123] S51: Perform enhanced fuzzification processing on the key parameters in the high-precision monitoring data;

[0124] The load is divided into five levels, namely low load, slightly low load, medium load, slightly high load and high load; the slope is divided into five levels, namely steep downhill, downhill, flat, uphill and steep uphill;

[0125] In order to improve the control accuracy, in S51, the specific process of the enhanced fuzzification processing is as follows:

[0126] For the load : When 0 ≤ Loads in the range of < 150 kg / m are considered low loads, and 150 kg / m ≤ Loads in the range of < 225 kg / m are considered slightly low loads, and 225 kg / m ≤ Loads in the range of < 300 kg / m are considered medium loads, and 300 kg / m ≤ Loads in the range of < 350 kg / m are considered slightly high loads, and 350 kg / m ≤ ≤ 600 kg / m are considered high loads;

[0127] For the slope : Slopes in the range of < -8° are considered steep downhills, and -8° ≤ Slopes in the range of < -4° are considered downhills, and -4° ≤ Slopes in the range of < 4° are considered flat, and 4° ≤ Slopes in the range of < 8° are considered uphills, and 8° ≤ Slopes in the range of are considered steep uphills.

[0128] S52: According to the actual on-site operation situation, an enhanced fuzzy logic rule base is established using the empirical induction method, as shown in Table 2;

[0129] Table 2: Enhanced Fuzzy Logic Rule Base

[0130]

[0131] Among them, 0 represents the normal working speed, the negative sign represents deceleration, and the specific negative value corresponds to the number of gear positions for reducing the speed. The positive number represents acceleration, and the specific value corresponds to the number of gear positions for increasing the speed.

[0132] The enhanced fuzzy logic rule base is represented in the form of "IF THEN" as follows:

[0133] IF load = low load AND slope = steep uphill THEN conveying speed = increase by 2 gear positions;

[0134] IF load = low load AND slope = uphill THEN conveying speed = increase by 3 gear positions;

[0135] IF load = low load AND slope = flat THEN conveying speed = increase by 4 gear positions;

[0136] IF load = low load AND slope = downhill THEN conveying speed = increase by 2 gear positions;

[0137] IF the load = low load AND the slope = steep downhill THEN the conveying speed = normal operating speed;

[0138] IF the load = slightly low load AND the slope = steep uphill THEN the conveying speed = normal operating speed;

[0139] IF the load = slightly low load AND the slope = uphill THEN the conveying speed = increase by 1 gear speed;

[0140] IF the load = slightly low load AND the slope = flat THEN the conveying speed = increase by 2 gear speeds;

[0141] IF the load = slightly low load AND the slope = downhill THEN the conveying speed = increase by 1 gear speed;

[0142] IF the load = slightly low load AND the slope = steep downhill THEN the conveying speed = decrease by 1 gear speed;

[0143] IF the load = medium load AND the slope = steep uphill THEN the conveying speed = decrease by 2 gear speeds;

[0144] IF the load = medium load AND the slope = uphill THEN the conveying speed = normal operating speed;

[0145] IF the load = medium load AND the slope = flat THEN the conveying speed = normal operating speed;

[0146] IF the load = medium load AND the slope = downhill THEN the conveying speed = normal operating speed;

[0147] IF the load = medium load AND the slope = steep downhill THEN the conveying speed = decrease by 1 gear speed;

[0148] IF the load = slightly high load AND the slope = steep uphill THEN the conveying speed = decrease by 3 gear speeds;

[0149] IF the load = slightly high load AND the slope = uphill THEN the conveying speed = decrease by 2 gear speeds;

[0150] IF the load = slightly high load AND the slope = flat THEN the conveying speed = decrease by 1 gear speed;

[0151] IF Load = Slightly High Load AND Gradient = Downhill THEN Conveyor Speed = Normal Operating Speed;

[0152] IF Load = Slightly High Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduce by 2 - gear Speed;

[0153] IF Load = High Load AND Gradient = Steep Uphill THEN Conveyor Speed = Reduce by 4 - gear Speed;

[0154] IF Load = High Load AND Gradient = Uphill THEN Conveyor Speed = Reduce by 3 - gear Speed;

[0155] IF Load = High Load AND Gradient = Flat THEN Conveyor Speed = Reduce by 2 - gear Speed;

[0156] IF Load = High Load AND Gradient = Downhill THEN Conveyor Speed = Reduce by 3 - gear Speed;

[0157] IF Load = High Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduce by 4 - gear Speed.

[0158] In the present invention, first, the sensor monitoring network is used to continuously monitor the operating state of the conveyor, and the sensor data is optimized by Kalman filtering, which can remove the noise and other interference data in the monitoring data, effectively ensuring the accuracy of the monitoring data. Then, the key parameters in the monitoring data are fuzzified and fuzzy inference is carried out according to the preset fuzzy rule base. At the same time, the centroid method is used to convert the fuzzy output into an accurate speed adjustment instruction, and then the speed adjustment instruction is fed back to the motor drive control system to realize the closed - loop control of accelerating, maintaining, and decelerating the motor. This method can ensure that the optimal control instruction is obtained efficiently and accurately, thereby significantly improving the control accuracy and greatly shortening the control response time. By introducing the state credibility, the filtering and fuzzy control parameters can be updated in real - time according to the actual operating state during the operation process, effectively ensuring that the conveyor can still maintain stable and accurate speed regulation performance under dynamic working conditions such as sudden load changes or transient gradient changes. The present invention adopts a distributed Kalman filtering architecture to optimize the sensor data, combines a fuzzy logic control strategy and state credibility evaluation, effectively reducing the sensor noise interference and improving the accuracy of state estimation. Compared with the traditional speed regulation method, the present invention is more adaptable to sudden load changes and environmental changes in a long - distance transportation environment, has higher steady - state accuracy and stronger robustness, and significantly improves the transportation efficiency and equipment service life of the conveyor.

[0159] This method can effectively reduce the interference of noise, has high control precision and short response time, can significantly improve the robustness of the control system, and further can significantly improve the transportation efficiency of the belt conveyor, and is applicable to the energy-saving optimization control of the long-distance two-way belt conveyor under complex working conditions in coal mines.

Claims

1. An intelligent speed regulation method for a long-distance two-way belt conveyor used in mines, characterized in that, It includes the following steps: Step 1: Establish a sensor monitoring network; Install a variety of sensors at multiple key monitoring nodes on the long-distance two-way belt conveyor for mining, and form a sensor monitoring network through a variety of sensors; Step 2: Collect operation data and perform filtering processing; During the operation of the long-distance two-way belt conveyor for mining, real-time collect monitoring signals through the sensor monitoring network, use the signal preprocessing module for amplification and preliminary denoising processing, then perform the fusion of multi-modal data, and use the distributed Kalman filter architecture for filtering processing to output high-precision monitoring data; Step 3: Fuzzify the key parameters in the high-precision monitoring data; The specific process of fuzzification is as follows: The load is divided into three levels: low load, medium load, and high load; for the load : The load in the range of 0 ≤ < 150 kg / m is regarded as a low load, the load in the range of 150 kg / m ≤ < 300 kg / m is regarded as a medium load, and the load in the range of 300 kg / m ≤ ≤ 600 kg / m is regarded as a high load; The slope is divided into three grades, namely downhill, relatively flat, and uphill; for the slope : The slope within the range of < -5° is regarded as downhill, the slope within the range of -5° ≤ < 5° is regarded as relatively flat, and the slope within the range of 5° ≤ is regarded as uphill; the conveying speed is divided into three grades, namely deceleration, normal working speed, and acceleration; for the speed : Keeping the speed unchanged is regarded as the normal working speed, increasing the speed is regarded as acceleration, and decreasing the speed is regarded as deceleration; Step 4: Establish a standard fuzzy logic rule base according to the actual on-site operation situation by using the empirical induction method; IF Load = Low load AND Gradient = Uphill THEN Conveyor speed = Increase by 1 gear speed; IF Load = Low load AND Gradient = Relatively flat THEN Conveyor speed = Increase by 2 gear speeds; IF Load = Low load AND Gradient = Downhill THEN Conveyor speed = Decrease by 1 gear speed; IF Load = Medium load AND Gradient = Uphill THEN Conveyor speed = Normal operating speed; IF Load = Medium load AND Gradient = Relatively flat THEN Conveyor speed = Increase by 1 gear speed; IF Load = Medium load AND Gradient = Downhill THEN Conveyor speed = Decrease by 1 gear speed; IF Load = High load AND Gradient = Uphill THEN Conveyor speed = Decrease by 1 gear speed; IF Load = High load AND Gradient = Relatively flat THEN Conveyor speed = Normal operating speed; IF Load = High load AND Gradient = Downhill THEN Conveyor speed = Decrease by 2 gear speeds; Step 5: Expand the fuzzy logic control rules; Introduce state credibility, and divide the state credibility into low, medium, and high according to the size of the diagonal elements of the error covariance matrix to assist in judging the reliability of the monitoring data and adjust the fuzzy control rules; Step 6: Output a speed regulation command to perform intelligent control on the long-distance two-way belt conveyor for mining; Use the centroid method to convert the output of the fuzzy controller into an accurate speed regulation command, and send the speed regulation command to the motor drive control system of the long-distance two-way belt conveyor for mining to achieve precise control of the speed of the long-distance two-way belt conveyor for mining by using the speed regulation command.

2. The intelligent speed regulation method of a long-distance two-way belt conveyor for mine use according to claim 1, characterized in that, In Step 1, the variety of sensors include load sensors, speed sensors, current sensors, gradient sensors, temperature sensors, and torque sensors.

3. The intelligent speed regulation method of a long-distance two-way belt conveyor for mine use according to claim 1, wherein In Step 2, the process of using the distributed Kalman filter architecture for filtering processing is as follows: S21: During the operation of the long-distance two-way belt conveyor for mining, real-time collect monitoring signals through the sensor monitoring network, and use the signal preprocessing module for amplification and preliminary denoising processing; S22: Define the state vector according to formula (1) , establish the state equation based on the motion model of the long-distance two-way belt conveyor for mine use according to formula (2), and establish the observation equation based on the motion model of the long-distance two-way belt conveyor for mine use according to formula (3); （1）； （2）； （3）； In the formula, is the speed of the conveyor belt at time is the slope of the conveyor belt at time is the load of the conveyor belt at time is the transpose operation; is the state transition matrix based on the motion model of the long-distance two-way belt conveyor for mine use, , is the sampling time interval, is the acceleration due to gravity, is the sine value of the slope of the conveyor belt at time is the friction coefficient of the conveyor belt, is the load of the conveyor belt at time is the state at time is the control input matrix, is the equivalent mass of the conveyor belt, is the motor torque control quantity, is the process noise; is the sensor measurement value at time is the observation matrix, , is the calibration coefficient of the load sensor; is the observation noise; is the state at time; S23: Continuously correct the state equation through the observation equation, and use the Kalman gain K K to recursively update the state estimate and covariance matrix, and output high-precision monitoring data.

4. An intelligent speed regulation method for a long-distance two-way belt conveyor for mine use according to claim 1, characterized in that In step five, calculate the state credibility according to formula (4). ; When < 0.3, use an external PID for compensation to multiply the output of the fuzzy controller by a decay coefficient of 0.

5. When 0.3 ≤ < 0.6, use the standard fuzzy rule base for control. When 0.6 ≤ , enable the enhanced fuzzy rule base to reduce the response speed; （4）； In the formula, is the trace of the error covariance matrix, the sum of the diagonal elements, is the error covariance matrix of the Kalman filter, is the set maximum acceptable error matrix.

5. The intelligent speed regulation method of a long-distance two-way belt conveyor for mine use according to claim 2, characterized in that, In Step 1, the load sensor is installed on the idler support in the middle section of the long-distance bidirectional belt conveyor for underground mining, and is used to collect the actual load signal in real time; the speed sensor is installed at the driving wheel of the long-distance bidirectional belt conveyor for underground mining, and is used to collect the conveying speed signal in real time; the current sensor is installed on the driving motor of the long-distance bidirectional belt conveyor for underground mining, and is used to collect the current signal in real time; multiple slope sensors are installed on the frame of the long-distance bidirectional belt conveyor for underground mining at intervals of a set distance in sequence, and are used to collect the belt slope signal of the area where they are located in real time; the temperature sensor is installed on the driving motor and controller of the long-distance bidirectional belt conveyor for underground mining, and is used to collect the temperature signal in real time; the torque sensor is installed on the driving motor of the long-distance bidirectional belt conveyor for underground mining, and is used to collect the torque signal in real time.

6. The intelligent speed regulation method of a long-distance two-way belt conveyor for mine use according to claim 4, characterized in that In Step 5, the construction process of the enhanced fuzzy rule base is as follows: S51: Perform enhanced fuzzification processing on the key parameters in the high-precision monitoring data; The load is divided into five levels, namely low load, slightly low load, medium load, slightly high load, and high load; the slope is divided into five levels, namely steep downhill, downhill, flat, uphill, and steep uphill; S52: Establish an enhanced fuzzy logic rule base using the empirical induction method according to the actual on-site operation conditions; IF load = low load AND slope = steep uphill THEN conveying speed = increase by 2 gear speeds; IF load = low load AND slope = uphill THEN conveying speed = increase by 3 gear speeds; IF load = low load AND slope = flat THEN conveying speed = increase by 4 gear speeds; IF load = low load AND slope = downhill THEN conveying speed = increase by 2 gear speeds; IF load = low load AND slope = steep downhill THEN conveying speed = normal operating speed; IF load = slightly low load AND slope = steep uphill THEN conveying speed = normal operating speed; IF load = slightly low load AND slope = uphill THEN conveying speed = increase by 1 gear speed; IF load = slightly low load AND slope = flat THEN conveying speed = increase by 2 gear speeds; IF load = slightly low load AND slope = downhill THEN conveying speed = increase by 1 gear speed; IF load = slightly low load AND slope = steep downhill THEN conveying speed = decrease by 1 gear speed; IF load = medium load AND slope = steep uphill THEN conveying speed = decrease by 2 gear speeds; IF load = medium load AND slope = uphill THEN conveying speed = normal operating speed; IF Load = Medium Load AND Gradient = Flat THEN Conveyor Speed = Normal Operating Speed; IF Load = Medium Load AND Gradient = Downhill THEN Conveyor Speed = Normal Operating Speed; IF Load = Medium Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduce by 1 Gear Speed; IF Load = Slightly Higher Load AND Gradient = Steep Uphill THEN Conveyor Speed = Reduce by 3 Gear Speeds; IF Load = Slightly Higher Load AND Gradient = Uphill THEN Conveyor Speed = Reduce by 2 Gear Speeds; IF Load = Slightly Higher Load AND Gradient = Flat THEN Conveyor Speed = Reduce by 1 Gear Speed; IF Load = Slightly Higher Load AND Gradient = Downhill THEN Conveyor Speed = Normal Operating Speed; IF Load = Slightly Higher Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduce by 2 Gear Speeds; IF Load = High Load AND Gradient = Steep Uphill THEN Conveyor Speed = Reduce by 4 Gear Speeds; IF Load = High Load AND Gradient = Uphill THEN Conveyor Speed = Reduce by 3 Gear Speeds; IF Load = High Load AND Gradient = Flat THEN Conveyor Speed = Reduce by 2 Gear Speeds; IF Load = High Load AND Gradient = Downhill THEN Conveyor Speed = Reduce by 3 Gear Speeds; IF Load = High Load AND Gradient = Steep Downhill THEN Conveyor Speed = Reduce by 4 Gear Speeds.

7. An intelligent speed regulation method for a long-distance two-way belt conveyor for mine use according to claim 6, characterized in that, In S51, the specific process of enhancing the fuzzification process is as follows: For the load : The load within the range of 0 ≤ < 150 kg / m is regarded as a low load, the load within the range of 150 kg / m ≤ < 225 kg / m is regarded as a slightly low load, the load within the range of 225 kg / m ≤ < 300 kg / m is regarded as a medium load, the load within the range of 300 kg / m ≤ < 350 kg / m is regarded as a slightly high load, the load within the range of 350 kg / m ≤ ≤ 600 kg / m is regarded as a high load; For the slope : Consider the slope within the range of < -8° as a steep downhill slope, the slope within the range of -8° ≤ < -4° as a downhill slope, the slope within the range of -4° ≤ < 4° as a flat slope, the slope within the range of 4° ≤ < 8° as an uphill slope, and the slope within the range of 8° ≤ as a steep uphill slope.

Images