Intelligent control method of train gondola car spiral unloading machine

Abstract

The invention discloses an intelligent control method for a train gondola car spiral unloading machine, and relates to the technical field of automatic control. The problems that due to manual operation of a train gondola car spiral unloading machine, spiral head position control is not accurate, load dynamic unbalance is caused, and intelligence is lacked are solved. Comprising the steps of vehicle type recognition, accurate positioning, height control, load adjustment, end collision prevention and automatic discharging. The system is suitable for bulk cargo unloading scenes of railway freight stations and large factory and mine enterprises, and has remarkable advantages in promotion of intelligent logistics and unmanned port construction.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automation control technology, and is a kind of intelligent control method of train open wagon spiral unloading machine. BACKGROUND

[0002] The existing train open wagon spiral unloading machine relies on manual visual operation. In the harsh working conditions of dust diffusion and uncertain light, the spiral head position control is inaccurate due to the inability to accurately identify the diversified car models and real-time perception of material property changes, resulting in frequent collision with the compartment wall or residual discharge. At the same time, the lack of load regulation leads to motor overload and burning or underload idling. Moreover, due to the lack of data interface and intelligent decision-making layer, the equipment becomes an information island, which seriously restricts the construction process of smart factory. The existing technology has defects such as poor anti-interference and response lag, and has not been able to break through the systematic shackles of relying on manual experience. SUMMARY

[0003] To solve the problems of spiral head position control inaccuracy, load dynamic imbalance and lack of intelligence caused by manual operation of the train open wagon spiral unloading machine, the present application provides an intelligent control method of train open wagon spiral unloading machine, comprising:

[0004] S1. Car model identification, the high-definition camera collects the side identification image of the open wagon car, identifies the car model based on machine vision, and retrieves the pre-stored car geometric parameters;

[0005] S2. Relative position coordinate calculation, the first and second car detection sensors are symmetrically installed to detect the upper edge position of the car side, and the device displacement data collected by the walking encoder in real time is combined with the car geometric parameters and device displacement data retrieved in S1 to calculate the position coordinates of the spiral head relative to the open wagon car;

[0006] S3. Spiral head height dynamic control, the first and second rotary encoders are used to monitor the rotation angle of the lifting sprocket in real time, the lifting height of the spiral arm is calculated based on the chain wheel circumference, and the spiral head height is controlled to keep a predetermined gap between the spiral head and the car bottom;

[0007] S4. Load adaptive adjustment, the current transformer is used to collect the working current of the spiral head motor in real time. When the current exceeds the upper limit of the rated threshold, the device walking speed is reduced to reduce the load. When the current is lower than the lower limit of the rated threshold, the walking speed is increased to increase the load;

[0008] S5. Car end anti-collision control, when the detection screw rod of the car end detection device touches the car end and triggers the proximity switch, the device walking is forced to stop;

[0009] S6. Automatic unloading operation is performed, based on the position coordinates of S2, the height of the screw head of S3 and the load adjustment result of S4, the screw unloading machine is controlled to complete the unloading operation according to the preset path.

[0010] Further, in S4, the load self-adaptive adjustment is specifically realized by the following algorithm:

[0011] When I 实时 ≥ I 基准 + I 偏移

[0012] Output frequency = gamma x (-I 偏移 ) + delta

[0013] When I 实时 ≤ I 基准 -I 偏移

[0014] Output frequency = gamma x (I 偏移 ) + delta

[0015] When I 基准 -I 偏移 ≤ I 实时 ≤ I 基准 + I 偏移

[0016] Output frequency = gamma x (I 基准 -I 实时 ) + delta

[0017] Wherein, the reference working current I 基准 is the reference working current, which is 0.8In; I 偏移 is the offset current, which is obtained by subtracting the reference working current from the working current of the screw head motor under maximum load; I 实时 is the real-time working current of the screw head motor; gamma is the gain coefficient, which is the amplification multiple determined according to the actual working state of the screw head; delta is the speed constant, which is the frequency corresponding to the walking speed of the frequency converter according to the actual working state.

[0018] The beneficial effects of the present application: the present application realizes safe and efficient unloading operation with no human intervention in the whole process through precise identification of vehicle type by machine vision, precise positioning constructed by ultrasonic sensor and walking encoder linkage, dynamic speed regulation of current transmitter, and instantaneous anti-collision of buffer spring triggering proximity switch, completely solves the industry pain points of high accident rate of manual operation, equipment overload shutdown and inability to connect intelligent factory. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flow chart of the intelligent control method of the train open car screw unloading machine;

[0020] Figure 2 It is a whole structure diagram of a spiral unloading machine for a train open wagon.

[0021] Figure 3 It is a front view of a spiral unloading machine for a train open wagon.

[0022] Figure 4 It is a structure diagram of a car end detection device.

[0023] Figure 5 It is a diagram of motor output frequency changing with real-time current;

[0024] In the figure, 1 is a first rotary encoder, 2 is a second rotary encoder, 3 is a walking encoder, 4 is a high-definition camera, 5 is a car end detection device, 6 is a first car detection sensor, 7 is a second car detection sensor, 8 is a proximity switch, 9 is a buffer spring, 10 is a device shell, 11 is an adjusting nut, and 12 is a detection lead screw. DETAILED DESCRIPTION

[0025] The technical solutions of the present application will be further described below in combination with examples, but are not limited thereto, any modification or equivalent replacement to the technical solutions of the present application without departing from the spirit and scope of the present application shall be covered in the protection scope of the present application. The process equipment or device not specifically mentioned in the following examples all adopts the conventional equipment or device in the field, and the raw materials used in the examples of the present application are all commercially available if not specifically mentioned; and the technical means used in the examples of the present application are all conventional means familiar to those skilled in the art if not specifically mentioned.

[0026] Example 1, in combination Figure 1 It is illustrated that the intelligent control method of a spiral unloading machine for a train open wagon comprises the following steps:

[0027] S1. Car model identification, the high-definition camera 4 collects the side identification image of the open wagon car, identifies the car model based on machine vision, and calls the pre-stored car geometric parameters;

[0028] S2. Relative position coordinate calculation, the first car detection sensor 6 and the second car detection sensor 7 symmetrically installed are used to detect the upper edge position of the car side, combined with the real-time equipment displacement data collected by the walking encoder 3, the position coordinates of the spiral head relative to the open wagon car are calculated according to the car geometric parameters and the equipment displacement data called in S1;

[0029] S3. Spiral head height dynamic control, the first rotary encoder 1 and the second rotary encoder 2 are used to monitor the rotation angle of the lifting sprocket in real time, the lifting height of the spiral arm is calculated combined with the circumference of the sprocket, and then the height of the spiral head is controlled, so that the spiral head always maintains a preset gap with the car bottom;

[0030] S4. Load self-adaptive adjustment, real-time acquisition of screw head motor working current through current transducer, when the current exceeds the upper limit of rated threshold, reduce the walking speed of the device to reduce the load; when the current is lower than the lower limit of rated threshold, increase the walking speed to increase the load;

[0031] S5. Carriage end anti-collision control, when the detection wire rod of the carriage end detection device 5 touches the carriage end and triggers the proximity switch, the device is forced to stop walking;

[0032] S6. Automatic unloading operation execution, based on the position coordinates of S2, the screw head height of S3 and the load adjustment result of S4, the screw unloader is controlled to complete the unloading operation according to the preset path.

[0033] Specifically, in combination with Figure 1 And Figure 2 It can be seen that the walking encoder is connected with the center of the driven wheel through the elastic coupling, that is, the encoder also synchronously rotates one circle when the driven wheel rotates one circle, and the encoder has 600 pulses per circle. The PLC can calculate the rotation angle of the walking wheel by calculating the number of received pulses, and can calculate the walking distance of the device under the condition of knowing the circumference of the walking wheel. Moreover, the synchronization of the encoder and the driven wheel can effectively avoid the situation that the encoder calculation is wrong due to the slipping of the driving wheel of the device, thereby ensuring the accuracy of the calculation of the walking distance. At the same time, when the driving wheel slips due to the sudden increase of the unloading resistance, the driven wheel does not rotate, thereby ensuring the accuracy of the position data of the walking encoder connected with the driven wheel. If the walking encoder is installed on the driving wheel, when the driving wheel of the device slips, the real position of the device body does not move, but the walking encoder connected with the driving wheel rotates with the driving wheel, thereby causing the position data error of the device and unable to work normally.

[0034] The first carriage detection sensor 6 and the second carriage detection sensor 7 are installed at the intersection of the horizontal projection of the center line of the two screw head horizontal symmetry axes and the upper edge of the train carriage side, which is 1200 mm away from the upper edge of the carriage. The first carriage detection sensor 6 and the second carriage detection sensor 7 detect the upper edge of the carriage and send an action signal to the PLC. When the first carriage detection sensor 6 and the second carriage detection sensor 7 are at the connection position of the two carriages, the signal of the upper edge of the carriage disappears, and the sensor sends an action disappearance signal to the PLC. In this way, the relative position origin of the device and the carriage can be determined, and the relative position of the device and the carriage can be calculated through the information of the walking encoder.

[0035] The rotary encoder is directly connected to the through shaft of the lifting height limiter, and is connected to the center of the lifting sprocket, that is, the lifting sprocket rotates one circle, and the encoder also rotates one circle synchronously, and the encoder has 600 pulses per circle, so that the PLC can calculate the rotation angle of the lifting sprocket by calculating the number of received pulses, and the height of the spiral arm can be calculated under the condition of knowing the circumference of the lifting sprocket.

[0036] As shown in Figure 4 The car end detection device 5 is composed of a device shell, a detection screw rod, a buffer spring, an adjusting nut and a proximity switch. When the spiral head runs to the end of the car, the detection screw rod first touches the end of the car and is compressed to the inside of the device under the buffer of the spring, so that the proximity switch detects the signal and transmits it to the PLC, so that the control equipment stops walking, to solve the problem that the manual operation spiral unloading machine often overshoots the car wall or does not reach the position.

[0037] In S4, the load adaptive adjustment is specifically realized by the following algorithm:

[0038] When I 实时 ≥I 基准 +I 偏移

[0039] Output frequency = γ × (-I 偏移 ) + δ

[0040] When I 实时 ≤I 基准 -I 偏移

[0041] Output frequency = γ × (I 偏移 ) + δ

[0042] When I 基准 -I 偏移 ≤I 实时 ≤I 基准 +I 偏移

[0043] Output frequency = γ × (I 基准 -I 实时 ) + δ

[0044] Wherein, the reference working current I 基准 is the reference working current, which is 0.8In; I 偏移 is the offset current, which is obtained by subtracting the reference working current from the working current of the spiral head motor under maximum load; I 实时 is the real-time working current of the spiral head motor; γ is the gain coefficient, which is the amplification multiple determined according to the actual working state of the spiral head; δ is the speed constant, which is the frequency corresponding to the walking speed of the frequency converter according to the actual working state.

[0045] For example, the common spiral motor usually uses 22kW motor, the rated current is about 40A, the reference working current is 0.8x40=32A, and the offset current is 40-32=8A.

[0046] When I 实时 ≥I 基准 +I 偏移

[0047] I 实时 ≥32+8

[0048] I 实时 ≥40

[0049] Output frequency=γ × (-I 偏移 ) + δ

[0050] According to the actual situation, γ=2.5; δ=25

[0051] Output frequency=2.5 × (-8) + 25

[0052] Output frequency=5Hz

[0053] When I 实时 ≤I 基准 -I 偏移

[0054] I 实时 ≤24

[0055] Output frequency=2.5 × (8) + 25

[0056] Output frequency=45Hz

[0057] When I 基准 -I 偏移 ≤I 实时 ≤I 基准 +I 偏移

[0058] I 实时 24A≤I real time ≤40A

[0059] Output frequency=2.5 × (32-I 实时 ) + 25

[0060] At this time, the output frequency changes linearly with the real-time current, as shown in Figure 5

[0061] Example 2, illustrating the automatic unloading process of spiral unloading machine:

[0062] The spiral unloading machine is in the initial working position, the "working mode" knob on the driver's room operation panel is switched from "manual" to "automatic", and the "unloading direction" knob is switched to "left to right" or "right to left" according to the actual needs on the spot. The following is an example of "right to left":

[0063] 1. The driver presses the "one-key start" button, and the equipment starts to automatically walk to the left to find the car, when the first car detection sensor 6 and the second car detection sensor 7 detect the car, the first car detection sensor and the second car detection sensor 7 send a signal to the PLC, at this time the PLC starts to calculate the real-time position of the two spiral head center points relative to the train car (calculated by collecting the number of walking encoder pulses, at this time the equipment displacement relative to the car is 0mm, the equipment displacement increases when walking to the left, and the equipment displacement decreases when walking to the right).

[0064] 2. When the equipment displacement relative to the car is 7300mm (i.e. the two spiral heads move to the middle door of the car), the equipment stops walking and starts the two spiral heads, at the same time the two spiral heads start to descend synchronously, descend to 1250mm (the spiral head descends to the lower limit, the coordinate is 0mm, at this time the spiral head is 50mm away from the car bottom - reserve safety distance) Wait 5s, continue to descend to 600mm and wait 5s, continue to descend to 0mm and wait 6s.

[0065] 3. The right spiral head is lifted to 900mm, and the left spiral head is lifted to 450mm.

[0066] 4. The equipment walks to the right, when the equipment displacement relative to the car is 3000mm (i.e. the right spiral head moves to 500mm away from the car end), the two spiral heads start to descend synchronously, descend to 0mm and wait for 5s.

[0067] 5. The equipment walks to the right, when the detection screw rod of the right car end detection device touches the car end and triggers the proximity switch, the equipment stops walking and waits for 3s.

[0068] 6. The equipment walks to the left, when the equipment displacement relative to the car is 7300mm, the equipment stops walking, the left spiral head is lifted to 900mm, and the right spiral head is lifted to 450mm.

[0069] 7. The equipment walks to the left, when the equipment displacement relative to the car is 11000mm (i.e. the left spiral head moves to 500mm away from the car end), the two spiral heads start to descend synchronously, descend to 0mm and wait for 5s.

[0070] 8. The equipment walks to the left, when the detection screw rod of the left car end detection device touches the car end and triggers the proximity switch, the equipment stops walking and waits for 3s.

[0071] 9. The device walks to the right, when the device is displaced 11000mm relative to the car, both screw heads stop working, and at the same time, rise to the upper limit.

[0072] 10. The device walks to the left, the car detection sensor I and the car detection sensor II detect the train car, and the real-time position of the center points of both screw heads relative to the train car is calculated again, and the above steps are repeated.

Claims

1. An intelligent control method for a spiral unloading machine for a train open car, characterized in that: include: S1. Carriage model recognition: using a high-definition camera (4) to capture the side logo image of the open car, identify the car model based on machine vision, and retrieve pre-stored car geometry parameters; S2. Calculate the relative position coordinates by using the symmetrically installed first carriage detection sensor (6) and the second carriage detection sensor (7) to detect the upper edge position of the carriage side, and calculate the position coordinates of the spiral head relative to the open car carriage based on the carriage geometric parameters and equipment displacement data retrieved in S1 in combination with the equipment displacement data collected in real time by the travel encoder (3); S3. Dynamic control of the height of the spiral head, using the first rotary encoder (1) and the second rotary encoder (2) to monitor the rotation angle of the lifting sprocket in real time, and calculating the lifting height of the spiral arm in combination with the circumference of the sprocket, thereby controlling the height of the spiral head so that the spiral head always maintains a preset gap with the bottom of the carriage; S4. Load adaptive adjustment: The current transmitter collects the operating current of the screw head motor in real time. When the current exceeds the upper rated threshold, the equipment's travel speed is reduced to reduce the load; when the current falls below the lower rated threshold, the travel speed is increased to increase the load. S5. Carriage end anti-collision control, when the detection screw of the carriage end detection device (5) touches the carriage end and triggers the proximity switch, the equipment is forced to stop moving; S6. Automatic unloading operation is executed. Based on the position coordinates of S2, the screw head height of S3 and the load adjustment result of S4, the screw unloader is controlled to complete the unloading operation according to the preset path.

2. The intelligent control method for a train open car spiral unloading machine according to claim 1, characterized in that: In S4, load adaptive regulation is implemented through the following algorithm: This I 实时 ≥I 基准 +I 偏移 Output frequency = γ × (-I 偏移 ) + δ This I 实时 ≤I 基准 -I 偏移 Output frequency = γ × (I 偏移 ) + δ This I 基准 -I 偏移 ≤I 实时 ≤I 基准 +I 偏移 Output frequency = γ × (I 基准 -I 实时 ) + δ Among them, the reference operating current I 基准 is the reference working current, which is 0.8In; I 偏移 is the offset current, which is obtained by subtracting the reference working current from the working current of the spiral head motor at maximum load; I 实时 is the real-time working current of the spiral head motor; γ is the gain coefficient, which is the magnification determined according to the actual working state of the spiral head; δ is the speed constant, which is the frequency corresponding to the inverter walking speed determined according to the actual working state.

Images