Pneumatic conveying method and slag conveying device based on slag state recognition

Abstract

This invention relates to a pneumatic conveying method and device based on the identification of slag conditions. The method utilizes negative pressure generated by a blower to pump and discharge slag from a vertical shaft. The pneumatic conveying method includes the following steps: classifying different types of slag and establishing a database; collecting image information of slag at preset locations within the shaft; matching slag parameters corresponding to the image information in the database; determining the slag weight based on the image information and the slag parameters; and adjusting the frequency of the blower according to the slag weight to match the corresponding wind speed. This invention can identify the slag condition, provide feedback on different slag condition information, and adjust the power required for conveying the slag using frequency conversion based on the feedback information, effectively reducing energy waste while ensuring smooth slag output.

Description

Technical Field

[0001] This invention relates to the field of vertical shaft construction technology, and more particularly to a pneumatic conveying method and a soil conveying device based on soil condition identification. Background Technology

[0002] Currently, dry muck removal by vertical shaft tunneling machines (such as using grab buckets, scrapers, screw conveyors, etc.) has disadvantages such as incomplete muck removal, heavy load on the front end of the tunneling machine, and large space occupation, thus limiting the structural layout and directional adjustment functions of vertical shaft construction machinery. Wet muck removal, on the other hand, is difficult to implement in special geological conditions or water-scarce areas. Therefore, vacuum negative pressure muck removal is often used for vertical shaft muck removal. A muck conveying pipeline is set between the inside of the shaft and a preset muck removal location outside the shaft. A suction fan is installed on the muck conveying pipeline to create negative pressure, thereby drawing the muck into the pipeline for external discharge. This structure has advantages such as simple structure, small size, flexible process layout, and easy automation. However, for engineering efficiency considerations (to achieve better muck removal efficiency), the muck removal system is often set to the maximum operating frequency during actual muck removal. Although the preset muck removal effect can be achieved, the frequency cannot be adjusted according to different muck conditions during actual muck removal, resulting in a large amount of energy waste.

[0003] There is currently no effective solution to the problem that the negative pressure suction slag removal method in related technologies cannot adjust the operating frequency according to different slag conditions, resulting in energy waste.

[0004] Therefore, based on years of experience and practice in related industries, the inventor proposes a pneumatic conveying method and a waste conveying device based on the identification of waste soil conditions, in order to overcome the shortcomings of the prior art. Summary of the Invention

[0005] The purpose of this invention is to provide a pneumatic conveying method and a pneumatic conveying device based on the identification of the state of construction waste. This method can identify the state of construction waste, provide feedback on different state information, and adjust the power required for conveying construction waste according to the feedback information. This effectively reduces energy waste while ensuring smooth output of construction waste.

[0006] The objective of this invention can be achieved through the following methods:

[0007] This invention provides a pneumatic conveying method based on the state recognition of construction waste, which uses negative pressure generated by a blower to pump and discharge construction waste from a vertical shaft. The pneumatic conveying method includes the following steps:

[0008] Different types of slag and stone are classified and a database is established;

[0009] Collect image information of slag and rock at preset locations inside the vertical shaft;

[0010] Based on the slag image information, match the slag parameters corresponding to the slag image information in the database;

[0011] The weight of the slag is determined based on the slag image information and the slag parameters;

[0012] The frequency of the blower is adjusted according to the weight of the slag and stone to match the corresponding wind speed.

[0013] In a preferred embodiment of the present invention, the step of classifying different types of slag and establishing a database includes:

[0014] Images of different types of slag were extracted and a dataset was created.

[0015] The images of the different types of slag are repositioned to expand the dataset;

[0016] A convolutional neural network model is built based on the data in the dataset, and the database includes the convolutional neural network model.

[0017] In a preferred embodiment of the present invention, the step of establishing a convolutional neural network model based on the data in the dataset includes:

[0018] The data in the dataset are normalized.

[0019] Define neural network parameter variables;

[0020] Set corresponding convolution kernels for different types of slag;

[0021] At least one convolutional layer is added for each of the different convolutional kernels to construct feature maps.

[0022] Based on the original convolutional layer, multiple convolutional layers are created for different convolutional kernels, and the input data of each convolutional layer is set as the output data of the previous convolutional layer.

[0023] The convolutional neural network model is formed by training the convolutional neural network using the set input data and output data. The input data includes images of different types of slag and the output data includes the category of slag.

[0024] A recognition script file is created based on the convolutional neural network model, and the database includes the recognition script file.

[0025] In a preferred embodiment of the present invention, the neural network parameter variables include the iteration period of images of different types of slag and / or the number of images of different types of slag processed.

[0026] In a preferred embodiment of the present invention, setting corresponding convolution kernels for different types of slag includes: setting the size of the convolution kernel.

[0027] In a preferred embodiment of the present invention, the images of different types of slag include collected pictures of different types of slag.

[0028] The image with a fixed width and height is fed into the convolutional layer to construct feature maps with different convolutional kernels.

[0029] In a preferred embodiment of the invention, the position adjustment includes rotation, horizontal translation, and / or vertical translation.

[0030] In a preferred embodiment of the present invention, the acquisition of slag and rock image information at a preset location within the shaft includes:

[0031] Images of slag and rock at preset locations within the vertical shaft are collected from different angles.

[0032] In a preferred embodiment of the present invention, the step of matching slag parameters corresponding to the slag image information in the database according to the slag image information includes:

[0033] Calculate at least the length, width, and height of the slag based on the slag image information collected from different angles;

[0034] The length, width, and height values ​​of the slag obtained from different angles were compared respectively;

[0035] The maximum length, maximum width, and maximum height values ​​of the slag obtained from different angles are extracted to form a cubic envelope of the slag.

[0036] Calculate the volume of the cubic envelope of the slag.

[0037] In a preferred embodiment of the present invention, determining the weight of the slag based on the slag image information and the slag parameters includes:

[0038] The weight of the slag is obtained based on the type, density, and volume of the cubic envelope of the slag.

[0039] In a preferred embodiment of the present invention, adjusting the frequency of the blower according to the weight of the slag and stone to match the corresponding wind speed includes:

[0040] Establish a correspondence between the weight of the slag and the frequency of the blower, wherein the blower can pump and discharge the slag of the corresponding weight at the corresponding frequency;

[0041] The frequency of the blower is obtained based on the weight of the slag and stone.

[0042] The frequency of the blower is adjusted so that the frequency of the blower and the weight of the slag meet the specified correspondence.

[0043] This invention provides a slag conveying device, which uses negative pressure suction to pump and discharge slag from a vertical shaft. The slag conveying device includes:

[0044] A pre-storage unit is used to classify different types of slag and establish a database;

[0045] The image acquisition unit is used to acquire image information of slag and rock at preset locations inside the shaft;

[0046] The proportioning unit is used to match the slag parameters corresponding to the slag image information in the database according to the slag image information.

[0047] The parameter determination unit is used to determine the weight of the slag based on the slag image information and the slag parameters;

[0048] The adjustment unit is used to adjust the frequency of the blower used for pumping the slag according to the weight of the slag, so as to match the corresponding wind speed.

[0049] In a preferred embodiment of the present invention, the image acquisition unit includes a plurality of image acquisition devices, which are disposed on a vertical shaft excavation equipment and can move along the depth direction with the vertical shaft excavation equipment.

[0050] In a preferred embodiment of the present invention, the slag conveying device includes a suction nozzle, a slag conveying pipeline, a blower, and a slag storage device. The suction nozzle is disposed on or near the cutter head of the vertical shaft excavation equipment. The suction nozzle is connected to one end of the slag conveying pipeline, and the other end of the slag conveying pipeline is connected to the slag storage device. The blower is disposed on the slag conveying pipeline or on the slag storage device.

[0051] In a preferred embodiment of the present invention, a dust removal device is provided upstream of the blower along the conveying direction of the slag.

[0052] As described above, the features and advantages of the pneumatic conveying method and slag conveying device based on slag state recognition of the present invention are as follows: Before pumping and discharging slag from the shaft, slag of different weights is collected and classified in advance. During the actual excavation process, image information of slag at a preset location in the shaft is collected. The corresponding slag parameters are matched in the database based on the slag image information. The weight of slag at the preset location in the shaft can be determined based on the collected slag image information and the corresponding slag parameters. The frequency of the blower used to pump the slag is adjusted according to the feedback slag weight to match the corresponding wind speed, ensuring that the slag can be smoothly pumped out by the blower while maintaining a suitable frequency, effectively reducing energy waste. Attached Figure Description

[0053] The accompanying drawings are intended only to illustrate and explain the present invention and do not limit the scope of the invention.

[0054] in:

[0055] Figure 1 This is one of the flowcharts for the pneumatic conveying method based on the state recognition of slag and soil according to the present invention.

[0056] Figure 2 This is the second flowchart of the pneumatic conveying method based on the state recognition of slag and soil according to the present invention.

[0057] Figure 3 This is the third flowchart of the pneumatic conveying method based on the state recognition of slag in this invention.

[0058] Figure 4 This is the fourth flowchart of the pneumatic conveying method based on the state recognition of slag in this invention.

[0059] Figure 5 This is one of the structural block diagrams of the slag conveying device of the present invention.

[0060] Figure 6 This is the second structural block diagram of the slag conveying device of the present invention.

[0061] Figure 7 This is the third structural block diagram of the slag conveying device of the present invention.

[0062] Figure 8 This is the fourth structural block diagram of the slag conveying device of the present invention.

[0063] Figure 9 This is a schematic diagram showing the distribution of multiple image acquisition devices in the slag conveying device of the present invention.

[0064] Figure 10 : This is a connection structure diagram of the slag conveying device of the present invention.

[0065] The reference numerals in the accompanying drawings of this invention are:

[0066] 1. Fan; 2. Shaft;

[0067] 3. Image acquisition device; 4. Suction nozzle;

[0068] 5. Slag conveying pipeline; 6. Slag storage device;

[0069] 7. Dust removal equipment; 701. Cyclone dust collector;

[0070] 702. Baghouse dust collector; 8. Control device;

[0071] 9. Frequency converter;

[0072] 10. Pre-storage unit; 11. Image acquisition module;

[0073] 12. Data augmentation module; 13. Convolutional neural network model building module;

[0074] 20. Image acquisition unit;

[0075] 30. Proportioning unit; 31. Parameter acquisition module;

[0076] 32. Parameter comparison module; 33. Envelope formation module;

[0077] 34. Volume acquisition module;

[0078] 40. Parameter determination unit;

[0079] 50. Adjustment unit; 51. Correspondence establishment module;

[0080] 52. Fan frequency acquisition module; 53. Fan frequency adjustment module. Detailed Implementation

[0081] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described with reference to the accompanying drawings.

[0082] Implementation Method 1

[0083] like Figure 1 As shown, this invention provides a pneumatic conveying method based on the state recognition of construction waste. This method uses negative pressure generated by a blower 1 to pump and discharge construction waste from a vertical shaft 2. The pneumatic conveying method based on the state recognition of construction waste includes the following steps:

[0084] Step S1: Classify different types of slag and establish a database;

[0085] In an optional embodiment of the present invention, such as Figure 2 As shown, step S1 includes:

[0086] Step S101: Extract images of different types of slag and build a dataset;

[0087] Images of different types of slag can be obtained by collecting photos of different types of slag, or by taking photos of different types of slag encountered during the excavation of shaft 2. In this invention, the specific method of obtaining images of different types of slag is not limited; it is sufficient to collect as many types of slag as possible during the excavation of shaft 2.

[0088] Step S102: Adjust the positions of images of different types of slag to expand the dataset;

[0089] Specifically, adjusting the position of images of different types of slag includes rotating, translating horizontally, and / or translating vertically the collected images of different types of slag, and acquiring the image data when the images are rotated, translated horizontally, and / or translated vertically, and collecting them in the dataset. This increases the amount of image data of different types of slag in the dataset, which helps improve the accuracy of the parameters corresponding to different types of slag. For subsequent development of convolutional neural network models, the expansion of the dataset can improve training accuracy.

[0090] Step S103: Establish a convolutional neural network model based on the image data of different types of slag collected in the dataset, wherein the database includes the convolutional neural network model.

[0091] In this embodiment, as Figure 3 As shown, step S103 further includes:

[0092] Step S1031: Normalize the data in the dataset to improve the accuracy of data processing;

[0093] Step S1032: Define neural network parameter variables; wherein, the neural network parameter variables include the iteration period of images of different types of slag and / or the number of images of different types of slag processed.

[0094] Step S1033: Set corresponding convolution kernels for different types of slag; wherein, setting corresponding convolution kernels for different types of slag includes setting the size of the convolution kernel, and the activation function that can be used can be, but is not limited to, the ReLU activation function.

[0095] Step S1034: Add at least one convolutional layer corresponding to different convolutional kernels to construct feature maps with different convolutional kernels; specifically, based on the collected images of different types of slag, images with fixed width and height can be passed into the convolutional layer to construct feature maps with different convolutional kernels.

[0096] Step S1035: Based on the existing convolutional layers, create multiple convolutional layers corresponding to different convolutional kernels, setting the input data of each convolutional layer to the output data of the previous convolutional layer; among these multiple convolutional layers, pooling layers may be included, which perform dimensionality reduction and feature extraction on the output of the convolutional layers. Specifically, pooling layers can reduce the dimensionality of the output feature map by maximizing or averaging the local regions of the convolutional layer output, reducing the number of parameters and improving the model's generalization ability.

[0097] Step S1036: Train the convolutional neural network using the set input and output data to form a convolutional neural network model. The input data includes images of different types of slag and the output data includes the category of slag.

[0098] Step S1037: Create a recognition script file (a file with the .py extension) based on the convolutional neural network model. The database includes the recognition script file, and perform image recognition.

[0099] Between steps S1035 and S1036, an optimizer (such as the ADAM optimizer) can be created to optimize the convolutional neural network model using the cross-entropy cost function.

[0100] In this invention, the method of establishing a convolutional neural network model based on image data used in step S103 (i.e., from step S1031 to step S1037) is a common method for establishing convolutional neural network models in the prior art. It is only applied to the establishment of a convolutional neural network model for slag in this invention. Of course, other methods can also be used to establish a convolutional neural network model in this invention. It is possible to ensure that different types of slag can be classified through the convolutional neural network model, and that the parameters of the corresponding type of slag can be found based on the actual collected slag image data. The specific steps and methods for establishing the convolutional neural network model are not limited in this invention.

[0101] Step S2: Collect image information of slag and rock at a preset location inside shaft 2.

[0102] The image information of slag and rock at preset locations within shaft 2 includes: acquiring images of slag and rock at preset locations within shaft 2 from different angles. During actual acquisition, multiple image acquisition devices 3 can be installed on the shaft 2 excavation equipment. These devices can move along the shaft depth direction with the equipment, thereby acquiring images of slag and rock at preset locations from different angles. Preferably, three image acquisition devices 3 are used, positioned horizontally in the same plane, at the three vertices of a triangle, and distributed at 120° angles on the same circumference. This arrangement of three devices allows for the acquisition of slag and rock images at various locations within the shaft, avoiding blind spots.

[0103] Step S3: Based on the slag image information, match the slag parameters corresponding to the slag image information in the database;

[0104] In an optional embodiment of the present invention, such as Figure 3 As shown, step S3 includes:

[0105] Step S301: Collect slag images from different angles using multiple image acquisition devices 3, and calculate the parameters of the slag corresponding to the slag images from different angles. The parameters of the slag include at least the length, width, and height of the slag. Since the volume of the slag in the shaft 2 needs to be known, the length, width, and height of the slag need to be obtained from the collected slag images. The specific method of obtaining these values ​​can be calculated by comparing the proportion of the slag in the image with the actual slag in the shaft 2.

[0106] In step S301, the parameters of the slag obtained are the parameters of the slag within a certain area of ​​the shaft 2, and the length, width and height values ​​of the slag are also the length, width and height values ​​of the slag contained in the corresponding area of ​​the shaft 2 (not the length, width and height values ​​of a single piece of slag).

[0107] Step S302: Compare the length, width, and height values ​​of the slag obtained from different angles;

[0108] Step S303: Extract the maximum length, maximum width, and maximum height values ​​of the slag obtained from different angles to form a cubic envelope of the slag;

[0109] Step S304: Calculate the volume of the cubic envelope of the slag.

[0110] In this embodiment, the volume of the cubic envelope of the slag is calculated (i.e., the maximum possible volume of the slag is obtained by calculating the maximum length, maximum width and maximum height of the slag), thereby ensuring that the volume of the slag in the area acquired by the multiple image acquisition devices 3 is within the volume range of the cubic envelope, so that the corresponding suction force can be set according to the volume of the cubic envelope, so as to achieve the purpose of transporting all the slag in the area acquired by the multiple image acquisition devices 3 to the outside.

[0111] Step S4: Determine the weight of the slag based on the slag image information and slag parameters;

[0112] In this embodiment, the weight of the slag is obtained based on the type, density, and volume of the cubic envelope of the slag. Since a type of slag matching the image data of the actual slag is found in the database, the density of that type of slag can be determined. Furthermore, since the density and volume of the slag are known, the weight of the slag in the image information collected in shaft 2 can be calculated according to the formula: mass = density × volume (i.e., M = ρV).

[0113] Step S5: Adjust the frequency of blower 1 according to the weight of the slag and stone to match the corresponding wind speed.

[0114] In an optional embodiment of the present invention, such as Figure 4 As shown, step S5 includes:

[0115] Step S501: Establish the correspondence between the weight of slag and the frequency of blower 1. Blower 1 can suck and discharge slag of the corresponding weight at the corresponding frequency. The correspondence between the weight of slag and the frequency of blower 1 includes the weight of slag and the wind speed load spectrum of blower. When the weight of slag is known, the wind speed of blower 1 can be directly obtained through the weight of slag and the wind speed load spectrum of blower.

[0116] In a specific embodiment of the present invention, the load spectrum of slag weight and blower wind speed can be obtained by the following method:

[0117] Step S5011: Place slag of different weights on a pre-set vertical shaft pneumatic conveying simulation test platform, and place an air velocity sensor at the slag.

[0118] Step S5012: By changing the output air volume of blower 1, the air velocity at the slag is recorded by the air velocity sensor during different slag pneumatic conveying processes.

[0119] Step S5013: Multiply the recorded air velocity at the slag and the safety threshold (which can be set by the user) to obtain the target air velocity, thereby forming the load spectrum of slag weight and target air velocity (i.e., the load spectrum of slag weight and blower wind speed).

[0120] Step S502: Since the weight of the slag is calculated, the frequency of the blower 1 required for slag suction and discharge can be obtained from the pre-established correspondence based on the slag weight. Specifically, based on the slag weight information, interpolation can be used, but is not limited to, to obtain the wind speed data corresponding to the slag weight and the blower wind speed load spectrum. The frequency of the blower 1 is then adjusted by the control device 8 according to this wind speed data to adapt to the wind speed. The control device 8 can be, but is not limited to, a host computer.

[0121] Step S503: Adjust the frequency of blower 1 so that the frequency of blower 1 corresponds to the weight of slag and stone, thereby achieving the purpose of suctioning and discharging the slag and stone. Specifically, the control device 8 sends a control signal to blower 1 based on the received data to adjust the frequency of blower 1, thus completing the adjustment of blower 1 in this invention.

[0122] The features and advantages of the pneumatic conveying method based on the state identification of construction waste in this invention are as follows:

[0123] In this pneumatic conveying method based on slag condition recognition, before pumping and discharging the slag in shaft 2, slag of different weights is collected and classified in advance. During the actual excavation process, image information of slag at preset locations in shaft 2 is collected. Based on the slag image information, the corresponding slag parameters are matched in the database. Based on the collected slag image information and the corresponding slag parameters, the weight of the slag at the preset location in shaft 2 can be determined. Based on the feedback slag weight, the frequency of the blower 1 used to pump the slag is adjusted to match the corresponding wind speed. This ensures that the slag can be smoothly pumped and discharged by the blower 1 while maintaining a suitable frequency, without having to run at the maximum frequency all the time, effectively reducing energy waste.

[0124] Implementation Method 2

[0125] like Figure 5 As shown, the present invention provides a slag conveying device, which uses negative pressure suction to pump and discharge slag from the vertical shaft 2. The slag conveying device includes:

[0126] The pre-storage unit 10 is used to classify different types of slag and establish a database;

[0127] Image acquisition unit 20 is used to acquire image information of slag and rock at preset locations within shaft 2;

[0128] The proportioning unit 30 is used to match the slag parameters corresponding to the slag image information in the database based on the slag image information.

[0129] The parameter determination unit 40 is used to determine the weight of the slag based on the slag image information and slag parameters;

[0130] The regulating unit 50 is used to adjust the frequency of the blower 1 used for pumping slag according to the weight of the slag, so as to match the corresponding wind speed.

[0131] In an optional embodiment of the present invention, such as Figure 6 As shown, the pre-storage unit 10 includes an image acquisition module 11, a data expansion module 12, and a convolutional neural network model building module 13. The image acquisition module 11 is used to acquire and extract images of different types of slag and establish a dataset. Images of different types of slag can be acquired by collecting photographs of different types of slag, or by taking photos of different types of slag encountered during the excavation of the shaft 2. The data expansion module 12 is used to adjust the positions of images of different types of slag to expand the dataset. Specifically, adjusting the positions of images of different types of slag includes rotating, translating horizontally, and / or translating vertically the collected images of different types of slag, and acquiring the image data during these processes, which is then collected into the dataset. This increases the amount of image data of different types of slag contained in the dataset, helping to improve the accuracy of the parameters corresponding to different types of slag. The convolutional neural network model building module 13 is used to build a convolutional neural network model based on the data in the dataset, and the database includes the convolutional neural network model.

[0132] In an optional embodiment of the present invention, such as Figure 7 As shown, the proportioning unit 30 includes a parameter acquisition module 31, a parameter comparison module 32, an envelope formation module 33, and a volume acquisition module 34. The parameter acquisition module 31 is used to calculate at least the length, width, and height values ​​of the slag based on slag image information acquired from different angles. The parameters of the slag include at least the length, width, and height values. Since the volume of the slag within the shaft 2 needs to be determined, the length, width, and height values ​​need to be obtained from the acquired slag image information. The specific acquisition method can be derived by calculating the ratio between the slag in the image and the actual slag within the shaft 2. The parameter comparison module 32 is used to compare the length, width, and height values ​​of the slag calculated from different angles. The envelope formation module 33 is used to extract the maximum length, maximum width, and maximum height values ​​of the slag calculated from different angles, forming a cubic envelope of the slag. The volume acquisition module 34 is used to calculate the volume of the cubic envelope of the slag.

[0133] In this embodiment, the volume of the cubic envelope of the slag is calculated (i.e., the maximum possible volume of the slag is obtained by calculating the maximum length, maximum width and maximum height of the slag), thereby ensuring that the volume of the slag in the area acquired by the multiple image acquisition devices 3 is within the volume range of the cubic envelope, so that the corresponding suction force can be set according to the volume of the cubic envelope, so as to achieve the purpose of transporting all the slag in the area acquired by the multiple image acquisition devices 3 to the outside.

[0134] In an optional embodiment of the present invention, such as Figure 8 As shown, the adjustment unit 50 includes a correspondence establishment module 51, a fan frequency acquisition module 52, and a fan frequency adjustment module 53. The correspondence establishment module 51 is used to establish a correspondence between the weight of slag and the frequency of fan 1. Fan 1 can pump and discharge slag of a corresponding weight at the corresponding frequency. The correspondence between the weight of slag and the frequency of fan 1 includes the weight of slag and the fan wind speed load spectrum. Knowing the weight of slag, the corresponding wind speed of fan 1 is directly obtained through the weight of slag and the fan wind speed load spectrum. The fan frequency acquisition module 52 is used to obtain the corresponding frequency of fan 1 based on the weight of slag. Specifically, based on the slag weight information, the wind speed data corresponding to the weight of slag and the fan wind speed load spectrum can be obtained using, but is not limited to, interpolation methods. The control device 8 adjusts the frequency of fan 1 according to the wind speed data to achieve wind speed adaptation. The fan frequency adjustment module 53 is used to adjust the frequency of fan 1 so that the frequency of fan 1 and the weight of slag satisfy the aforementioned correspondence.

[0135] In an optional embodiment of the present invention, such as Figure 9 , Figure 10 As shown, the image acquisition unit 20 includes multiple image acquisition devices 3, which are installed on the shaft excavation equipment. The multiple image acquisition devices 3 can move along the shaft depth direction with the shaft excavation equipment, thereby acquiring image information of slag and rock at preset positions from different angles through the multiple image acquisition devices 3.

[0136] Specifically, such as Figure 10 As shown, there are three image acquisition devices 3. These three devices are positioned horizontally in the same plane, at the three vertices of a triangle, and distributed at 120° angles on the same circle. This arrangement of three devices allows for the acquisition of images of the slag and rock at various locations within the shaft, avoiding blind spots. Each image acquisition device 3 includes an industrial camera, with its image acquisition end facing the slag and rock at the bottom of the shaft 2.

[0137] In an optional embodiment of the invention, such as Figure 10As shown, the slag conveying device includes a suction nozzle 4, a slag conveying pipeline 5, a blower 1, and a slag storage device 6. The suction nozzle 4 is located on or near the cutter head of the vertical shaft excavation equipment. The suction nozzle 4 moves with the cutter head of the vertical shaft excavation equipment and can suck up the cut slag. One end of the suction nozzle 4 is connected to the slag conveying pipeline 5, and the other end of the slag conveying pipeline is connected to the slag storage device 6. The blower is located on the slag conveying pipeline or on the slag storage device.

[0138] Furthermore, the slag conveying pipeline 5 includes at least a telescopic pipeline located near the suction nozzle 4. The telescopic pipeline is directly connected to the outlet of the suction nozzle 4. By setting the telescopic pipeline, the position of the suction nozzle 4 can be adjusted within a certain range, thereby meeting the requirements for suction and discharge of different external slag stones in the vertical shaft 2.

[0139] Furthermore, such as Figure 10 As shown, to meet the slag storage requirements, multiple slag storage devices 6 can be installed in parallel. These devices can store and discharge slag simultaneously, or at least one can be reserved as a backup. When other storage devices 6 are storing slag, no slag is fed into the backup device; when other storage devices 6 are discharging slag, slag is fed into the backup device. That is, during normal slag discharge, other storage devices and the backup device work alternately. The slag storage device 6 can be, but is not limited to, a slag storage tank. When the tank is full, it can be lifted to the outside of the shaft 2 using an existing lifting device.

[0140] Furthermore, such as Figure 10 As shown, there are multiple blowers 1 connected in parallel to provide sufficient airflow as needed to ensure the suction and discharge of slag and stone. Among them, blower 1 is a vacuum blower that can be, but is not limited to, frequency converter.

[0141] In an optional embodiment of the present invention, such as Figure 10 As shown, a dust removal device 7 is installed on the slag conveying pipeline 5 upstream of the blower 1 along the slag conveying direction. Specifically, the dust removal device 7 is located between the slag storage device 6 and the blower 1 (i.e., downstream of the slag storage device 6 and upstream of the blower 1). The dust removal device 7 can intercept debris in the slag conveying pipeline 5 to prevent dust, slag, and other debris from entering the downstream blower 1 after the slag falls into the slag storage device 6, thus preventing damage to the blower 1.

[0142] Furthermore, the dust removal device 7 includes a cyclone dust collector 701 and a bag dust collector 702 connected in series on the slag conveying pipeline 5.

[0143] In an optional embodiment of the present invention, such as Figure 10As shown, the slag conveying device also includes a control device 8 and a frequency converter 9. The image signal receiving end of the control device 8 is electrically connected to the image signal output end of the image acquisition device 3, and the control signal output end of the control device 8 is electrically connected to the control end of the frequency converter 9. The frequency converter 9 is used to control the frequency of the blower 1. In actual operation, the image acquisition device 3 acquires image information of the slag and stone inside the shaft 2 and sends the acquired image information to the control device 8. The control device 8 outputs a control signal to the frequency converter 9 according to the pneumatic conveying method based on slag and stone state recognition. The frequency converter 9 adjusts the operating frequency of the blower 1 according to the control signal, so that the slag and stone can be smoothly sucked out by the blower 1 while the blower 1 can maintain a suitable frequency, without having to run at the maximum frequency all the time, effectively reducing energy waste. The control device 8 can be, but is not limited to, a host computer.

[0144] The slag conveying device of the present invention has the characteristics and advantages of the above-mentioned pneumatic conveying method based on slag state identification, which will not be repeated here.

[0145] Implementation Method 3

[0146] The present invention provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned pneumatic conveying method based on the identification of the state of slag.

[0147] Specifically, the computer device can be a computer terminal, a server, or a similar computing device.

[0148] Implementation Method 4

[0149] The present invention provides a computer-readable storage medium storing a computer program that executes the above-described pneumatic conveying method based on the identification of the state of slag and soil.

[0150] Specifically, computer-readable storage media include both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable storage media does not include transient media, such as modulated data signals and carrier waves.

[0151] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0153] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0155] The above description is merely an illustrative embodiment of the present invention and is not intended to limit the scope of the invention. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A pneumatic conveying method based on the state recognition of construction waste, which uses negative pressure generated by a blower to pump and discharge construction waste from a vertical shaft, characterized in that... The pneumatic conveying method includes the following steps: Different types of slag and stone are classified and a database is established; Collect image information of slag and rock at preset locations inside the vertical shaft; Images of slag and rock at preset locations within the vertical shaft are collected from different angles. Based on the slag image information, match the slag parameters corresponding to the slag image information in the database; Calculate at least the length, width, and height of the slag based on the slag image information collected from different angles; The length, width, and height values ​​of the slag obtained from different angles were compared respectively; The maximum length, maximum width, and maximum height values ​​of the slag obtained from different angles are extracted to form a cubic envelope of the slag. Calculate the volume of the cubic envelope of the slag; The weight of the slag is determined based on the slag image information and the slag parameters; The frequency of the blower is adjusted according to the weight of the slag and stone to match the corresponding wind speed.

2. The pneumatic conveying method based on slag condition identification as described in claim 1, characterized in that, The classification of different types of slag and the establishment of a database include: Acquire images of different types of slag and create a dataset; The images of the different types of slag are repositioned to expand the dataset; A convolutional neural network model is built based on the data in the dataset, and the database includes the convolutional neural network model.

3. The pneumatic conveying method based on slag condition identification as described in claim 2, characterized in that, The step of building a convolutional neural network model based on the data in the dataset includes: The data in the dataset are normalized. Define neural network parameter variables; Set corresponding convolution kernels for different types of slag; At least one convolutional layer is added for each of the different convolutional kernels to construct feature maps. Based on the original convolutional layer, multiple convolutional layers are created for different convolutional kernels, and the input data of each convolutional layer is set as the output data of the previous convolutional layer. The convolutional neural network model is formed by training the convolutional neural network using the set input data and output data. The input data includes images of different types of slag and the output data includes the category of slag. A recognition script file is created based on the convolutional neural network model, and the database includes the recognition script file.

4. The pneumatic conveying method based on slag condition identification as described in claim 3, characterized in that, The neural network parameter variables include the iteration cycle of images of different types of slag and / or the number of images of different types of slag processed.

5. The pneumatic conveying method based on the state recognition of construction waste as described in claim 3, characterized in that, Setting corresponding convolution kernels for different types of slag includes: setting the size of the convolution kernel.

6. The pneumatic conveying method based on slag condition identification as described in claim 3, characterized in that, The images of different types of slag and stone include pictures of different types of slag and stone collected. The image with a fixed width and height is fed into the convolutional layer to construct feature maps with different convolutional kernels.

7. The pneumatic conveying method based on slag condition identification as described in claim 2, characterized in that, The position adjustment includes rotation, translation along the horizontal direction and / or translation along the vertical direction.

8. The pneumatic conveying method based on slag condition identification as described in claim 1, characterized in that, Determining the weight of the slag based on the slag image information and the slag parameters includes: The weight of the slag is obtained based on the type, density, and volume of the cubic envelope of the slag.

9. The pneumatic conveying method based on the state recognition of construction waste as described in claim 1 or 8, characterized in that, The step of adjusting the frequency of the blower according to the weight of the slag and stone to match the corresponding wind speed includes: Establish a correspondence between the weight of the slag and the frequency of the blower, wherein the blower can pump and discharge the slag of the corresponding weight at the corresponding frequency; The frequency of the blower is obtained based on the weight of the slag and stone. The frequency of the blower is adjusted so that the frequency of the blower and the weight of the slag meet the specified correspondence.

10. A slag conveying device, wherein the slag conveying device uses the pneumatic conveying method based on slag state identification as described in any one of claims 1 to 9 for slag conveying, and uses negative pressure suction to suction and discharge slag from the vertical shaft, characterized in that, The slag conveying device includes: A pre-storage unit is used to classify different types of slag and establish a database; The image acquisition unit is used to acquire image information of slag and rock at preset locations inside the shaft; The proportioning unit is used to match the slag parameters corresponding to the slag image information in the database according to the slag image information. The parameter determination unit is used to determine the weight of the slag based on the slag image information and the slag parameters; The adjustment unit is used to adjust the frequency of the blower used for pumping the slag according to the weight of the slag, so as to match the corresponding wind speed.

11. The slag conveying device as described in claim 10, characterized in that, The image acquisition unit includes multiple image acquisition devices, which are mounted on the shaft excavation equipment and can move along the shaft depth direction with the shaft excavation equipment.

12. The slag conveying device as described in claim 11, characterized in that, The slag conveying device includes a suction nozzle, a slag conveying pipeline, a blower, and a slag storage device. The suction nozzle is located on or near the cutter head of the vertical shaft excavation equipment. The suction nozzle is connected to one end of the slag conveying pipeline, and the other end of the slag conveying pipeline is connected to the slag storage device. The blower is located on the slag conveying pipeline or on the slag storage device.

13. The slag conveying device as described in claim 12, characterized in that, A dust removal device is installed upstream of the blower along the conveying direction of the slag and stone.

Images