A slag leakage construction method and system based on deep learning algorithm

Abstract

The present invention discloses a slag leakage construction method and system based on a deep learning algorithm, which belongs to the field of data acquisition and processing technology. Geological data is collected by sensors integrated on the bucket teeth of an excavator, and the geological data is identified by a first deep learning algorithm to determine lithologic stratification data. Based on the lithologic stratification data, the excavation rate of the excavator is intelligently adjusted, and the excavator is controlled to perform slag leakage excavation according to the adjusted excavation rate, which can effectively avoid the occurrence of collapse accidents and ensure the safety of slag leakage construction. After the slag leakage excavation, the slag leakage is transported to the outside through a vertical slag leakage channel, and a stroboscopic camera is used to collect real-time slag leakage images at the slag leakage outlet. Based on the real-time slag leakage images, a second deep learning algorithm is used to predict the probability of blockage of the slag leakage outlet due to the current slag leakage amount, and the opening size of the slag leakage outlet is adjusted according to the predicted probability, thereby avoiding the problem of excessive dust or blockage.

Description

Technical Field

[0001] The present invention belongs to the technical field of data acquisition and processing, and specifically relates to a slag leakage construction method and system based on a deep learning algorithm. Background Art

[0002] Traditional layered excavation and vertical transportation techniques are typically used for deep foundation pit excavation. Layered excavation is manually controlled, and inclined or vertical transportation equipment such as gantry cranes, tower cranes, and conveyor belts are deployed to lift the slag to the surface for stacking and transportation. This method is currently widely used in deep foundation pit construction for urban subways and urban structures. Traditional deep foundation pit excavation typically follows a fixed layer thickness (e.g., 2 meters per layer), failing to account for construction in environments with interbedded sandstone and mudstone, cracks, and contact surfaces. Site clearance and excavation speeds often rely on manual judgment based on engineering experience. Complex strata, such as those with widespread soft rock, can easily lead to collapse, posing significant engineering risks. Furthermore, the slag outlets are manually set to a fixed opening size, which can easily be too large or too small, resulting in excessive dust and blockage. Summary of the Invention

[0003] The present invention provides a slag leakage construction method and system based on a deep learning algorithm to solve the problems existing in the prior art.

[0004] In a first aspect, the present invention provides a slag leakage construction method based on a deep learning algorithm, comprising:

[0005] During the construction of the subway station ventilation shaft foundation pit, geological data is collected through sensors integrated on the excavator bucket teeth, and the first deep learning algorithm is used to identify the geological data to determine the lithologic stratification data;

[0006] Based on the lithologic stratification data, the excavator's advance rate is intelligently adjusted, and the excavator is controlled to perform slag excavation according to the adjusted advance rate;

[0007] After the slag is excavated, the slag is transported to the outside through a vertical slag channel, and a stroboscopic camera is used at the slag outlet to capture real-time slag images;

[0008] Based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will cause the slag leakage outlet to be blocked, and the opening size of the slag leakage outlet is adjusted according to the predicted probability to complete the slag leakage construction.

[0009] In a possible implementation, the method further includes:

[0010] The dust concentration of the slag leakage outlet is collected by a dust sensor, and when the dust concentration of the slag leakage is greater than a preset concentration threshold, the atomizing nozzle arranged at the slag leakage outlet is turned on to spray for dust reduction treatment.

[0011] In one possible implementation, during the construction of a subway station ventilation shaft foundation pit, geological data is collected using sensors integrated into the excavator's bucket teeth, including:

[0012] During the construction of the subway station ventilation shaft foundation pit, pressure sensing data was collected through the pressure sensor integrated on the excavator bucket teeth;

[0013] Preliminarily determining lithologic layering data based on the pressure sensing data; wherein the lithologic layering data includes whether the current excavation layer is a sandstone layer, a mudstone layer, or a plain fill layer;

[0014] When the current excavation layer is a sandstone layer or a mudstone layer, the fracture density data is collected by the geological radar integrated on the bucket tooth of the excavator, and based on the fracture density data, the lithologic stratification data is corrected by using the first deep learning algorithm to obtain the corrected lithologic stratification data.

[0015] In one possible implementation, based on the lithologic layering data, intelligently adjusting the excavation rate of the excavator, and controlling the excavator to perform slag leakage excavation according to the adjusted excavation rate, includes:

[0016] The preset excavation rate corresponding to the lithologic stratification data is dispatched from the database, and the excavator is controlled to perform slag excavation according to the adjusted excavation rate.

[0017] In one possible embodiment, the process of controlling the excavator to perform slag excavation according to the adjusted excavation rate further includes: continuously identifying lithologic layer data and determining the number of excavation layers; wherein the number of excavation layers is increased by one when the lithologic layer data changes once;

[0018] The lithologic stratification data and the number of excavation layers are simultaneously transmitted to the staff so that the staff can manage the construction site of the subway station ventilation shaft foundation pit.

[0019] In one possible implementation, based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will cause the slag leakage outlet to be blocked, and the opening size of the slag leakage outlet is adjusted according to the predicted probability, including:

[0020] Acquire the slag leakage particle size data in the real-time slag leakage image;

[0021] Based on the slag particle size data, slag leakage amount, and slag leakage outlet opening degree, a second deep learning algorithm is used to predict the probability that the slag leakage outlet will be blocked due to the current slag leakage amount;

[0022] When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet.

[0023] In a possible implementation, when the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet, including:

[0024] When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, the opening size of the slag leakage outlet is adjusted according to a preset adjustment amount to obtain the opening size of the slag leakage outlet after adjustment;

[0025] Based on the opening size of the slag outlet after adjustment, the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is obtained again, and it is determined whether the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is greater than the preset probability threshold. If so, return to the previous step, otherwise continue monitoring.

[0026] In a possible implementation, the method further includes: when the slag particle size data satisfies that the proportion of mudstone powder exceeds a preset threshold, directly adjusting the opening size of the slag outlet to a preset opening size value.

[0027] In a possible implementation, the first deep learning algorithm is set to the RF algorithm; the second deep learning algorithm is set to the CNN-LSTM algorithm.

[0028] In a possible implementation, the second deep learning algorithm is optimized before use by using an intelligent optimization algorithm to train hyperparameters of the second deep learning algorithm.

[0029] In a second aspect, the present invention provides a slag leakage construction system based on a deep learning algorithm, comprising: a field data acquisition module, an excavation rate control module, a slag leakage image acquisition module, and an intelligent slag discharge control module;

[0030] The field data acquisition module is used to collect geological data through sensors integrated on the bucket teeth of the excavator during the construction of the subway station ventilation shaft foundation pit, and to identify the geological data using the first deep learning algorithm to determine the lithologic stratification data;

[0031] The excavation rate control module is used to intelligently adjust the excavation rate of the excavator based on the lithologic stratification data, and control the excavator to perform slag leakage excavation according to the adjusted excavation rate;

[0032] The slag leakage image acquisition module is used to transport the slag leakage to the outside through the vertical slag leakage channel after the slag leakage excavation is carried out, and to use a stroboscopic camera to collect real-time slag leakage images at the slag leakage outlet;

[0033] The intelligent slag discharge control module is used to predict the probability of the current slag leakage amount causing the slag leakage outlet to be blocked based on the real-time slag leakage image using a second deep learning algorithm, and adjust the opening size of the slag leakage outlet according to the predicted probability to complete the slag leakage construction.

[0034] The present invention provides a slag excavation construction method and system based on a deep learning algorithm. The method collects geological data through sensors integrated on the bucket teeth of the excavator, and uses a first deep learning algorithm to identify the geological data to determine the lithologic stratification data. Based on the lithologic stratification data, the excavator's excavation rate is intelligently adjusted, and the excavator is controlled to perform slag excavation according to the adjusted excavation rate, which can effectively avoid the occurrence of collapse accidents and ensure the safety of slag excavation. After the slag excavation, the slag is transported to the outside through a vertical slag excavation channel, and a stroboscopic camera is used to collect real-time slag images at the slag outlet. Based on the real-time slag images, a second deep learning algorithm is used to predict the probability of blockage of the slag outlet due to the current slag amount, and the opening size of the slag outlet is adjusted according to the predicted probability, avoiding the problem of excessive dust or blockage. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0036] Figure 1 A flowchart of a slag leakage construction method based on a deep learning algorithm provided in an embodiment of the present invention.

[0037] Figure 2 A structural schematic diagram of a slag leakage construction system based on a deep learning algorithm provided in an embodiment of the present invention.

[0038] Among them, 201-field data acquisition module, 202-excavation rate control module, 203-slag leakage image acquisition module, and 204-intelligent slag discharge control module.

[0039] The above drawings illustrate specific embodiments of the present invention, which will be described in more detail below. These drawings and the accompanying description are not intended to limit the scope of the present invention in any way, but rather to illustrate the concept of the present invention to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0040] Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0041] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0042] like Figure 1 As shown, an embodiment of the present invention provides a slag leakage construction method based on a deep learning algorithm, comprising:

[0043] S101. During construction of a subway station ventilation shaft foundation pit, collect geological data using sensors integrated on an excavator bucket tooth, and use a first deep learning algorithm to identify the geological data to determine lithologic stratification data.

[0044] S102, intelligently adjusting the excavation rate of the excavator based on the lithologic stratification data, and controlling the excavator to perform slag-leaking excavation according to the adjusted excavation rate;

[0045] S103, after the slag is excavated, the slag is transported to the outside through a vertical slag channel, and a stroboscopic camera is used at the slag outlet to capture a real-time slag image;

[0046] S104. Based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will cause the slag leakage outlet to be blocked, and the opening size of the slag leakage outlet is adjusted according to the predicted probability to complete the slag leakage construction.

[0047] In a possible implementation, the method further includes:

[0048] The dust concentration of the slag leakage outlet is collected by a dust sensor, and when the dust concentration of the slag leakage is greater than a preset concentration threshold, the atomizing nozzle arranged at the slag leakage outlet is turned on to spray for dust reduction treatment.

[0049] In one possible implementation, during the construction of a subway station ventilation shaft foundation pit, geological data is collected using sensors integrated into the excavator's bucket teeth, including:

[0050] During the construction of the subway station ventilation shaft foundation pit, pressure sensing data was collected through the pressure sensor integrated on the excavator bucket teeth;

[0051] Preliminarily determining lithologic layering data based on the pressure sensing data; wherein the lithologic layering data includes whether the current excavation layer is a sandstone layer, a mudstone layer, or a plain fill layer;

[0052] Different pressure zones can be set for sandstone layers, mudstone layers or plain fill layers. By judging in which pressure interval the pressure sensing data is located, the current lithologic stratification data can be determined.

[0053] When the current excavation layer is a sandstone layer or a mudstone layer, the fracture density data is collected by the geological radar integrated on the bucket tooth of the excavator, and based on the fracture density data, the lithologic stratification data is corrected by using the first deep learning algorithm to obtain the corrected lithologic stratification data.

[0054] For example, the first deep learning algorithm can be used to classify the lithologic stratification data, thereby further determining the lithologic stratification data and improving the accuracy of the lithologic stratification data.

[0055] In one possible implementation, based on the lithologic layering data, intelligently adjusting the excavation rate of the excavator, and controlling the excavator to perform slag leakage excavation according to the adjusted excavation rate, includes:

[0056] The preset excavation rate corresponding to the lithologic stratification data is dispatched from the database, and the excavator is controlled to perform slag excavation according to the adjusted excavation rate.

[0057] By intelligently adjusting the excavator's advance rate, collapse accidents can be effectively avoided and engineering risks can be reduced.

[0058] In one possible embodiment, the process of controlling the excavator to perform slag excavation according to the adjusted excavation rate further includes: continuously identifying lithologic layer data and determining the number of excavation layers; wherein the number of excavation layers is increased by one when the lithologic layer data changes once;

[0059] The lithologic stratification data and the number of excavation layers are simultaneously transmitted to the staff so that the staff can manage the construction site of the subway station ventilation shaft foundation pit.

[0060] For example, workers can select different reinforcement schemes based on the rock stratification data and the number of excavation layers to improve construction safety.

[0061] In one possible implementation, based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will cause the slag leakage outlet to be blocked, and the opening size of the slag leakage outlet is adjusted according to the predicted probability, including:

[0062] Acquire the slag leakage particle size data in the real-time slag leakage image;

[0063] During the free fall of the slag, the camera continuously scans at a line frequency of 172kHz. To reduce the smearing caused by camera flicker during the falling slag, the light source's strobe frequency is synchronized with the slag's movement speed to eliminate smear. Particle outlines can be extracted from the real-time slag image, separating interlocking particles to improve segmentation accuracy. The area-equivalent diameter is then used to generate a particle size distribution curve for the slag.

[0064] Based on the slag particle size data, slag leakage amount, and slag leakage outlet opening degree, a second deep learning algorithm is used to predict the probability that the slag leakage outlet will be blocked due to the current slag leakage amount;

[0065] Among them, the amount of slag leakage can be expressed by the working rate / working power of the slag extraction equipment or transportation equipment, and then the slag particle size data, slag leakage amount and slag outlet opening degree can be used as input data of the second deep learning algorithm to realize the probability prediction of blockage.

[0066] When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet.

[0067] The output of the second deep learning algorithm is essentially the probability of blockage and non-blockage. The blockage probability can be used as the prediction probability, and the larger the prediction probability, the greater the possibility that the current slag leakage will cause the slag outlet to be blocked.

[0068] In a possible implementation, when the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet, including:

[0069] When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, the opening size of the slag leakage outlet is adjusted according to a preset adjustment amount to obtain the opening size of the slag leakage outlet after adjustment;

[0070] Based on the opening size of the slag outlet after adjustment, the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is obtained again, and it is determined whether the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is greater than the preset probability threshold. If so, return to the previous step, otherwise continue monitoring.

[0071] Optionally, in addition to adjusting the slag outlet, the angle of the slag discharge channel can also be adjusted to better control the slag discharge efficiency.

[0072] In addition to using a feedback control algorithm to adjust the opening size of the slag outlet, direct adjustment is also possible. For example, when the proportion of large-grained sandstone (>300mm) is between 30% and 50%, the opening size of the slag outlet is controlled to 70%, with the slag channel angled at 45° from the horizontal. When the proportion of large-grained sandstone (>300mm) is between 50% and 70%, the opening size of the slag outlet is controlled to 80%, with the slag channel angled at 50° from the horizontal, to improve slag discharge efficiency and prevent blockage. When the proportion of large-grained sandstone (>300mm) exceeds 70%, the opening size of the slag outlet is controlled to 90%, with the slag channel angled at 60° from the horizontal, to improve slag discharge efficiency and prevent blockage. In other cases, slag discharge can be carried out at the preset opening size, as blockage is less likely and dust generation is minimal.

[0073] In a possible implementation, the method further includes: when the slag particle size data satisfies that the proportion of mudstone powder exceeds a preset threshold, directly adjusting the opening size of the slag outlet to a preset opening size value.

[0074] For example, slag with a particle size of less than 10 mm can be considered as mudstone dust. When the proportion of mudstone dust exceeds 70%, the opening and closing size of the slag outlet can be reduced to 60%, and the angle of the slag discharge channel can be controlled to 40% to control the slag discharge efficiency and reduce dust pollution.

[0075] In a possible implementation, the first deep learning algorithm is set to the RF (Random Forest) algorithm; the second deep learning algorithm is set to the CNN-LSTM (Convolutional Neural Network-Long Short-Term Memory) algorithm.

[0076] It is worth noting that, in addition to the above-mentioned deep learning algorithms, other deep learning algorithms can also be used as the first deep learning algorithm and the second deep learning algorithm.

[0077] In a possible implementation, the second deep learning algorithm is optimized before use by using an intelligent optimization algorithm to train hyperparameters of the second deep learning algorithm.

[0078] In an embodiment of the present invention, using an intelligent optimization algorithm to train the hyperparameters of the second deep learning algorithm may include:

[0079] A1. Initialize the hyperparameters of the second deep learning algorithm using a random initialization method or a chaotic mapping method to obtain multiple hyperparameter individuals;

[0080] The hyperparameter individuals include all or part of the hyperparameters to be trained of the second deep learning algorithm;

[0081] A2. Obtain the loss function value corresponding to each hyperparameter individual and determine the hyperparameter individual with the smallest loss function value as the optimal individual;

[0082] The loss function value can be obtained through the root mean square loss function or the cross entropy loss function. When obtaining the loss function value corresponding to each hyperparameter individual, it is necessary to use the data pair of training data and training labels. The training data can be the leakage particle size data, leakage amount and leakage outlet opening and closing degree at historical moments, and the training label can be the actual blockage situation under the conditions of the training data. If it is blocked, it is set to 1, and if it is not blocked, it is set to 0.

[0083] A3. Based on the optimal individual, a search is performed on the hyperparameter individual. The hyperparameter individual obtained after the search is:

[0084]

[0085]

[0086] in, Indicates the t During the training n Hyperparameter individuals, Indicates the n The hyperparameters after one update, n =1,2,…,N, N represents the total number of hyperparameter individuals, Indicates the n The inertia weight corresponding to each hyperparameter individual, represents the optimal individual, u represents the first helical constant, v represents the second helical constant, e represents a natural constant, means (0,2 π ), sin represents the sine function, cos represents the cosine function, and exp represents the exponential function with the natural constant e as the base. Indicates the n The fitness corresponding to each hyperparameter individual, Represents the fitness corresponding to the optimal individual, and the fitness is obtained by taking the inverse of the loss function value;

[0087] Optionally, when obtaining the fitness, in order to avoid the denominator being zero, the loss function value may be first added to a preset constant. In order to prevent the preset constant from affecting the fitness result, the preset constant may be set to 0.0001.

[0088] This one-shot search allows the original hyperparameter individuals to merge with the optimal individual, occupying a new position in the solution space. This not only effectively avoids collisions during the search process but also prevents the algorithm from falling into local optima. Adaptive fusion based on fitness allows the algorithm to achieve higher convergence accuracy in the later stages. Finally, a spiral search function is introduced, which allows the search to proceed along an irregular path, improving the ability to find the global optimum.

[0089] A4. Perform a secondary search on the hyperparameter individuals after the first search. The hyperparameter individuals after the second search are:

[0090]

[0091] in, Indicates the t During the training h The hyperparameter individuals after one search, Expressed as h After a search, the hyperparameter individuals are randomly matched with the information fusion individuals. The information fusion individuals and the hyperparameter individuals different, h =1,2,…,N, M represents a positive integer generated by Levy flight, , Indicates a random number between (0,1) generated by Levy flight, represents the floor function, represents the first random number between (0,1), Represents the second random number between (0,1);

[0092] This secondary search process can randomly integrate the position information of other hyperparameter individuals and introduce an exponential function for control, which can effectively improve the search ability in unfamiliar areas. At the same time, it searches around the optimal position. When the hyperparameter individual is outside the group, it will accelerate to approach. Otherwise, it will switch to a fine search, improving the search ability of the algorithm.

[0093] A5. Perform three searches on the hyperparameter individuals after the second search. The hyperparameter individuals after the third search are:

[0094]

[0095]

[0096] in, Indicates the t During the training k Hyperparameter individuals after secondary search, Indicates the kThe hyperparameter individuals after three searches, k =1,2,…,N, represents the third random number between (0,1), represents the fourth random number between (0,1), represents pi, Expressed as k After the secondary search, the hyperparameter individual randomly matched with the loss function value of the random hyperparameter individual is smaller. When is the optimal individual, Set to the hyperparameter individual with the second smallest loss function value; represents the optimal information learning factor, represents the fifth random number between (0,1), Indicates the preset maximum number of training times;

[0097] This three-step search process enables the hyperparameter individual to learn the position information of the optimal individual in the solution space at an adaptive rate while learning information about other better positions, thereby improving the search accuracy and speed of the algorithm.

[0098] A6. Perform four searches on the hyperparameter individuals after three searches. The hyperparameter individuals after four searches are:

[0099]

[0100] in, Indicates the t During the training q The hyperparameter individuals after three searches, Indicates the q The hyperparameter individuals after four searches, q =1,2,…,N; Represents the worst individual, that is, the hyperparameter individual with the largest loss function value; Represents the sixth random number between (0,1).

[0101] The four-step search process can perform mutually exclusive information fusion, so that the hyperparameter individuals move randomly under the attraction of the best individuals and the repulsion of the worst individuals, thereby achieving position updates, improving the diversity of the population, and enabling the algorithm to escape from the local optimal solution.

[0102] Optionally, the four searches may be accepted only if the loss function value of the hyperparameter individual after the four searches decreases, otherwise the four searches are rejected. By introducing the greedy strategy, the convergence ability and convergence speed of the algorithm can be effectively guaranteed.

[0103] A7. Determine whether the current number of training times is greater than or equal to the preset maximum number of training times. If so, redetermine the optimal individual based on the hyperparameter individual after four searches, and use the hyperparameters in the redetermined optimal individual as the final hyperparameters of the second deep learning algorithm. Otherwise, return to the step of determining the optimal individual.

[0104] The embodiment of the present invention provides an intelligent optimization algorithm for training the hyperparameters of the second deep learning algorithm. Compared with the existing technology, it has a powerful global search capability, is not easily trapped in the local optimum, and effectively improves the training speed and training accuracy, ensuring that the second deep learning algorithm after training can effectively perform predictions, and ultimately improves the prediction accuracy of the current slag leakage amount causing slag leakage outlet blockage, thereby preventing the occurrence of slag leakage blockage.

[0105] The present invention provides a slag excavation construction method based on a deep learning algorithm. The method collects geological data through sensors integrated on the bucket teeth of an excavator, and uses a first deep learning algorithm to identify the geological data to determine rock layer data. Based on the rock layer data, the excavator's excavation rate is intelligently adjusted, and the excavator is controlled to perform slag excavation according to the adjusted excavation rate. This can effectively avoid the occurrence of collapse accidents and ensure the safety of slag excavation. After the slag excavation, the slag is transported to the outside through a vertical slag excavation channel, and a stroboscopic camera is used to collect real-time slag images at the slag outlet. Based on the real-time slag images, a second deep learning algorithm is used to predict the probability of blockage of the slag outlet due to the current slag amount, and the opening size of the slag outlet is adjusted according to the predicted probability, thereby avoiding the problem of excessive dust or blockage.

[0106] like Figure 2 As shown, the embodiment of the present invention provides a slag leakage construction system based on a deep learning algorithm, comprising: a field data acquisition module 201, an excavation rate control module 202, a slag leakage image acquisition module 203, and an intelligent slag discharge control module 204;

[0107] The field data acquisition module 201 is used to collect geological data through sensors integrated on the bucket teeth of an excavator during the construction of a subway station ventilation shaft foundation pit, and to identify the geological data using a first deep learning algorithm to determine lithologic stratification data;

[0108] The excavation rate control module 202 is used to intelligently adjust the excavation rate of the excavator based on the lithologic layering data, and control the excavator to perform slag leakage excavation according to the adjusted excavation rate;

[0109] The slag leakage image acquisition module 203 is used to transport the slag leakage to the outside through the vertical slag leakage channel after the slag leakage excavation is carried out, and to use a stroboscopic camera to collect real-time slag leakage images at the slag leakage outlet;

[0110] The intelligent slag discharge control module 204 is used to predict the probability of the current slag leakage amount causing the slag leakage outlet to be blocked based on the real-time slag leakage image using a second deep learning algorithm, and adjust the opening size of the slag leakage outlet according to the predicted probability to complete the slag leakage construction.

[0111] Figure 2 The slag application system based on deep learning algorithm shown can execute the above method. Its principles and beneficial effects are similar and will not be repeated here.

[0112] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0114] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1A step that specifies a function in one or more boxes.

[0116] Those skilled in the art will understand that all or part of the steps in implementing the above facts and methods can be completed by instructing relevant hardware through a program, and the program involved or the program can be stored in a computer-readable storage medium. When the program is executed, it includes the following steps: the corresponding method steps are then brought out, and the storage medium can be ROM / RAM, a disk, an optical disk, etc.

[0117] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A slag leakage construction method based on deep learning algorithm, characterized in that: include: During the construction of the subway station ventilation shaft foundation pit, geological data is collected through sensors integrated on the excavator bucket teeth, and the first deep learning algorithm is used to identify the geological data to determine the lithologic stratification data; Based on the lithologic stratification data, the excavator's advance rate is intelligently adjusted, and the excavator is controlled to perform slag excavation according to the adjusted advance rate; After the slag is excavated, the slag is transported to the outside through a vertical slag channel, and a stroboscopic camera is used at the slag outlet to capture real-time slag images; Based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will lead to blockage of the slag leakage outlet, and the opening size of the slag leakage outlet is adjusted according to the predicted probability to complete the slag leakage construction; During the construction of the subway station ventilation shaft foundation pit, sensors integrated into the excavator bucket teeth were used to collect geological data, including: During the construction of the subway station ventilation shaft foundation pit, the pressure sensor data was collected through the pressure sensor integrated on the excavator bucket teeth; Preliminarily determining lithologic layering data based on the pressure sensing data; wherein the lithologic layering data includes whether the current excavation layer is a sandstone layer, a mudstone layer, or a plain fill layer; When the current excavation layer is a sandstone layer or a mudstone layer, fracture density data is collected by a geological radar integrated on the bucket tooth of the excavator. Based on the fracture density data, the lithologic layer data is corrected using a first deep learning algorithm to obtain the corrected lithologic layer data. The first deep learning algorithm is set to be an RF algorithm. Based on the real-time slag leakage image, a second deep learning algorithm is used to predict the probability that the current slag leakage amount will cause the slag leakage outlet to be blocked, and the opening size of the slag leakage outlet is adjusted according to the predicted probability, including: Acquire the slag leakage particle size data in the real-time slag leakage image; Based on the slag leakage particle size data, the amount of slag leakage, and the opening and closing degree of the slag leakage outlet, a second deep learning algorithm is used to predict the probability that the slag leakage outlet will be blocked due to the current slag leakage amount, and the second deep learning algorithm is set to be a CNN-LSTM algorithm; When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet.

2. The slag leakage construction method based on deep learning algorithm according to claim 1 is characterized in that: Also includes: The dust concentration of the slag leakage outlet is collected by a dust sensor, and when the dust concentration of the slag leakage is greater than a preset concentration threshold, the atomizing nozzle arranged at the slag leakage outlet is turned on to spray for dust reduction treatment.

3. The slag leakage construction method based on deep learning algorithm according to claim 1 is characterized in that: Based on the lithologic stratification data, the excavator's advance rate is intelligently adjusted, and the excavator is controlled to perform slag excavation according to the adjusted advance rate, including: The preset excavation rate corresponding to the lithologic stratification data is dispatched from the database, and the excavator is controlled to perform slag excavation according to the adjusted excavation rate.

4. The slag leakage construction method based on deep learning algorithm according to claim 3 is characterized in that: The process of controlling the excavator to perform slag excavation according to the adjusted excavation rate also includes: continuously identifying lithologic layer data and determining the number of excavation layers; wherein the number of excavation layers increases by one when the lithologic layer data changes once; The lithologic stratification data and the number of excavation layers are simultaneously transmitted to the staff so that the staff can manage the construction site of the subway station ventilation shaft foundation pit.

5. The slag leakage construction method based on deep learning algorithm according to claim 3 is characterized in that: When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, a feedback adjustment algorithm is used to adjust the opening size of the slag leakage outlet, including: When the predicted probability that the current slag leakage amount causes the slag leakage outlet to be blocked is greater than a preset probability threshold, the opening size of the slag leakage outlet is adjusted according to a preset adjustment amount to obtain the opening size of the slag leakage outlet after adjustment; Based on the opening size of the slag outlet after adjustment, the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is obtained again, and it is determined whether the predicted probability that the current slag leakage amount will cause the slag outlet to be blocked is greater than the preset probability threshold. If so, return to the previous step, otherwise continue monitoring.

6. The slag leakage construction method based on deep learning algorithm according to claim 4 is characterized in that: Also includes: When the slag particle size data satisfies the condition that the proportion of mudstone powder exceeds a preset threshold value, the opening size of the slag outlet is directly adjusted to the preset opening size value.

7. The slag leakage construction method based on deep learning algorithm according to claim 4 is characterized in that: Before use, the second deep learning algorithm is optimized by using an intelligent optimization algorithm to train the hyperparameters of the second deep learning algorithm.

8. A slag leakage construction system based on a deep learning algorithm, which is capable of executing the slag leakage construction method based on a deep learning algorithm according to any one of claims 1 to 7, characterized in that: include: On-site data acquisition module, excavation rate control module, slag leakage image acquisition module and intelligent slag discharge control module; The field data acquisition module is used to collect geological data through sensors integrated on the bucket teeth of the excavator during the construction of the subway station ventilation shaft foundation pit, and to identify the geological data using the first deep learning algorithm to determine the lithologic stratification data; The excavation rate control module is used to intelligently adjust the excavation rate of the excavator based on the lithologic stratification data, and control the excavator to perform slag leakage excavation according to the adjusted excavation rate; The slag leakage image acquisition module is used to transport the slag leakage to the outside through the vertical slag leakage channel after the slag leakage excavation is carried out, and to use a stroboscopic camera to collect real-time slag leakage images at the slag leakage outlet; The intelligent slag discharge control module is used to predict the probability of the current slag leakage amount causing the slag leakage outlet to be blocked based on the real-time slag leakage image, using a second deep learning algorithm, and adjust the opening size of the slag leakage outlet according to the predicted probability to complete the slag leakage construction.

Images