A cloud prediction-based shield pressure maintaining regulation method and system

Abstract

The application discloses a cloud prediction-based shield pressure maintaining regulation method and system, relates to the technical field of shield regulation, and comprises the following steps: constructing a three-dimensional geological model of a construction area, acquiring control parameters and attitude parameters of a shield machine, and constructing a three-dimensional simulation model; acquiring an overbreak area and an overbreak volume of the shield machine; constructing a volume prediction model in combination with equipment parameters; acquiring historical overbreak volumes, historical internal pressure data and historical external pressure data; constructing a pressure evaluation model in combination with geological feature information; acquiring predicted overbreak volumes by using the volume prediction model; acquiring predicted internal pressure data and predicted external pressure data by using the pressure evaluation model; and respectively regulating the internal and external sides in a curve tunneling stage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield regulation, and particularly relates to a shield pressure maintaining regulation method and system based on cloud prediction. BACKGROUND

[0002] Shield construction is a widely used technology in underground tunnel construction, and has the advantages of fast construction speed, small influence on the ground, high automation degree, etc. With the acceleration of urbanization and the continuous advancement of infrastructure construction, shield construction is facing more and more complex geological conditions and higher engineering requirements.

[0003] In the shield construction process, the stability of the excavation face is the key to ensure construction safety and engineering quality. Since the original stratum stress balance is destroyed, it is necessary to effectively control the pressure of the excavation face. The existing technology for pressure maintaining regulation of the shield machine in the straight tunneling stage is quite mature, but the pressure maintaining regulation of the shield machine in the curved tunneling stage is obviously insufficient.

[0004] Due to the particularity of the curved tunneling stage, different pressure changes often occur on both sides of the curved line, so the regulation mechanism of the straight tunneling stage cannot be used. The existing technology lacks effective means to obtain the overbreak area and overbreak volume on the outside of the curved line, and cannot combine the overbreak volume with other related parameters to realize the advance regulation of the pressure on both sides. In view of the deficiencies of the existing technology, the present application provides a shield pressure maintaining regulation method and system based on cloud prediction. SUMMARY

[0005] The present application provides a shield pressure maintaining regulation method and system based on cloud prediction.

[0006] The purpose of the present application can be achieved by the following technical solutions: a shield pressure maintaining regulation system based on cloud prediction, comprising the following modules:

[0007] A geological modeling module is used to obtain geological feature information of a construction area, and a three-dimensional geological model is constructed using the geological feature information.

[0008] A tunneling simulation module is used to obtain control parameters and attitude parameters of the shield machine in the curved tunneling stage, and the tunneling process of the shield machine is simulated in the three-dimensional geological model to obtain a three-dimensional simulation model.

[0009] A volume prediction module is used to obtain an overbreak area and an overbreak volume of the shield machine in the curved tunneling stage in the three-dimensional simulation model, and a volume prediction model is constructed in combination with equipment parameters of the shield machine.

[0010] The pressure evaluation module is configured to obtain a historical overbreak volume by using the volume prediction model, obtain historical inner pressure data and historical outer pressure data of the shield tunneling machine in different curve tunneling stages, and construct a pressure evaluation model according to different historical overbreak volumes, historical inner pressure data, historical outer pressure data, and geological feature information.

[0011] The pressure adjustment module is configured to obtain a predicted overbreak volume by using the volume prediction model, obtain predicted inner pressure data and predicted outer pressure data by using the pressure evaluation model, and control the shield tunneling machine to adjust the inner and outer sides of the curve tunneling stage.

[0012] Further, the process of obtaining geological feature information of the construction area and constructing a three-dimensional geological model includes:

[0013] The geological feature information refers to rock types, geological structures, hydrological conditions, topographic features, and stratigraphic sequences of the construction area, and a three-dimensional structure model is constructed according to the topographic features and the stratigraphic sequences by using three-dimensional modeling software, and the geological structures, rock types, and hydrological conditions are imported into the three-dimensional structure model to obtain the three-dimensional geological model.

[0014] Further, the process of obtaining control parameters and attitude parameters of the shield tunneling machine in the curve tunneling stage and simulating the tunneling process of the shield tunneling machine in the three-dimensional geological model to obtain a three-dimensional simulation model includes:

[0015] The curve tunneling stage refers to a construction stage in which the shield tunneling machine tunnels along a preset curve route, the control parameters include a pushing speed, a total thrust, a cutter head rotating speed, and a cutter head torque, and the attitude parameters include an azimuth angle, a pitch angle, a roll angle, and a shield tail gap.

[0016] Obtaining equipment parameters of the shield tunneling machine, the equipment parameters including geometrical shapes and sizes of different components of the shield tunneling machine, and constructing a three-dimensional physical model of the shield tunneling machine according to the equipment parameters;

[0017] Importing the three-dimensional physical model of the shield tunneling machine into the three-dimensional geological model by using simulation software, and controlling the three-dimensional physical model to tunnel according to the control parameters and the attitude parameters to obtain a three-dimensional simulation model.

[0018] Further, the process of obtaining an overbreak area and an overbreak volume of the shield tunneling machine in the curve tunneling stage in the three-dimensional simulation model and constructing a volume prediction model in combination with the equipment parameters of the shield tunneling machine includes:

[0019] In the curve tunneling stage of the shield tunneling machine, the cutting range outside the curve route is actively expanded to form a gap, and the shield body is passively extruded against the surrounding rock on the inside of the curve route.

[0020] The overbreak area refers to a void area formed in a preset fixed period after the shield machine actively expands the cutting range, and the overbreak volume refers to the volume size of the overbreak area in the preset fixed period, and the region where the stress of the soil is increased due to extrusion is marked as a loss region;

[0021] In the three-dimensional simulation model, the overbreak area and the overbreak volume of the shield machine in a single preset fixed period under different control parameters and attitude parameters are obtained, a volume prediction set is generated in combination with the equipment parameters of the shield machine, and the volume prediction set is divided into a first training set and a first test set;

[0022] A first convolutional neural network is constructed, the equipment parameters and the different control parameters and attitude parameters in the first training set are taken as input data of the first convolutional neural network, and the corresponding overbreak volume in the first training set is taken as output data of the first convolutional neural network;

[0023] The first convolutional neural network is trained by using the first training set to obtain an initial first convolutional neural network, the initial first convolutional neural network is model-verified by using the first test set, and the initial first convolutional neural network with a first test error threshold less than or equal to a preset value is output as a volume prediction model.

[0024] Further, the process of obtaining the historical overbreak volume by using the volume prediction model and obtaining the historical inner pressure data and the historical outer pressure data of the shield machine in different curve tunneling stages includes:

[0025] The control parameters and the attitude parameters of the shield machine in the past curve tunneling stages are obtained, and the control parameters and the attitude parameters are input into the volume prediction model in combination with the equipment parameters of the shield machine to obtain the historical overbreak volume of the shield machine in a single preset fixed period in the corresponding curve tunneling stage;

[0026] The historical inner pressure data and the historical outer pressure data corresponding to the historical overbreak volume in the single preset fixed period are obtained, the historical inner pressure data refers to the earth pressure and the pore water pressure on the inner side of the curve line in the corresponding curve tunneling stage, and the historical outer pressure data refers to the earth pressure and the pore water pressure on the outer side of the curve line in the corresponding curve tunneling stage.

[0027] Further, the process of constructing a pressure evaluation model according to different historical overbreak volumes, historical inner pressure data, and historical outer pressure data in combination with geological feature information includes:

[0028] The pressure evaluation set is generated according to the historical overbreak volume, the historical inner pressure data, the historical outer pressure data, the equipment parameters, the control parameters, the attitude parameters, and the geological feature information in different preset fixed periods in different curve tunneling stages, and the pressure evaluation set is divided into a second training set and a second test set;

[0029] A second convolutional neural network is constructed, using different historical over-excavation volumes, geological features, equipment parameters, control parameters, and attitude parameters from the second training set as input data, and corresponding historical internal pressure data and historical external pressure data from the second training set as output data.

[0030] The second convolutional neural network is trained using the second training set to obtain an initial second convolutional neural network. The initial second convolutional neural network is validated using the second test set, and the initial second convolutional neural network with an error less than or equal to the preset second test error threshold is output as the stress assessment model.

[0031] Furthermore, the process of using a volume prediction model to obtain the predicted over-excavation volume, and using a pressure assessment model to obtain the predicted internal pressure data and predicted external pressure data, and controlling the tunnel boring machine to adjust the pressure on the inner and outer sides of the curved tunneling stage respectively includes:

[0032] Before the tunnel boring machine enters the curved tunneling stage, the preset control parameters and attitude parameters for achieving curved tunneling are obtained. Combined with the equipment parameters of the tunnel boring machine, the data is input into the volume prediction model to obtain the predicted over-excavation volume within a single preset fixed period after entering the curved tunneling stage.

[0033] The predicted over-excavation volume, control parameters, and attitude parameters of the tunnel boring machine (TBM) within a single preset fixed period after entering the curved tunneling stage are combined with geological feature information and equipment parameters and input into the pressure assessment model to obtain the predicted internal pressure data P within the single preset fixed period after entering the curved tunneling stage. n Predicted external pressure data P w ;

[0034] For the outer side of the curved section, inject a quick-setting grout into the over-excavated area and obtain the grouting pressure P. out and grouting volume V out For the inner side of the curved line, a retarding grout is injected into the loss area, and the grouting pressure P is obtained. in and grouting volume V in ;

[0035] ;

[0036] ;

[0037] ;

[0038] ;

[0039] For the preset correction coefficient, The value range is 0.03-0.05 MPa. The value range is 0.02-0.03 MPa. These are the static earth pressure and the overlying earth pressure in the construction area, respectively. The value is 3.14, where D is the diameter of the shield body. L represents the shield tail gap, and L represents the segment width.

[0040] After the tunnel boring machine enters the curved tunneling stage, the predicted over-excavation volume, predicted internal pressure data, and predicted external pressure data of the tunnel boring machine in the next preset fixed period are continuously acquired. Then, the grouting pressure and grouting volume on the outside of the curved line in the next preset fixed period, as well as the grouting pressure and grouting volume on the inside of the curved line in the next preset fixed period, are acquired. Based on the two grouting pressures and grouting volumes acquired, the pressure on the inside and outside of the curved tunneling stage is adjusted respectively.

[0041] A shield tunneling pressure control method based on cloud prediction includes the following steps:

[0042] Step S1: Obtain geological feature information of the construction area and construct a three-dimensional geological model using the geological feature information;

[0043] Step S2: Obtain the control parameters and attitude parameters of the tunnel boring machine during the curved tunneling stage, and simulate the tunneling process of the tunnel boring machine in the three-dimensional geological model to obtain a three-dimensional simulation model;

[0044] Step S3: Obtain the over-excavation area and its over-excavation volume of the tunnel boring machine in the curved tunneling stage in the three-dimensional simulation model, and construct a volume prediction model in combination with the equipment parameters of the tunnel boring machine;

[0045] Step S4: Use the volume prediction model to obtain the historical over-excavation volume, obtain the historical internal pressure data and historical external pressure data of the tunnel boring machine at different curve tunneling stages, and construct a pressure assessment model based on the different historical over-excavation volumes, historical internal pressure data, and historical external pressure data, combined with geological feature information.

[0046] Step S5: Use the volume prediction model to obtain the predicted over-excavation volume, use the pressure assessment model to obtain the predicted internal pressure data and predicted external pressure data, and control the tunnel boring machine to adjust the pressure on the inner and outer sides of the curved tunneling stage respectively.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] This invention constructs a three-dimensional geological model of the construction area and combines it with the equipment parameters, control parameters, and attitude parameters of the tunnel boring machine to simulate its tunneling process and obtain a three-dimensional simulation model. It can obtain the over-excavation volume of the over-excavation area of ​​the tunnel boring machine in the curved tunneling stage in the three-dimensional simulation model, which solves the problem that it is difficult to obtain the over-excavation area and over-excavation volume in the past. It uses deep learning technology to construct a volume prediction model, which can directly obtain the corresponding over-excavation volume according to different equipment parameters, control parameters, and attitude parameters.

[0049] By acquiring historical over-excavation volume, historical internal pressure data, and historical external pressure data of the tunnel boring machine (TBM) during previous curved tunneling stages, and combining these with other relevant parameters, a pressure assessment model was constructed. This model is beneficial for assessing the predicted internal and external pressure data of the TBM in the next preset fixed cycle based on the predicted over-excavation volume and corresponding relevant parameters. On this basis, the grouting pressure and grouting volume on both sides of the curved line can be obtained, enabling advance control of subsequent pressure changes. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the modules of the present invention. Detailed Implementation

[0051] like Figure 1 As shown, a shield tunneling pressure control system based on cloud prediction includes the following modules:

[0052] The geological modeling module is used to acquire geological feature information of the construction area and to construct a three-dimensional geological model using the geological feature information.

[0053] The tunneling simulation module is used to obtain the control parameters and attitude parameters of the tunnel boring machine during the curved tunneling stage, and to simulate the tunneling process of the tunnel boring machine in a three-dimensional geological model to obtain a three-dimensional simulation model.

[0054] The volume prediction module is used to obtain the over-excavation area and its over-excavation volume of the tunnel boring machine in the curved tunneling stage in the three-dimensional simulation model, and to build a volume prediction model in combination with the equipment parameters of the tunnel boring machine.

[0055] The pressure assessment module is used to obtain historical over-excavation volume using a volume prediction model, and to obtain historical internal pressure data and historical external pressure data of the tunnel boring machine at different curve excavation stages. Based on different historical over-excavation volumes, historical internal pressure data, and historical external pressure data, and combined with geological feature information, a pressure assessment model is constructed.

[0056] The pressure regulation module is used to obtain the predicted over-excavation volume using the volume prediction model and the predicted internal pressure data and predicted external pressure data using the pressure assessment model, and to control the tunnel boring machine to regulate the pressure on the inner and outer sides of the curved tunneling stage respectively.

[0057] It needs to be further explained that in the specific implementation process, the process of acquiring the geological feature information of the construction area and constructing the three-dimensional geological model using the geological feature information includes:

[0058] The geological feature information refers to the rock type (such as igneous rock, sedimentary rock, metamorphic rock) of the construction area, geological structure (including fault, fold, joint), hydrological condition (such as groundwater distribution, water quantity and water quality), geomorphic feature, stratigraphic sequence, etc.;

[0059] The construction area is divided into a three-dimensional grid using three-dimensional modeling software, a three-dimensional structure model of the construction area is constructed according to the geomorphic feature and the stratigraphic sequence, including the position and thickness of the stratigraphic interface, the geological structure is imported into the constructed three-dimensional structure model, including the position and strike of the fault, the shape of the fold and joint, and then the rock type and the hydrological condition are synchronized to the three-dimensional structure model, and the three-dimensional structure model at this time is taken as the three-dimensional geological model of the construction area.

[0060] It needs to be further explained that in the specific implementation process, the process of acquiring the control parameters and attitude parameters of the shield tunneling machine in the curve tunneling stage and simulating the tunneling process of the shield tunneling machine in the three-dimensional geological model to acquire the three-dimensional simulation model includes:

[0061] In the tunnel construction process, the shield tunneling machine generally adopts the straight tunneling mode, however, due to the influence of the terrain, buildings and underground obstacles, the tunnel line often cannot be completely straight, and needs to be partially designed as a curve line, at this time the shield tunneling machine needs to perform curve tunneling, and the curve tunneling stage refers to the construction stage in which the shield tunneling machine tunnels along the preset curve line;

[0062] The control parameters refer to various parameters related to the control of the shield tunneling machine in the curve tunneling stage, including the pushing speed, total thrust, cutter head rotating speed and cutter head torque, and the attitude parameters refer to various parameters related to the attitude of the shield tunneling machine in the curve tunneling stage, including the azimuth angle, pitch angle, roll angle and shield tail gap;

[0063] The device parameters of the shield tunneling machine are acquired, the device parameters are used to reflect the actual size and structure of the shield tunneling machine, including the geometric shape and size of the cutter head, shield body, jack, grouting system and other components, and a three-dimensional physical model of the shield tunneling machine is constructed according to the device parameters;

[0064] The three-dimensional physical model of the shield tunneling machine is imported into the three-dimensional geological model using simulation software, the three-dimensional physical model is controlled to tunnel according to the acquired control parameters and attitude parameters to acquire its three-dimensional simulation model, and the three-dimensional simulation model can simulate the tunneling process of the shield tunneling machine.

[0065] It needs to be further explained that, in the specific implementation process, the process of acquiring the overbreak area and the overbreak volume of the shield tunneling machine in the curve tunneling stage in the three-dimensional simulation model, combining with the equipment parameters of the shield tunneling machine to construct the volume prediction model includes:

[0066] In the curve tunneling stage, the shield tunneling machine needs to adjust the attitude of the cutter head and the shield body when turning. In order to reduce the pushing resistance, the cutting range on the outside of the curve line will be actively expanded to form an additional gap. Correspondingly, when the shield tunneling machine turns, the shield body will be passively extruded against the surrounding rock on the inside of the curve line. The extrusion will cause the stress of the soil in this area to increase;

[0067] The overbreak area refers to the gap area formed by the shield tunneling machine actively expanding the cutting range within a preset fixed period. The overbreak volume refers to the volume of the overbreak area within the preset fixed period. The area where the stress of the soil increases due to extrusion is marked as a loss area;

[0068] In the three-dimensional simulation model, the overbreak area and the overbreak volume formed by the shield tunneling machine within a single preset fixed period under different control parameters and attitude parameters are acquired. A corresponding volume prediction set is generated combining the equipment parameters of the shield tunneling machine, and the volume prediction set is divided into a first training set and a first test set;

[0069] A first convolutional neural network is constructed. The equipment parameters and different control parameters and attitude parameters in the first training set are used as the input data of the first convolutional neural network, and the corresponding overbreak volume in the first training set is used as the output data of the first convolutional neural network;

[0070] The first training set is used to train the first convolutional neural network to obtain an initial first convolutional neural network. The first test set is used to verify the model of the initial first convolutional neural network, and an initial first convolutional neural network with a first test error threshold less than or equal to a preset first test error threshold is output as the corresponding volume prediction model.

[0071] It needs to be further explained that, in the specific implementation process, the process of acquiring the historical overbreak volume using the volume prediction model and acquiring the historical inner pressure data and the historical outer pressure data of the shield tunneling machine in different curve tunneling stages includes:

[0072] The control parameters and the attitude parameters of the shield tunneling machine in the past curve tunneling stages are acquired, and the equipment parameters of the shield tunneling machine are input into the volume prediction model to acquire the historical overbreak volume within a single preset fixed period in the corresponding curve tunneling stage;

[0073] obtain historical inner pressure data and historical outer pressure data corresponding to the historical overbreak volume in a single preset fixed period, the historical inner pressure data refers to the stratum soil pressure and pore water pressure on the inner side of the curved tunneling stage, and the historical outer pressure data refers to the stratum soil pressure and pore water pressure on the outer side of the curved tunneling stage, both of which can be obtained by sensors arranged on the surface of the shield machine shield body, and both of which are bound to the control parameters and attitude parameters corresponding to the curved tunneling stage.

[0074] It needs to be further explained that, in the specific implementation process, according to different historical overbreak volumes, historical inner pressure data, historical outer pressure data, and geological feature information, the process of constructing a pressure evaluation model includes:

[0075] According to the historical overbreak volume, the historical inner pressure data, the historical outer pressure data, the equipment parameters, the control parameters, the attitude parameters and the geological feature information of the construction area in different preset fixed periods under different curved tunneling stages, a pressure evaluation set is generated, and it is divided into a second training set and a second test set;

[0076] A second convolutional neural network is constructed, different historical overbreak volumes, geological feature information, equipment parameters, control parameters, and attitude parameters in the second training set are taken as input data of the second convolutional neural network, and corresponding historical inner pressure data and historical outer pressure data in the second training set are taken as output data of the second convolutional neural network;

[0077] The second training set is used to train the second convolutional neural network to obtain an initial second convolutional neural network, the initial second convolutional neural network is verified by the second test set, and the initial second convolutional neural network with a second test error threshold less than or equal to a preset value is output as the corresponding pressure evaluation model.

[0078] It needs to be further explained that, in the specific implementation process, the process of obtaining the predicted overbreak volume by using the volume prediction model, obtaining the predicted inner pressure data and the predicted outer pressure data by using the pressure evaluation model, and controlling the shield machine to adjust the inner and outer sides of the curved tunneling stage includes:

[0079] Before the shield machine enters the curved tunneling stage, the preset control parameters and attitude parameters for realizing the curved tunneling are obtained, which are input into the volume prediction model in combination with the equipment parameters of the shield machine, to obtain the predicted overbreak volume of the shield machine in a single preset fixed period after entering the curved tunneling stage;

[0080] The predicted overbreak volume of the shield machine in a single preset fixed period after entering the curved tunneling stage, the control parameters and the attitude parameters, are input into the pressure evaluation model in combination with the geological feature information and the equipment parameters, to obtain the predicted inner pressure data and the predicted outer pressure data of the shield machine in a single preset fixed period after entering the curved tunneling stage, and are respectively recorded as Pn and P w ;

[0081] For the outside of the curved line, the quick-setting slurry is injected in the overbreak area to compensate for the stratum loss, inhibit the settlement, and obtain the grouting pressure P out and the grouting volume V out For the inside of the curved line, the slow-setting slurry is injected in the loss area to release the stress of the soil body, prevent the segment from being extruded, and obtain the grouting pressure P in and the grouting volume V in ;

[0082] ;

[0083] ;

[0084] ;

[0085] ;

[0086] is a preset correction coefficient, the value range of which is 0.03-0.05 Mpa, the value range of which is 0.02-0.03 Mpa, respectively the static earth pressure and the overburden pressure of the construction area, the value of which is 3.14, D is the diameter of the shield body, is the shield tail gap, and L is the segment width;

[0087] After the shield machine enters the curved tunneling stage, the predicted overbreak volume, the predicted internal pressure data, and the predicted external pressure data of the shield machine in the next preset fixed period are continuously obtained, and then the grouting pressure and the grouting volume of the outside of the curved line in the next preset fixed period and the grouting pressure and the grouting volume of the inside of the curved line in the next preset fixed period are obtained, and the two grouting pressures and the two grouting volumes are used to respectively adjust the pressure of the inside and the outside of the curved tunneling stage.

[0088] In the embodiment of the application, the method further comprises a shield pressure maintaining and regulating method based on cloud prediction, which comprises the following steps:

[0089] Step S1: Obtain the geological feature information of the construction area, and construct a three-dimensional geological model by using the geological feature information;

[0090] Step S2: Obtain the control parameters and the attitude parameters of the shield machine in the curved tunneling stage, and simulate the tunneling process of the shield machine in the three-dimensional geological model to obtain a three-dimensional simulation model;

[0091] Step S3: Obtain the overbreak area and overbreak volume of the shield tunneling machine in the curve tunneling stage in the three-dimensional simulation model, and construct a volume prediction model in combination with the equipment parameters of the shield tunneling machine;

[0092] Step S4: Obtain the historical overbreak volume by using the volume prediction model, obtain the historical inner pressure data and historical outer pressure data of the shield tunneling machine in different curve tunneling stages, and construct a pressure evaluation model in combination with the geological feature information according to different historical overbreak volumes, historical inner pressure data and historical outer pressure data;

[0093] Step S5: Obtain the predicted overbreak volume by using the volume prediction model, obtain the predicted inner pressure data and predicted outer pressure data by using the pressure evaluation model, and control the shield tunneling machine to adjust the inner and outer sides in the curve tunneling stage.

[0094] The above embodiments are only used to illustrate the technical method of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A cloud prediction-based shield pressure maintaining regulation system, characterized in that, The method comprises the following modules: a geological modeling module, configured to acquire geological feature information of a construction area, and to construct a three-dimensional geological model by using the geological feature information; a tunneling simulation module, configured to acquire control parameters and attitude parameters of a shield machine in a curve tunneling stage, and to simulate a tunneling process of the shield machine in the three-dimensional geological model to acquire a three-dimensional simulation model; a volume prediction module, configured to acquire an overbreak area and an overbreak volume of the shield machine in the curve tunneling stage in the three-dimensional simulation model, and to construct a volume prediction model in combination with equipment parameters of the shield machine; a pressure evaluation module, configured to acquire historical overbreak volumes by using the volume prediction model, to acquire historical internal pressure data and historical external pressure data of the shield machine in different curve tunneling stages, and to construct a pressure evaluation model in combination with the geological feature information according to different historical overbreak volumes, historical internal pressure data and historical external pressure data; a pressure adjustment module, configured to acquire a predicted overbreak volume by using the volume prediction model, to acquire predicted internal pressure data and predicted external pressure data by using the pressure evaluation model, and to control the shield machine to adjust internal and external pressures in the curve tunneling stage.

2. The cloud prediction-based shield pressure maintaining regulation system according to claim 1, wherein, The process of acquiring the geological feature information and constructing the three-dimensional geological model comprises: The geological feature information refers to rock types, geological structures, hydrological conditions, topographic features and stratigraphic sequences of the construction area, a three-dimensional structure model is constructed according to the topographic features and the stratigraphic sequences by using a three-dimensional modeling software, and the geological structures, the rock types and the hydrological conditions are imported into the three-dimensional structure model to acquire the three-dimensional geological model.

3. The cloud prediction-based shield pressure maintaining regulation system according to claim 2, wherein, The process of acquiring the control parameters and the attitude parameters and simulating the three-dimensional simulation model comprises: The curve tunneling stage refers to a construction stage in which the shield machine tunnels along a preset curve route, the control parameters include a pushing speed, a total thrust, a cutter head rotating speed and a cutter head torque, and the attitude parameters include an azimuth angle, a pitch angle, a roll angle and a shield tail gap; Equipment parameters of the shield machine are acquired, the equipment parameters include geometrical shapes and sizes of different components of the shield machine, and a three-dimensional physical model of the shield machine is constructed according to the equipment parameters; The three-dimensional physical model of the shield machine is imported into the three-dimensional geological model by using a simulation software, and the three-dimensional physical model is controlled to tunnel according to the control parameters and the attitude parameters to acquire the three-dimensional simulation model.

4. The cloud prediction-based shield pressure maintaining regulation system according to claim 3, wherein, The process of acquiring the overbreak area and the overbreak volume and constructing the volume prediction model comprises: In the curve tunneling stage, a cutting range outside the curve route is actively expanded to form a gap, and a shield body is passively extruded against surrounding rocks inside the curve route; The overbreak area refers to a gap area formed in a preset fixed period after the cutting range is actively expanded by the shield machine, and the overbreak volume refers to a volume size of the overbreak area in the preset fixed period, and a region with increased soil stress due to extrusion is marked as a loss region; In the three-dimensional simulation model, the overbreak area and the overbreak volume formed by the shield machine in a single preset fixed period under different control parameters and attitude parameters are acquired, a volume prediction set is generated in combination with the equipment parameters of the shield machine, and the volume prediction set is divided into a first training set and a first test set; The first convolutional neural network is constructed, and the equipment parameters and different control parameters and attitude parameters in the first training set are taken as input data of the first convolutional neural network, and the corresponding overbreak volume in the first training set is taken as output data of the first convolutional neural network. The first convolutional neural network is trained by using the first training set to obtain an initial first convolutional neural network, the initial first convolutional neural network is verified by using the first test set, and the initial first convolutional neural network less than or equal to a preset first test error threshold is output as a volume prediction model.

5. The cloud prediction-based shield pressure maintaining regulation system according to claim 4, wherein, The process of obtaining historical overbreak volume, historical internal pressure data and historical external pressure data comprises: The control parameters and attitude parameters of the shield machine in the past curve tunneling stage are obtained, and the equipment parameters of the shield machine are input into the volume prediction model to obtain the historical overbreak volume in a single preset fixed period corresponding to the curve tunneling stage; The historical internal pressure data and the historical external pressure data corresponding to the historical overbreak volume in the single preset fixed period are obtained, the historical internal pressure data refers to the stratum soil pressure and pore water pressure on the inside of the curve line in the corresponding curve tunneling stage, and the historical external pressure data refers to the stratum soil pressure and pore water pressure on the outside of the curve line in the corresponding curve tunneling stage.

6. The cloud prediction-based shield pressure maintaining regulation system according to claim 5, wherein, The process of constructing the pressure evaluation model comprises: According to the historical overbreak volume, the historical internal pressure data, the historical external pressure data, the equipment parameters, the control parameters, the attitude parameters and the geological feature information in different preset fixed periods in different curve tunneling stages, a pressure evaluation set is generated, and the pressure evaluation set is divided into a second training set and a second test set; The second convolutional neural network is constructed, and the different historical overbreak volumes, the geological feature information, the equipment parameters, the control parameters and the attitude parameters in the second training set are taken as input data of the second convolutional neural network, and the corresponding historical internal pressure data and historical external pressure data in the second training set are taken as output data of the second convolutional neural network; The second convolutional neural network is trained by using the second training set to obtain an initial second convolutional neural network, the initial second convolutional neural network is verified by using the second test set, and the initial second convolutional neural network less than or equal to a preset second test error threshold is output as a pressure evaluation model.

7. The cloud prediction-based shield pressure maintaining regulation system according to claim 6, wherein, The process of adjusting the internal and external sides of the curve tunneling stage comprises: Before the shield machine enters the curve tunneling stage, the preset control parameters and attitude parameters for realizing the curve tunneling are obtained, and the equipment parameters of the shield machine are input into the volume prediction model to obtain the predicted overbreak volume in a single preset fixed period after the shield machine enters the curve tunneling stage; The predicted overbreak volume, control parameters, attitude parameters of the shield machine in a single preset fixed period after entering the curve tunneling stage are input into the pressure evaluation model combined with geological feature information and equipment parameter to obtain predicted internal pressure data P of the shield machine in the single preset fixed period after entering the curve tunneling stage n , predicted external pressure data P w ; For the outside of the curve line, inject the quick-setting slurry into the over-excavation area and obtain the injection pressure P out and the injection volume V out For the inside of the curve line, inject the slow-setting slurry into the loss area and obtain the injection pressure P in and the injection volume V in ; ； ； ； ； the preset correction coefficient, the value range of the preset correction coefficient is 0.03-0.05Mpa, the value range of the preset correction coefficient is 0.02-0.03Mpa, respectively the static earth pressure and the overburden pressure of the construction area, the value is 3.14, D is the diameter of the shield body, is the shield tail gap, L is the segment width; After the shield machine enters the curve tunneling stage, the predicted overbreak volume, the predicted internal pressure data and the predicted external pressure data of the shield machine in the next preset fixed period are obtained, and then the grouting pressure and the grouting amount of the outside of the curve line in the next preset fixed period and the grouting pressure and the grouting amount of the inside of the curve line in the next preset fixed period are obtained, and the internal and external sides of the curve tunneling stage are adjusted according to the two obtained grouting pressures and grouting amounts.

8. A cloud prediction-based shield pressure maintaining regulation method, implemented based on the shield pressure maintaining regulation system of any one of claims 1-7, characterized in that, The method comprises: Step S1: Obtain geological feature information of a construction area, and construct a three-dimensional geological model by using the geological feature information; Step S2: Obtain control parameters and attitude parameters of a shield machine in a curve tunneling stage, simulate a tunneling process of the shield machine in the three-dimensional geological model to obtain a three-dimensional simulation model; Step S3: Obtain an overbreak area and an overbreak volume of the shield machine in the curve tunneling stage in the three-dimensional simulation model, and construct a volume prediction model in combination with equipment parameters of the shield machine; Step S4: Obtain a historical overbreak volume by using the volume prediction model, obtain historical internal pressure data and historical external pressure data of the shield machine in different curve tunneling stages, and construct a pressure evaluation model in combination with the geological feature information according to different historical overbreak volumes, the historical internal pressure data and the historical external pressure data; Step S5: Obtain a predicted overbreak volume by using the volume prediction model, obtain predicted internal pressure data and predicted external pressure data by using the pressure evaluation model, and control the shield machine to adjust pressures on inner and outer sides of the curve tunneling stage, respectively.

Images