Shield segment assembly error quantification and position prediction method, system and device

Abstract

The invention discloses a shield tunnel segment assembly error quantification and segment position prediction method, and relates to the technical field of shield subway tunnel construction. The position deviation of a point B is calculated through coordinate data of a central position A of a segment ring where a laser target is located, the distance of a central point B of a segment ring on the outermost side and the position of the laser target actually measured on site in combination with the horizontal and vertical coordinate orientation of the laser target, and then the measured data is compared with the data of the actual position of the segment. And central position data and error rules of the segment assembly rings under different tunneling stratums are obtained through calculation. Error correction is carried out by combining data comparison of a trial tunneling section, a segment position prediction model is established through an artificial neural network algorithm, and the relative position of a next segment is determined. According to the method, the relative position relation between the duct piece position and the tunnel line type can be judged in the tunneling process, the relative position of the duct piece of the next ring can be predicted, errors in the duct piece splicing process can be remarkably reduced, and the splicing quality and the forming quality of the shield tunnel duct pieces are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield subway tunnel construction, in particular to a shield segment assembly error quantification and position prediction method, system and device. BACKGROUND

[0002] The core of tunnel construction quality lies in the accurate assembly of segments, which directly determines the integrity and durability of the tunnel structure. In traditional shield tunnel construction, segment assembly usually relies on the propulsion cylinder, excavation parameters and the positioning system of the shield machine. These methods are often passive and lack real-time performance.

[0003] The overall structure of the shield machine is complex, the propulsion cylinder has a wide distribution range, and the connection between the cylinder and the corresponding sensor is complex, which can easily cause segment assembly errors. At the same time, due to manual operation, there are certain subjective factors, and the position of the segment cannot be determined before assembly, and the deviation caused by shield excavation and assembly process is prone to cause segment linear quality problems such as misalignment, leakage, damage and cracking. These problems not only increase the difficulty and cost of post-repair, but also pose a serious threat to construction safety and engineering quality.

[0004] In addition, the physical and mechanical properties of different strata differ greatly, and have a significant impact on segment assembly errors. However, the existing technology lacks a mechanism to effectively combine stratum parameters with error calculation, resulting in limited error calculation accuracy. At the same time, when errors occur abnormally, there is no perfect early warning and automatic adjustment mechanism, making it difficult to quickly respond and handle error problems.

[0005] Through the above analysis, the use of existing technology for shield tunnel construction faces many challenges: low construction efficiency, high cost, and difficult to guarantee quality. Especially in complex strata, how to accurately control the assembly position of the segment, reduce errors and improve construction accuracy has become a technical problem to be solved. SUMMARY

[0006] The technical problem to be solved by the present application is that the construction accuracy of the prior art is low and the assembly position of the segment is not accurate. The purpose is to provide a shield segment assembly error quantification and position prediction method, system and device, which can more accurately guide the assembly of the segment, has the beneficial effects of accurate prediction, high construction efficiency and high construction quality.

[0007] The present application is achieved by the following technical solutions: The first invention, the present application provides a shield segment assembly error quantification and position prediction method, which comprises, S1, the shield posture condition is monitored in real time by the guide system laser target in the shield machine excavation process; S2, coordinate data (xA, yA) of a center position A of a segment ring at a position of a laser target is acquired; S3, a distance L between the laser target position and a center point B of an outermost segment ring is acquired; S4, a position deviation (xB, yB) of the B point is calculated in combination with a horizontal trend (kx) and a vertical trend (ky) measured by the shield machine, and error data (Δx, Δy) is obtained by comparing the calculated data with position data (x1, y1) of the re-measured segment of the measurement group; wherein: xB=xA+L×kx, yB=yA+L×ky; Δx=|x1-xB|; Δy=|y1-yB|; S5, the center position data of the segment assembling ring and the error variation law under different tunneling strata are acquired by summarizing a plurality of (Δx, Δy) data; S6, a segment position prediction model is established through an artificial neural network algorithm, and the prediction model is continuously optimized by tracking the attitude data of the segment in the entire tunneling process to determine the relative position of the next ring segment.

[0008] Further, in the present application, in the above S4 step, a stratum parameter dynamic correction coefficient (λ) is introduced, the stratum parameter dynamic correction coefficient (λ) is determined based on the hardness, water content and porosity of the stratum, and is included in the B point position deviation calculation, and the corrected B point coordinate calculation formula is: xB=xA+L×kx×λ, yB=yA+L×ky×λ; and in the S5 step, an error early warning mechanism is established, when the error data (Δx, Δy) exceeds a preset threshold value, the early warning is automatically triggered and an error adjustment scheme is generated, and the preset threshold value is dynamically set according to the error variation law of different strata.

[0009] Further, in the present application, the hardness, water content and porosity data of different strata are acquired through indoor tests, the weights of the parameters are determined by using the analytic hierarchy process, and the stratum parameter dynamic correction coefficient (λ) is obtained by weighted calculation, and the specific calculation formula is λ=α×H+β×W+γ×P, wherein H is a stratum hardness parameter, W is a stratum water content parameter, P is a stratum porosity parameter, α, β and γ are weights of the respective parameters, and α+β+γ=1.

[0010] Further, in the present application, in the above error early warning mechanism, the dynamic setting mode of the preset threshold value is: based on the error variation law under different strata, the initial threshold value corresponding to each stratum is determined by using a statistical analysis method, and in the tunneling process, the threshold value is dynamically adjusted by an exponential smoothing method according to the deviation degree of the real-time error data and the initial threshold value.

[0011] Further, in the present application, in the above step S1, the data of the laser target is used for real-time monitoring of the shield attitude including the deviations in the horizontal and vertical directions, and in the step S6, the artificial neural network algorithm is a neuron Ai in the artificial neural network.

[0012] Further, in the present application, the artificial neural network algorithm comprises an input layer, a hidden layer and an output layer; The input layer vector is represented as X=(x1, x2, …, xk, …, xn)T, the hidden layer output vector is represented as Y=(y1, y2, …, yk, …, yn)T, and the output layer output vector is represented as O=(o1, o2, …, ok, …, on)T. The weight matrix between the input layer and the hidden layer is represented as V=(v1, v2, …, vk, …, vn) and W=(ω1, ω2, …, ωk, …, ωn), respectively, wherein vk and ωk represent the weight value of the kth neuron. The mathematical relationship of the hidden layer is wherein vij is the connection weight between the i th neuron of the input layer and the j th neuron of the hidden layer, xi is the input value of the i th neuron of the input layer, yj is the output value of the j th neuron of the hidden layer, and j is the number of the neuron of the hidden layer. The mathematical relationship of the output layer is

[0013] wherein ωjk is the connection weight between the j th neuron of the hidden layer and the k th neuron of the output layer, yj is the output value of the j th neuron of the hidden layer, ok is the output value of the k th neuron of the output layer, and k is the number of the neuron of the output layer; wherein f1(x) and f2(x) represent the transfer functions of the output layer and the hidden layer, respectively.

[0014] Further, in the present application, the input layer receives the input information (x0…xn) from the superior neurons and outputs Oi after processing the input information in the following manner:

[0015] wherein Ti represents the threshold value of the artificial neuron Ai, ωki represents the connection weight between the neuron Ak and Ai, and f(x) is the neuron transfer function.

[0016] In a second aspect, the present application further provides a shield segment assembling error quantification and position prediction system which adopts the shield segment assembling error quantification and position prediction method and comprises: (1) a posture monitoring module for monitoring the shield posture in real time through the laser target of the guiding system in the shield tunneling process; (2) a coordinate calculation module for calculating the coordinate data of the center position A of the segment ring where the laser target is located; (3) a distance measurement module for obtaining the distance between the laser target position and the center point B of the outermost segment ring; ​​​(4) an error calculation module, configured to calculate the position deviation of the B point in combination with the horizontal trend and the vertical trend measured by the shield machine, and to compare the position data of the re-measured segment to obtain error data; (5) a rule summary module, configured to summarize a plurality of error data to obtain the center position data of the segment assembly ring in different tunneling strata and error variation rules; (6) a prediction model module, configured to establish a segment position prediction model through an artificial neural network algorithm, and to continuously track the attitude data of the segment to optimize the prediction model; (7) a stratum parameter correction module, configured to obtain the hardness, water content and porosity data of the stratum, to calculate a stratum parameter dynamic correction coefficient (λ), and to apply the stratum parameter dynamic correction coefficient (λ) to the correction calculation of the position deviation of the B point; (8) an early warning and adjustment module, configured to set an error preset threshold, to issue a warning when the error data exceeds the threshold, and to generate an error adjustment scheme.

[0017] Further, in the present application, the above further includes a database configured to store the attitude monitoring data, coordinate calculation data, distance measurement data, error calculation data, rule summary data, prediction model data, stratum parameter data and early warning adjustment data of the shield machine.

[0018] In a third aspect, the present application further provides a shield segment assembly error quantification and position prediction device, which includes a shield segment assembly error quantification and position prediction system.

[0019] Compared with the prior art, the present application has the following advantages and beneficial effects: The present application introduces a stratum parameter dynamic correction coefficient (λ), fully considers the influence of different strata on segment assembly error, and by including key parameters such as the hardness, water content and porosity of the stratum in the B point position deviation calculation, makes the error calculation more accurate, and overcomes the problem of inaccurate error calculation caused by ignoring stratum differences in traditional methods.

[0020] The present application establishes an error early warning mechanism, which can issue a warning in time when the error exceeds the preset threshold, and automatically generate an adjustment scheme, so that the construction personnel can quickly take measures to correct, and avoid the accumulation of errors to seriously affect the quality of tunnel construction. At the same time, the preset threshold is dynamically set according to the error variation rules of different strata, which improves the pertinence and accuracy of the early warning.

[0021] The shield attitude can be monitored in real time by using the laser guiding system, the segment assembly error and variation law can be effectively calculated and analyzed in the shield tunneling construction process, and the segment assembly position of the next ring segment can be determined through the formula and real-time monitoring data. By reducing the segment assembly error and optimizing the segment selection, the construction efficiency can be significantly improved, the construction cost can be reduced, and the labor input of the workers can be reduced. Through accurate segment assembly position prediction and error control, the overall forming quality of the tunnel can be effectively guaranteed, and the need for post-repair and processing can be reduced.

[0022] Compared with the traditional prediction method, the artificial neural network algorithm is applied to establish the segment position prediction model, the prediction accuracy can be continuously optimized according to the real-time monitoring data. With the progress of the shield tunneling, the model can continuously absorb new error data, and self-learning and adaptation can be performed to correct and optimize the subsequent prediction data. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the example embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor. In the drawings: Figure 1 The method flowchart of the embodiment 1 of the present application; Figure 2 The ideal structure schematic diagram of the present application that the center point B of the outermost segment ring coincides with the center A of the segment ring where the laser target is located; Figure 3 The ideal structure schematic diagram of the present application that the center point B of the outermost segment ring does not coincide with the center A of the segment ring where the laser target is located; Figure 4 The geometric relationship schematic diagram of the position deviation X direction of the point B of the present application; Figure 5 The geometric relationship schematic diagram of the position deviation Y direction of the point B of the present application; Figure 6 The horizontal and vertical deviation diagram of the shield segment of the present application with the increase of the mileage; Figure 7 The schematic diagram of the artificial neuron receiving and processing information of the present application; Figure 8 The comparison diagram of the training and prediction structure of the present application; Figure 9 The stratum parameter dynamic correction coefficient calculation flowchart of the embodiment 2 of the present application; Figure 10 The error early warning mechanism working flowchart of the embodiment 2 of the present application. DETAILED DESCRIPTION

[0024] In order to make the objects, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and drawings. The schematic embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention. The following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] Example 1, the overall process of the present invention is as follows Figure 1 As shown in the figure, in the early stage of the shield tunneling process, the attitude data of the center position A of the segment ring where the laser target is located is monitored in real time through the laser target of the guidance system in the shield machine. Then, based on the distance L between the laser target position and the center point B of the outermost segment ring measured on site, the position deviation of point B is calculated in combination with the horizontal and vertical coordinate orientations of the shield machine itself. The measured data is compared with the data of the actual position of the segment, and the center position data and error law of the segment assembly ring in different excavation strata are summarized. Combined with the comparison error of the trial excavation section, correction is made, and the error data and change law of the segment assembly in different strata can be obtained. The relative position of the next ring of segments can be determined, providing a basis for segment selection.

[0026] like Figure 2 As shown, the left side shows the excavation section of the shield machine. During shield construction, the ideal positional relationship between the segment and the shield tail is when the center point B of the outermost segment ring coincides with the center point A of the segment ring where the laser target is located. However, in actual construction, the shield machine needs to excavate along the design axis. The shield tail and the center-front shield often form a "V" shape, meaning that the centerline of the shield tail does not coincide with the center point A of the segment ring where the laser target is located. In addition, during excavation, the shield machine corrects deviations to fit the design axis, resulting in random positions of the center points of each ring. Therefore, in actual construction, it is necessary to calculate the positional deviation of point B based on the horizontal and vertical coordinates of the shield machine itself to determine the horizontal or vertical positional relationship between the segment and the shield tail.

[0027] like Figure 3 As shown, the center point B of the outermost segment ring does not coincide with the center A of the segment ring where the laser target is located. At this time, it is necessary to analyze the posture data of the center position A of the segment ring where the laser target is located, and measure the distance L between the laser target position and the center point B of the outermost segment ring. Then, the position deviation of the center point B of the outermost segment ring is obtained through the trend of horizontal deviation and vertical deviation.

[0028] For example, after the completion of a certain ring advance, according to the laser measurement data, the front shield deviation is 0 mm, the middle shield deviation is 8 mm, and the rear shield deviation is 14 mm in the horizontal position, and the trend is -2.

[0029] Then the distance L between the laser target position measured on site and the outermost tube ring center point B is 10 mm.

[0030] From Figure 4 It can be known that xB=xA+L×kx=14+10×(-2)=-6 mm, which represents that the position deviation of the outermost tube ring center point B relative to the tube ring center position A of the laser target is 20 mm.

[0031] After the completion of a certain ring advance, according to the laser measurement data, the front shield deviation is -6 mm, the middle shield deviation is -24 mm, and the rear shield deviation is -30 mm in the vertical position, and the trend is 4.

[0032] Similarly, from Figure 5 It can be known that yB=yA+L×ky=-30+10×(4)=10 mm, which represents that the position deviation of the outermost tube ring center point B relative to the tube ring center position A of the laser target is 40 mm.

[0033] According to the above calculation process, the position deviation of point B is (-6, 10).

[0034] The position data of the tube piece retested by the measurement group is (-10, 20).

[0035] The error data is obtained by comparing the calculation data with the position data (x1, y1) of the tube piece retested by the measurement group.

[0036] The error data quantization formula is Δx=|x1-xB|; Δy=|y1-yB|.

[0037] Δx=|x1-xB|=|-10+6|=4 mm.

[0038] Δy=|y1-yB|=|10-20|=10 mm.

[0039] The application of the present application to a certain bid collects data as shown in Figure 6 According to the above method, the horizontal and vertical deviations (Δx, Δy) of the shield process posture data with the mileage increase are calculated, and the data is statistically analyzed through an EXCEL table.

[0040] As shown in Figure 7 The input information (x0…xn) from the superior neurons is received by the neurons Ai in the artificial neural network, and the output information , wherein: .

[0041] wherein Ti, Tj represent the threshold value of artificial neurons Ai and Aj, ωki and ωij represent the connection relationship between neurons Ak and Ai, Ai and Aj (here, the relative weight of the structural defect index) respectively, and f(x) is the neuron transfer function.

[0042] Through the special connection between artificial neurons, an artificial neural network composed of multiple network structures can be constructed, and through the identification processing of the input layer, hidden layer and output layer, complex real problems affected by multiple factors can be solved. The input layer vector can be represented as X = (x1, x2, …, xk, …, xn)T, the hidden layer output vector is represented as Y = (y1, y2, …, yk, …, yn)T, the output layer output vector is represented as O = (o1, o2, …, ok, …, on)T, the expected output vector is P = (p1, p2, …, pk, …, pn)T, and the weight matrix between the input layer and the hidden layer is represented as V = (v1, v2, …, vk, …, vn) and W = (ω1, ω2, …, ωk, …, ωn) respectively, wherein vk and ωk represent the weight value of the kth neuron.

[0043] The mathematical relationship of the output layer is .

[0044] The mathematical relationship of the hidden layer is .

[0045] wherein f1(x) and f2(x) represent the transfer functions of the output layer and the hidden layer respectively, also known as the activation function.

[0046] The segment position calculation formula is established by using a programming language, and the relative position of the next ring segment can be determined by real-time monitoring data through the guidance system.

[0047] The first 60 groups of data of the above segment quantitative error are selected and input into the software Matlab for calculation.

[0048] As shown in Figure 8 , the comparison of the prediction results of the training set can reflect that the true value and the predicted value are not much different, proving that the prediction model is suitable for segment position prediction.

[0049] At the same time, as the shield tunneling, new error data will also be generated, which will be added to the database and combined with the algorithm to continuously optimize the original model and correct the subsequent prediction data.

[0050] After the implementation of the above calculation method, the positional relationship and error of the shield segment linear point in the vertical and horizontal directions can be calculated. Combined with the segment position deviation law, the segment position prediction model and the subsequent correction data, the next ring segment assembly position can be determined, and the segment selection can be optimized.

[0051] The shield segment assembly error quantification and position prediction method of embodiment 2 is further optimized on the basis of embodiment 1, introduces a stratum parameter dynamic correction coefficient and establishes an error warning mechanism.

[0052] The calculation process of the stratum parameter dynamic correction coefficient (λ) is shown in FIG. 2. First, the hardness, water content and porosity data of the stratum are obtained through indoor tests, then the weights of the parameters are determined by using the analytic hierarchy process, and finally the stratum parameter dynamic correction coefficient (λ) is calculated according to the formula λ = α × H + β × W + γ × P. Figure 9 For example, for a certain sand layer, H = 0.6, W = 0.3 and P = 0.2 are measured by tests, α = 0.5, β = 0.3 and γ = 0.2 are determined by using the analytic hierarchy process, then λ = 0.5 × 0.6 + 0.3 × 0.3 + 0.2 × 0.2 = 0.3 + 0.09 + 0.04 = 0.43.

[0053] The stratum parameter dynamic correction coefficient (λ) is applied to the calculation of the position deviation of point B,

[0054] For example: after the completion of a certain ring, the laser measurement data show that the deviation of the front shield is 0 mm, the deviation of the middle shield is 8 mm, the deviation of the rear shield is 14 mm, the trend is -2, the distance L between the measured laser target position and the center point B of the outermost segment ring is 10 mm, and the stratum parameter dynamic correction coefficient λ = 0.43, then the corrected xB = xA + L × kx × λ = 14 + 10 × (-2) × 0.43 = 14 - 8.6 = 5.4 mm, after the completion of a certain ring, the laser measurement data show that the deviation of the front shield is -6 mm, the deviation of the middle shield is -24 mm, the deviation of the rear shield is -30 mm, the trend is 4, L = 10 mm, and λ = 0.43, the corrected yB = yA + L × ky × λ = -30 + 10 × 4 × 0.43 = -30 + 17.2 = -12.8 mm, According to the above calculation process, the position deviation of the corrected point B is (5.4, -12.8), The position data of the retested segment of the measurement group is (5.0, -13.0), The error data obtained by comparing the calculation data with the position data (x1, y1) of the retested segment of the measurement group is (Δx, Δy), The error data quantification formula is Δx = |x1 - xB|; Δy = |y1 - yB|, Δx = |5.0 - 5.4| = 0.4 mm, Δy = |-13.0 - (-12.8)| = 0.2 mm, Δx = |5.0 - 5.4| = 0.4 mm, It can be seen that after introducing the dynamic correction coefficient of formation parameters, the error data is smaller and the calculation results are more accurate.

[0055] The workflow of the error warning mechanism is as follows: Figure 10 As shown in the figure, based on the error variation patterns in different strata, statistical analysis methods are used to determine the initial thresholds corresponding to each stratum. For example, for clay layers, the initial thresholds for the horizontal direction are set at 5mm, and the initial thresholds for the vertical direction are set at 4mm. During the excavation process, error data is monitored in real time. When the horizontal error Δx = 6mm exceeds the initial threshold of 5mm, the threshold is dynamically adjusted using exponential smoothing. The adjusted horizontal threshold may be 5.5mm, and an early warning is issued.

[0056] In Example 3, the present invention provides a shield segment assembly error quantification and position prediction system, which adopts the shield segment assembly error quantification and position prediction method of Example 2, including: (1) Attitude monitoring module, which is used to monitor the attitude of the shield machine in real time through the laser target of the guidance system during the tunneling process; (2) Coordinate calculation module, used to calculate the coordinate data of the center position A of the segment ring where the laser target is located; (3) Distance measurement module, used to obtain the distance between the laser target position and the center point B of the outermost segment ring; (4) Error calculation module, which is used to calculate the position deviation of point B by combining the horizontal trend and vertical trend measured by the shield machine, and compare it with the position data of the re-measured segment to obtain the error data; (5) A rule summary module is used to summarize multiple error data and obtain the center position data of the segment assembly ring and the error change rules under different excavation strata; (6) Prediction model module, which is used to establish a segment position prediction model through artificial neural network algorithm and continuously track the segment posture data to optimize the prediction model; (7) Formation parameter correction module, which is used to obtain the hardness, water content and porosity data of the formation, calculate the dynamic correction coefficient (λ) of the formation parameters, and apply it to the correction calculation of the position deviation of point B; (8) Warning and adjustment module, which is used to set the preset error threshold, issue a warning when the error data exceeds the threshold, and generate an error adjustment plan.

[0057] The shield segment assembly error quantification and position prediction system also includes a database for storing the shield machine's posture monitoring data, coordinate calculation data, distance measurement data, error calculation data, rule summary data, prediction model data, formation parameter data, and early warning adjustment data.

[0058] Embodiment 4, the application provides a shield segment assembly error quantification and position prediction device, which adopts the shield segment assembly error quantification and position prediction system of embodiment 2.

[0059] The above specific embodiments further explain the purpose, technical scheme and beneficial effects of the application. It should be understood that the above description is only a specific embodiment of the application and is not used to limit the protection scope of the application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A method for quantifying and predicting the position of a shield segment assembly error, characterized in that, Comprising, S1, monitoring the shield posture in real time through the laser target of the guiding system in the tunneling process of the shield machine; S2, obtaining coordinate data (xA, yA) of the center position A of the segment ring at the position of the laser target; S3, obtaining the distance L between the position of the laser target and the center point B of the outermost segment ring; S4, combining the horizontal trend kx and the vertical trend ky measured by the shield machine to calculate the position deviation (xB, yB) of the point B, comparing the calculation data with the position data (x1, y1) of the re-measured segment of the measurement group to obtain error data (Δx, Δy); Wherein: xB=xA+L×kx, yB=yA+L×ky; Δx=|x1-xB|; Δy=|y1-yB|; S5, summarizing multiple (Δx, Δy) data to obtain the center position data of the segment assembly ring in different tunneling strata and the error variation law; S6, establishing a segment position prediction model through an artificial neural network algorithm, and continuously tracking the posture data of the segment in the entire tunneling process to optimize the prediction model and determine the relative position of the next segment. 2.The shield segment assembly error quantification and position prediction method of claim 1, wherein, In the S4 step, a stratum parameter dynamic correction coefficient λ is introduced, which is determined based on the hardness, water content and porosity of the stratum, and is included in the calculation of the position deviation of the point B. The corrected B point coordinate calculation formula is: xB=xA+L×kx×λ, yB=yA+L×ky×λ; And in the S5 step, an error warning mechanism is established. When the error data (Δx, Δy) exceeds the preset threshold value, the error adjustment scheme is automatically triggered and an error warning is generated. The preset threshold value is dynamically set according to the error variation law of different strata. 3.The shield segment assembly error quantification and position prediction method of claim 2, wherein, The hardness, water content and porosity data of different strata are obtained through indoor tests, the weights of each parameter are determined by the analytic hierarchy process, and the stratum parameter dynamic correction coefficient λ is calculated by weighting. The specific calculation formula is λ=α×H+β×W+γ×P, wherein H is the stratum hardness parameter, W is the stratum water content parameter, P is the stratum porosity parameter, α, β and γ are the weights of each parameter, and α+β+γ=1. 4.The shield segment assembly error quantification and position prediction method of claim 2, wherein, In the error warning mechanism, the dynamic setting method of the preset threshold value is: based on the error variation law of different strata, the initial threshold value corresponding to each stratum is determined by statistical analysis method. In the tunneling process, the threshold value is dynamically adjusted by the exponential smoothing method according to the deviation degree of the real-time error data and the initial threshold value. 5.The shield segment assembly error quantification and position prediction method of claim 1, wherein, In the step S1, the data of the laser target is used to monitor the shield posture in real time, including the deviation in the horizontal and vertical directions. In the step S6, the artificial neural network algorithm is a neuron Ai in the artificial neural network.

6. The shield segment assembly error quantification and position prediction method of claim 5, wherein, The artificial neural network algorithm includes an input layer, a hidden layer and an output layer; The input layer vector is represented as X=(x1, x2, …, xk, …, xn)T, the hidden layer output vector is represented as Y=(y1, y2, …, yk, …, yn)T, and the output layer output vector is represented as O=(o1, o2, …, ok, …, on)T, and the weight matrices between the input layer and the hidden layer are represented as V=(v1, v2, …, vk, …, vn) and W=(ω1, ω2, …, ωk, …, ωn) respectively, wherein vk and ωk represent the weight values of the kth neuron; The mathematical relationship of the hidden layer is , wherein vij is the connection weight between the i th neuron of the input layer and the j th neuron of the hidden layer, xi is the input value of the i th neuron of the input layer, yj is the output value of the j th neuron of the hidden layer, and j is the number of the neuron of the hidden layer; The mathematical relationship of the output layer is where ωjk is the connection weight between the jth neuron of the hidden layer and the kth neuron of the output layer, yj is the output value of the jth neuron of the hidden layer, ok is the output value of the kth neuron of the output layer, and k is the number of the neuron of the output layer. wherein f1(x) and f2(x) represent the transfer functions of the output layer and the hidden layer respectively.

7. The shield segment assembly error quantification and position prediction method of claim 6, wherein, The input layer receives input information (x0…xn) from the superior neurons and outputs Oi after processing the input information in the following manner: ， wherein Ti represents the threshold value of the artificial neuron Ai, ωki represents the connection weight between the neuron Ak and the neuron Ai, and f(x) is the neuron transfer function.

8. A shield segment assembly error quantification and position prediction system, characterized in that: The shield segment assembly error quantification and position prediction method of claim 2 comprises: (1) a posture monitoring module for monitoring the shield posture in real time through the guide system laser target in the shield tunneling process; (2) a coordinate calculation module for calculating the coordinate data of the center position A of the segment ring at the position of the laser target; (3) a distance measurement module for obtaining the distance between the position of the laser target and the center point B of the outermost segment ring; (4) an error calculation module for calculating the position deviation of the point B in combination with the horizontal trend and the vertical trend measured by the shield machine, and comparing the position data of the re-measured segment to obtain error data; (5) a rule summary module for summarizing a plurality of error data to obtain the center position data of the segment assembly ring and the error variation law under different tunneling strata; (6) a prediction model module for establishing a segment position prediction model through an artificial neural network algorithm, and continuously tracking the posture data of the segment to optimize the prediction model; (7) a stratum parameter correction module for obtaining the hardness, water content and porosity data of the stratum, calculating a dynamic correction coefficient λ of the stratum parameters, and applying the dynamic correction coefficient λ to the correction calculation of the position deviation of the point B; (8) a warning and adjustment module for setting an error preset threshold, issuing a warning when the error data exceeds the threshold, and generating an error adjustment scheme.

9. The system according to claim 8, wherein: Further comprising a database for storing the posture monitoring data, coordinate calculation data, distance measurement data, error calculation data, rule summary data, prediction model data, stratum parameter data and warning adjustment data of the shield machine.

10. A shield segment assembly error quantification and position prediction device, characterized by, The shield segment assembly error quantification and position prediction system of claim 9.

Images