A cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit

Abstract

The present invention discloses a cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit. S1: Obtain the original displacement data and historical influence data based on the influence factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall; S2: Obtain the trained original inclination displacement prediction model; S3: Obtain the trained disturbance inclination displacement prediction model; S4: Realize the displacement prediction of the foundation pit side wall based on the total displacement prediction model. The present invention uses the original inclination displacement prediction model to calculate the general situation of the on-site foundation pit displacement, and then uses the disturbance inclination displacement prediction model to predict the comprehensive influence of subsequent external influence factors on the foundation pit displacement, so as to scientifically and effectively avoid the occurrence of over-protection and collapse accidents commonly existing in the foundation pit construction process, save the manpower and material resources of on-site construction, ensure the safety of on-site construction, and better meet the actual application requirements of the construction site.

Description

Technical Field

[0001] The present invention relates to the technical field of foundation pit stability prediction, and particularly to a cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit. Background Art

[0002] At present, in the field of foundation pit construction, the analysis and prediction of the inclination deformation displacement of the foundation pit side wall have become one of the important bases for judging the stability of the construction results. Most of the existing analysis methods evaluate the stability of the foundation pit based on the on-site soil deformation detection results combined with artificial experience. This method not only costs a lot of manpower and material resources, but also the accuracy of the evaluation is not satisfactory, and ultimately may lead to overprotection or collapse accidents.

[0003] In fact, the deformation of the foundation pit side wall is formed under the combined action of the basic geological conditions and various external influencing factors (such as excavation depth (A), groundwater level (B), vibration influence (C), elastic modulus of rock and soil (D), temperature (E), and slope top load (F), etc.). Under the action of the basic geological conditions, it shows the primary displacement that changes with time, reflecting the main law of deformation, which is a non-stationary sequence; under the action of various influencing factors, it shows the disturbance displacement that changes randomly, which is a stationary noise sequence. The existence of this sequence will reduce the authenticity and accuracy of the prediction. With the development of foundation pit engineering towards larger, deeper, and more complex trends, the support technology is becoming more and more difficult, and the impact of foundation pit deformation on the surrounding environment is also more prominent. In the existing models during the prediction process, the interference of external influencing factors on the prediction results is often ignored, resulting in the prediction results being difficult to meet the actual engineering requirements. Summary of the Invention

[0004] The present invention provides a cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit to overcome the technical problem that the existing technology is prone to ignore the interference of external influencing factors on the prediction results, resulting in inaccurate prediction results for the deformation of the foundation pit side wall.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] A cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit, the specific steps include:

[0007] S1: Obtain the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall; and obtain the primary displacement data and historical influence data based on the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall, where the historical influence data includes disturbance displacement data and the influencing factor data of the inclination degree of the foundation pit side wall;

[0008] S2: Establish a primary inclination displacement prediction model, and train it based on the primary displacement data to obtain a trained primary inclination displacement prediction model;

[0009] S3: Establish a prediction model for disturbed tilt displacement considering the cross-time-domain influence characteristics of external factors, and train it based on historical influence data to obtain a trained prediction model for disturbed tilt displacement;

[0010] S4: Obtain a total displacement prediction model based on the trained original tilt displacement prediction model and the trained prediction model for disturbed tilt displacement, and realize the prediction of the displacement of the foundation pit side wall based on the total displacement prediction model.

[0011] Further, in S3, when establishing a prediction model for disturbed tilt displacement considering the cross-time-domain influence characteristics of external factors, the prediction model for disturbed tilt displacement includes formulas (1) to (7), and the process of training it based on historical influence data is as follows:

[0012] S31: Divide the historical influence data at equal intervals according to the time series of foundation pit deformation; and divide the equally spaced historical influence data into several calculation intervals according to the detection time;

[0013] S32: Substitute the historical influence data P of the H-th calculation interval H into the screening sequence formula to obtain the screened data Q H , expressed as:

[0014] Q H = X(M Q × U H-1 + N Q × P H + G Q ) (1)

[0015] where X is the sigmoid activation function, M Q and N Q are both coefficients of the linear relationship of the screening sequence, G Q is the error of the linear relationship of the screening sequence; U H-1 is an abstract expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0016] S33: Substitute the historical influence data P of the H-th calculation interval H into the entrance sequence formula, expressed as:

[0017] Y H = X(M Y × U H-1 + N Y × P H + G Y ) (2)

[0018] V H = tanh(M V×U H-1 +N V ×P H +G V ) (3)

[0019] In the formula, Y H and V H are the data that the entrance sequence can be optimally stored in the model for use, X is the sigmoid activation function, M Y , N Y , M V and N V are the coefficients of the linear relationship of the entrance sequence, G Y and G V are the errors of the linear relationship of the entrance sequence; U H-1 is the abstract expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0020] S34: Optimize the output result V H-1 in the (H - 1)-th calculation interval based on formulas (1), (2) and (3). The optimization formula is:

[0021] V H =V H-1 ×Q H +Y H ×V H (4)

[0022] S35: Substitute the data optimized by formula (4) and the historical influence data P H in the H-th calculation interval into the exit function formula, which is expressed as:

[0023] I H =X(M I ×U H-1 +N I ×P H +G I ) (5)

[0024] U H =I H ×tanh(V H ) (6)

[0025] In the formula, I H is the current data result output by the exit sequence, X is the sigmoid activation function, M I and N I are the coefficients of the linear relationship of the exit sequence, G I is the error of the linear relationship of the exit sequence, U H is the abstract expression matrix of the inclination degree of the foundation pit side wall in the H-th calculation interval; U H-1is an abstract expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0026] S36: Substitute U H into the result output formula to obtain the predicted disturbance displacement of the inclination degree of the foundation pit side wall on the (H + 1)-th day, expressed as:

[0027] Z * （H+1） = softmax(M S × U H + G S ) (7)

[0028] In the formula, M S is the relationship coefficient of the result output formula, and G S is the error of the result output linear relationship.

[0029] Furthermore, in S2, the established original inclination displacement prediction model is:

[0030] O H = X(M O × W H + G O ) (8)

[0031] W * （H+1） = softmax(M W × O H ) (9)

[0032] In the formula, O H is the model optimization representation coefficient, G O is the average error; W H is the original displacement data, and W * （H+1） is the predicted original displacement result on the (H + 1)-th day. M O and M W are the model relationship coefficients respectively.

[0033] Furthermore, in S1, the process of obtaining the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall, and obtaining the original displacement data and historical influence data based on the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall is:

[0034] Construct a monitoring data set J with a capacity of H, J = (J 1 , J 2 ...... J H );

[0035] where J 1 , J 2......J H represents the data of the inclined displacement of the foundation pit side wall and influencing factors for H days, and is not limited to consecutive H + 1 days;

[0036] J 1 =(P A 1 ,P B 1 ,P C 1 ,P D 1 ,P E 1 ,P F 1 ,S 1 ),

[0037] J 2 =(P A 2 ,P B 2 ,P C 2 ,P D 2 ,P E 2 ,P F 2 ,S 2 ), ...

[0038] J H =(P A H ,P B H ,P C H ,P D H ,P E H ,P F H ,S H );

[0039] P A H ,P B H ,P C H ,P D H ,P E H ,P F Hrespectively represent the excavation depth, groundwater level, vibration influence, elastic load of rock and soil mass, temperature, and the content of the influence factor of the slope top load on the H-th day, and are not limited to consecutive H days; S H is the data of the inclined displacement of the foundation pit side wall on the H-th day;

[0040] According to whether the data of the inclined displacement of the foundation pit side wall is affected by the influence factor of the inclination degree of the foundation pit, the data of the inclined displacement of the foundation pit side wall is divided into the original displacement data and the disturbed displacement data, and the disturbed displacement data and the data of the influence factor of the inclination degree of the foundation pit side wall are combined into the historical influence data.

[0041] Furthermore, the total displacement prediction model is expressed as:

[0042] S * (H+1) =W * (H+1) +Z * (H+1) (10)

[0043] In the formula, S * (H+1) is the total predicted displacement, and W * (H+1) is the original displacement predicted by the original inclined displacement prediction model; Z * (H+1) is the disturbed displacement predicted by the disturbed inclined displacement prediction model.

[0044] Beneficial effects: By decomposing the historical data of the inclined displacement of the foundation pit side wall, the present invention uses the original inclined displacement prediction model to calculate the general situation of the on-site foundation pit displacement, and then uses the disturbed inclined displacement prediction model to predict the comprehensive influence of external influence factors on the foundation pit displacement. Finally, the total displacement prediction model is used to obtain the accurate predicted value of the foundation pit deformation displacement. The present invention fully considers the interference of external influence factors on the prediction result, can scientifically and effectively avoid the occurrence of overprotection and collapse accidents commonly existing in the foundation pit construction process, save the manpower and material resources of on-site construction, and ensure the safety of on-site construction at the same time, which is more in line with the actual application requirements of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1It is a flowchart of a cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit in the present invention. Detailed implementation manners

[0047] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] This embodiment provides a cross-time-domain prediction method for the inclination trend of the side wall of a deep and large foundation pit, as Figure 1 shown, and the specific steps include:

[0049] S1: Obtain the data of the influencing factors of the inclination degree of the side wall of the foundation pit and the data of the inclination displacement of the side wall of the foundation pit; and obtain the original displacement data and historical influence data based on the data of the influencing factors of the inclination degree of the side wall of the foundation pit and the data of the inclination displacement of the side wall of the foundation pit, where the historical influence data includes disturbance displacement data and the data of the influencing factors of the inclination degree of the side wall of the foundation pit;

[0050] In a specific embodiment, in S1, the process of obtaining the data of the influencing factors of the inclination degree of the side wall of the foundation pit and the data of the inclination displacement of the side wall of the foundation pit; and obtaining the original displacement data and historical influence data based on the data of the influencing factors of the inclination degree of the side wall of the foundation pit and the data of the inclination displacement of the side wall of the foundation pit is as follows:

[0051] Construct a monitoring data set J = (J 1 , J 2 ...... J H ) with a capacity of H;

[0052] In this embodiment, the monitoring data set is sourced from the daily monitoring of the inclination displacement of the side wall of the foundation pit.

[0053] Among them, J 1 , J 2 ...... J H represents the inclination displacement of the side wall of the foundation pit and the influencing factor data for H days, and is not limited to consecutive H + 1 days;

[0054] J 1 =(P A 1 , P B 1 , P C 1 , P D 1 , P E 1,P F 1 ,S 1 ),J 2 =(P A 2 ,P B 2 ,P C 2 ,P D 2 ,P E 2 ,P F 2 ,S 2 ),...,J H =(P A H ,P B H ,P C H ,P D H ,P E H ,P F H ,S H );

[0055] P A H ,P B H ,P C H ,P D H ,P E H ,P F H respectively represent the excavation depth, groundwater level, vibration influence, elastic load of rock and soil mass, temperature, and the content of the influence factor of the slope top load on the H-th day, and are not limited to consecutive H days; S H is the data of the inclined displacement of the foundation pit side wall on the H-th day;

[0056] According to whether the data of the inclined displacement of the foundation pit side wall is affected by the influence factor of the inclination degree of the foundation pit, the data of the inclined displacement of the foundation pit side wall is divided into original displacement data and disturbed displacement data, and the disturbed displacement data and the data of the influence factor of the inclination degree of the foundation pit side wall are combined to form historical influence data.

[0057] S2: Establish an original inclined displacement prediction model, and train it based on the original displacement data to obtain the trained original inclined displacement prediction model;

[0058] In a specific embodiment, in S2, the established original inclined displacement prediction model is:

[0059] O H = X(M O × W H + G O ) (8)

[0060] W * （H+1） = softmax(M W × O H ) (9)

[0061] In the formula, O H is the model optimization representation coefficient, G O is the average error; W H is the original displacement data, and W * （H+1） is the predicted result of the original displacement on the (H + 1)-th day. M O and M W are the model relationship coefficients respectively.

[0062] S3: Establish a disturbance tilt displacement prediction model considering the cross-time domain influence characteristics of external factors, and train it based on historical influence data to obtain a trained disturbance tilt displacement prediction model;

[0063] In a specific embodiment, in S3, when establishing a disturbance tilt displacement prediction model considering the cross-time domain influence characteristics of external factors, the disturbance tilt displacement prediction model includes formulas (1) to (7), and the process of training it based on historical influence data is as follows:

[0064] S31: Divide the historical influence data at equal intervals according to the time series of foundation pit deformation; and divide the equally spaced historical influence data into several calculation intervals according to the detection time;

[0065] S32: Substitute the historical influence data P H of the H-th calculation interval into the screening sequence formula to obtain the screened data Q H , which is expressed as:

[0066] Q H = X(M Q × U H-1 + N Q × P H + G Q ) (1)

[0067] Among them, X is the sigmoid activation function, M Q and N Q are both the coefficients of the linear relationship of the screening sequence, G Q is the error of the linear relationship of the screening sequence; U H-1is an abstract expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0068] Specifically, in this embodiment, the calculation interval is divided to better train the model. The calculation interval is used to gradually update and optimize various parameters and gradually transfer the optimized parameters to subsequent calculation intervals to make various parameters reach the best. And during the process of optimizing the parameters of each calculation interval, it is necessary to first define predicting the next data from the previous several data, which requires dividing the input information and inputting it to different calculation intervals respectively. Therefore, the disturbance displacements input to each calculation interval are different.

[0069] S33: Substitute the historical influence data P of the H-th calculation interval H into the entrance sequence formula, expressed as:

[0070] Y H = X(M Y × U H-1 + N Y × P H + G Y ) (2)

[0071] V H = tanh(M V × U H-1 + N V × P H + G V ) (3)

[0072] In the formula, Y H and V H are the data that can be stored in the model in the best degree for the entrance sequence. X is the sigmoid activation function. M Y , N Y , M V and N V are the coefficients of the linear relationship of the entrance sequence. G Y and G V are the errors of the linear relationship of the entrance sequence. U H-1 is an abstract expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0073] S34: Optimize the output result V of the (H - 1)-th calculation interval based on formulas (1), (2) and (3). The optimization formula is: H-1 The optimization formula is:

[0074] V H = V H-1 × Q H + Y H × V H (4)

[0075] S35: Substitute the data optimized by formula (4) and the historical influence data P of the H-th calculation interval H into the outlet function formula, expressed as:

[0076] I H = X(M I ×U H-1 + N I ×P H + G I ) (5)

[0077] U H = I H × tanh(V H ) (6)

[0078] In the formula, I H is the current data result output by the outlet sequence, X is the sigmoid activation function, M I and N I are the coefficients of the linear relationship of the outlet sequence, G I is the error of the linear relationship of the outlet sequence, U H is the abstraction expression matrix of the inclination degree of the foundation pit side wall in the H-th calculation interval; U H-1 is the abstraction expression matrix of the inclination degree of the foundation pit side wall in the (H - 1)-th calculation interval;

[0079] Specifically, in this embodiment, the abstraction expression matrix of the initial inclination degree of the foundation pit side wall is a manually set value, and the subsequent is further optimized by formula (6).

[0080] S36: Substitute U H into the result output formula to obtain the predicted disturbance displacement of the inclination degree of the foundation pit side wall on the (H + 1)-th day, expressed as:

[0081] Z * （H+1） = softmax(M S ×U H + G S ) (7)

[0082] In the formula, M S is the relationship coefficient of the result output formula, G S is the error of the result output linear relationship.

[0083] Specifically, in this embodiment, the disturbance inclination displacement prediction model screens and stores the input information through the screening sequence and the inlet sequence, so as to effectively eliminate the hysteresis effect of the influencing factors of the inclination degree of the foundation pit side wall on the inclination displacement of the foundation pit side wall.

[0084] S4: Obtain the total displacement prediction model based on the trained native tilt displacement prediction model and the trained disturbance tilt displacement prediction model, and realize the prediction of the displacement of the foundation pit side wall based on the total displacement prediction model.

[0085] In a specific embodiment, the total displacement prediction model is expressed as:

[0086] S * (H+1) =W * (H+1) +Z * (H+1) (10)

[0087] In the formula, S * (H+1) is the total predicted displacement, W * (H+1) is the native displacement predicted by the native tilt displacement prediction model; Z * (H+1) is the disturbance displacement predicted by the disturbance tilt displacement prediction model.

[0088] In this embodiment, after training the total displacement prediction model in the present invention based on the data related to the inclination of the foundation pit side wall in Table 1, the deformation condition of the foundation pit is predicted based on the trained total displacement prediction model. In Table 1, the side wall inclination displacement values in the last column are obtained from the monitoring of a series of on-site equipment and facilities.

[0089] Table 1:

[0090]

[0091] Specifically, construct a monitoring data set J with a capacity of 100 = (J 1 , J 2, ..., J 100 ), and this data set is sourced from the monitoring of the inclination displacement of the foundation pit side wall in the experimental area of the Renqiu cable-stayed bridge.

[0092] Among them:

[0093] J 94 =(P A 94 , P B 94 , P C 94 , P D 94 , P E 94 , P F 94 , S 94 ),

[0094] J95 =(P A 95 ,P B 95 ,P C 95 ,P D 95 ,P E 95 ,P F 95 ,S 95 ), ...

[0095] J 100 =(P A 100 ,P B 100 ,P C 100 ,P D 100 ,P E 100 ,P F 100 ,S 100 ),

[0096] Among them: J 94 ,J 95 ......J 100 represents the monitoring data of the last 7 days, and is not limited to consecutive 7 days; P A 100 ,P B 100 ,P C 100 ,P D 100 ,P E 100 ,P F 100 respectively represent the excavation depth (A), groundwater level (B), vibration influence (C), elastic load of rock and soil mass (D), temperature (E) and slope top load (F) on the 100th day. The units of each factor are m, m, Hz, MN / m 2 、 o C, KN / m.

[0097] Using existing decomposition methods such as: decomposing the monitoring displacement by variational mode decomposition theory, wavelet optimization decomposition theory, etc., the original displacement data and disturbance displacement data are obtained as shown in Table 2 below, and the training data of the disturbance displacement prediction model are respectively constructed, denoted as: L 1 =(P A 1 ,P B 1,P C 1 ,P D 1 ,P E 1 ,P F 1 ,Z 1 ),P 2 =(P A 2 ,P B 2 ,P C 2 ,P D 2 ,P E 2 ,P F 2 ,Z 2 ),...,

[0098] L 100 =(P A 100 ,P B 100 ,P C 100 ,P D 100 ,P E 100 ,P F 100 ,Z 100 ), and the original displacement prediction model training data, expressed as: W 1 ,W 2 ,...,W 100Z。

[0099] Table 2:

[0100]

[0101] Specifically, in this embodiment, taking the 100th group of data as an example, the original foundation pit displacement is brought into the original inclination displacement prediction model, expressed as:

[0102] ,

[0103] ,

[0104] The disturbed foundation pit displacement and influencing factors are brought into the disturbed inclination displacement prediction model, expressed as:

[0105] ,

[0106] ,

[0107] ,

[0108] ,

[0109] ,

[0110] U H = I H × tanh(V H ) = 1.19 × tanh(30.82) = 5.79,

[0111] Z * （H+1） = softmax(M S × U H + G S ) = softmax(0.21 × 5.79 + 0.04) = 0.99,

[0112] S * (H+1) = W * (H+1) + Z * (H+1) = 0.99 + 16.49 = 17.4。

[0113] Specifically, Table 3 statistically analyzes the prediction errors of the test sample group. It can be seen from Table 3 that among the results predicted by the method proposed in the present invention, the maximum absolute error is 0.43, the maximum relative error is 2.3%, and the accuracy rate is relatively high, which can meet the requirements of the prediction and evaluation of the inclination trend of the foundation pit side wall. Therefore, it shows that the total displacement prediction model proposed based on this embodiment can directly, quickly and accurately calculate the inclination displacement of the foundation pit side wall based on the proportions of the currently measured excavation depth (A), groundwater level (B), vibration influence (C), elastic modulus of rock and soil (D), temperature (E) and the load on the top of the slope (F), without the need for a large number of complicated evaluations and calculations, greatly improving the efficiency of the prediction and evaluation calculation of the inclination trend of the foundation pit side wall.

[0114] Table 3:

[0115]

[0116] In summary, the present invention can predict the deformation displacement of the foundation pit completely based on two models, thereby realizing the determination of the stability degree of the foundation pit. It is an intelligent digital prediction method and improves the accuracy of the calculation results. Specifically, it also includes the following beneficial effects:

[0117] (1)Data sorting and statistics of historical monitoring datasets: In this embodiment, the data obtained from on-site monitoring is input into the overall model, which can automatically sort and summarize the monitoring data. Traditional monitoring data relies on abstract EXCEL spreadsheets or even paper documents for recording, making it prone to problems such as data loss and management chaos. With the model of the present invention, data can be sorted and processed, avoiding data loss and chaos.

[0118] (2)Accurate prediction of foundation pit deformation displacement: An important indicator for determining the stability of a foundation pit is the deformation displacement of the foundation pit sidewall at the current moment. By constructing a native tilt displacement prediction model and a disturbed tilt displacement prediction model, the present invention enables construction personnel to input the historical monitoring dataset into the model and quickly obtain the predicted value of the foundation pit deformation displacement at the next moment, avoiding the dependence on the technical experience of construction personnel and having high accuracy.

[0119] (3)Handling of time lag phenomena of external factors: During the prediction process of disturbed tilt displacement, the disturbed tilt displacement prediction model uses three-term sequences to store the prediction results of the input information in previous intervals, effectively solving the time lag problem of the response displacement of external factors. Compared with existing methods, this method is more stable, fully leveraging the computational advantages of the model, and the prediction results are more accurate.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features. However, such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the inclination trend of the sidewall of a deep and large foundation pit across time domain, characterized in that: The specific steps include: S1: Obtain influencing factor data of the inclination degree of the foundation pit side wall and inclination displacement data of the foundation pit side wall; and obtain original displacement data and historical influence data based on the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall, wherein the historical influence data includes disturbance displacement data and influencing factor data of the inclination degree of the foundation pit side wall; S2: establishing a native tilt displacement prediction model, and training it based on the native displacement data to obtain a trained native tilt displacement prediction model; S3: Establish a disturbance tilt displacement prediction model that takes into account the cross-time domain influence characteristics of external factors, and train it based on historical influence data to obtain a trained disturbance tilt displacement prediction model; S4: A total displacement prediction model is obtained based on the trained native tilt displacement prediction model and the trained disturbance tilt displacement prediction model, and the displacement prediction of the foundation pit side wall is realized based on the total displacement prediction model; The disturbance tilt displacement prediction model includes formulas (1) to (7), and the process of training it based on historical impact data is as follows: S31: dividing the historical impact data equally according to the time series of foundation pit deformation; and dividing the equally divided historical impact data into a number of calculation intervals according to the detection time; S32: The historical impact data P of the Hth calculation interval is H Substitute the screening series formula to obtain the screened data Q H , expressed as: Q H =X(M Q ×U H-1 +N Q ×P H +G Q ) (1) Among them, X is the sigmoid activation function, M Q and N Q are the coefficients of the linear relationship of the screening series, G Q To screen the error of linear relationship of the series; U H-1 is the abstract expression matrix of the inclination degree of the foundation pit side wall in the H-1th calculation interval; S33: The historical impact data P of the Hth calculation interval is H Substitute the entry sequence formula into it and it becomes: Y H =X(M Y ×U H-1 +N Y ×P H +G Y ) (2) V H =tanh(M V ×U H-1 +N V ×P H +G V ) (3) Where Y H and V H The entry sequence can store the data used by the model at the best level, X is the sigmoid activation function, M Y 、N Y 、M V and N V is the coefficient of the linear relationship of the input sequence, G Y and G V is the error of the linear relationship of the input series; S34: Output result V of the H-1th calculation interval based on formulas (1), (2) and (3) H-1 Optimize, the optimization formula is: V H =V H-1 ×Q H +Y H ×V H (4) S35: The data optimized by formula (4) and the historical impact data P of the Hth calculation interval are H Substituting into the export function formula, it is expressed as: I H =X(M I ×U H-1 +N I ×P H +G I ) (5) U H =I H ×tanh(V H ) (6) In the formula, I H is the current data result output by the export sequence, X is the sigmoid activation function, M I and N I is the coefficient of the linear relationship of the export series, G I is the error of the linear relationship of the export series, U H is the abstract expression matrix of the inclination degree of the foundation pit side wall in the Hth calculation interval; S36: U H Substitute the result output formula to obtain the disturbance displacement prediction result Z of the inclination degree of the foundation pit side wall on the H+1 day * (H+1) , expressed as: Z * (H+1) =softmax(M S ×U H +G S ) (7) Where M S The relationship coefficient of the result output formula, G S Output the error of the linear relationship for the result.

2. The cross-time domain prediction method for the inclination trend of the side wall of a deep and large foundation pit according to claim 1 is characterized in that: In S2, the original tilt displacement prediction model established is: O H =X(M O ×W H +G O ) (8) W * (H+1) =softmax(M W ×O H ) (9) In the formula, O H Optimize the representation coefficient for the model, G O is the average error; W H is the original displacement data, W * (H+1) is the original displacement prediction result on day H+1, M O and M W are the model relationship coefficients respectively.

3. The cross-time domain prediction method for the inclination trend of the side wall of a deep and large foundation pit according to claim 2 is characterized in that: In S1, the process of obtaining the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall, and obtaining the original displacement data and the historical influence data based on the influencing factor data of the inclination degree of the foundation pit side wall and the inclination displacement data of the foundation pit side wall is as follows: Construct a monitoring data set J with a capacity of H = (J1, J2...J H ); Among them, J1, J2...J H It represents the data of the pit sidewall tilt displacement and influencing factors on day H, and is not limited to consecutive days H+1; J1=(P A 1 ,P B 1 ,P C 1 ,P D 1 ,P E 1 ,P F 1 ,S1),J2=(P A 2 ,P B 2 ,P C 2 ,P D 2 ,P E 2 ,P F 2 ,S2),...,J H =(P A H ,P B H ,P C H ,P D H ,P E H ,P F H ,S H ); P A H ,P B H ,P C H ,P D H ,P E H ,P F H They represent the contents of the factors affecting the excavation depth, groundwater level, vibration effect, elastic load of rock and soil, temperature and top load on the Hth day, and are not limited to H consecutive days; S H The inclination displacement data of the foundation pit side wall on the Hth day; According to whether the inclination displacement data of the foundation pit side wall is affected by the influencing factors of the foundation pit inclination degree, the inclination displacement data of the foundation pit side wall is divided into original displacement data and disturbed displacement data, and the disturbed displacement data and the influencing factor data of the inclination degree of the foundation pit side wall are combined into historical impact data.

4. The cross-time domain prediction method for the inclination trend of the side wall of a deep and large foundation pit according to claim 3 is characterized in that: The total displacement prediction model is expressed as: S * (H+1) =W * (H+1) +Z * (H+1) (10) In the formula, S * (H+1) is the total predicted displacement, W * (H+1) Z is the native displacement predicted by the native tilt displacement prediction model; * (H+1) The disturbance displacement predicted by the disturbance tilt displacement prediction model.

Images