Tunnel step method-based surrounding rock deformation prediction method and system

Abstract

The embodiment of the application provides a surrounding rock deformation prediction method and system based on tunnel step method construction, the surrounding rock deformation sample data set is obtained, the tunnel surrounding rock deformation sample data set includes the accumulated deformation data time sequence of each step section, and the accumulated deformation data time sequence is obtained based on the surrounding rock deformation generated by step method construction on a reference tunnel. According to the initial surrounding rock deformation prediction function, data fitting is performed on the tunnel surrounding rock deformation sample data set, and the surrounding rock deformation prediction function corresponding to each step section is obtained. According to the to-be-constructed parameters of the to-be-constructed tunnel and the surrounding rock deformation prediction function, the surrounding rock deformation prediction result of the to-be-constructed tunnel is obtained, so that the accuracy and efficiency of surrounding rock deformation prediction during tunnel construction based on the step method are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel construction, in particular to a surrounding rock deformation prediction method and system based on tunnel bench method construction. BACKGROUND

[0002] With the continuous development of underground engineering construction, the surrounding rock geological conditions faced by tunnel construction are more complex. The engineering geological conditions and mechanical characteristics of soft rock areas make tunnel construction more difficult, and the deformation of soft rock has become a key problem in the construction field. When the tunnel is excavated, the deformation of surrounding rock will go through a process related to time and space, which is called the time-space effect of surrounding rock deformation. For different excavation methods, the characteristics of the time-space effect relationship curve of surrounding rock deformation are different. If the time-space effect of surrounding rock deformation during construction is ignored, there will be great safety hazards in the tunnel. Bench method excavation has the advantages of flexibility, strong adaptability, open operation surface, fast construction speed, temporary support for the working face, and is conducive to the stability of surrounding rock. Therefore, it is of great significance to study the time-space effect of surrounding rock deformation in bench method excavation to ensure construction safety, reasonably control the construction period and ensure the stability of the tunnel. However, at present, when tunnel construction is carried out based on the bench method, there are problems of low accuracy and efficiency of surrounding rock deformation prediction.

[0003] Therefore, how to improve the accuracy and efficiency of surrounding rock deformation prediction when tunnel construction is carried out based on the bench method is a problem to be solved. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a surrounding rock deformation prediction method and system based on tunnel bench method construction to improve the accuracy and efficiency of surrounding rock deformation prediction when tunnel construction is carried out based on the bench method.

[0005] In a first aspect, the present application provides a surrounding rock deformation prediction method based on tunnel bench method construction, comprising:

[0006] obtaining a tunnel surrounding rock deformation sample data set, the tunnel surrounding rock deformation sample data set including cumulative deformation data time series of each bench section, the cumulative deformation data time series being obtained based on the surrounding rock deformation generated by the bench method construction of a reference tunnel;

[0007] data fitting the tunnel surrounding rock deformation sample data set according to an initial surrounding rock deformation prediction function to obtain the surrounding rock deformation prediction function corresponding to each bench section, the initial surrounding rock deformation prediction function representing the mapping relationship between the surrounding rock deformation prediction result and the target influencing factor, the target influencing factor including the surrounding rock deformation result of the previous bench section, the surrounding rock deformation potential of the current bench section, the surrounding rock deformation rate of the current bench section, and the construction time of the current bench section;

[0008] According to the to-be-constructed parameter of the to-be-constructed tunnel and the surrounding rock deformation prediction function, a surrounding rock deformation prediction result of the to-be-constructed tunnel is obtained, and the to-be-constructed parameter includes a to-be-constructed bench segmentation and a construction time.

[0009] Optionally, the tunnel surrounding rock deformation sample data set is obtained by:

[0010] The surrounding rock deformation of the reference tunnel during the bench construction process is monitored, cumulative deformation data of different bench segmentations at different construction times is collected, and an original monitoring data set is formed;

[0011] The original monitoring data set is subjected to data screening processing, and a screened monitoring data set is obtained;

[0012] The screened monitoring data set is classified according to the bench segmentation, and the cumulative deformation data belonging to the same bench segmentation is arranged in time sequence to form a cumulative deformation data time sequence of each bench segmentation;

[0013] The cumulative deformation data time sequences of the bench segmentations are integrated to obtain the tunnel surrounding rock deformation sample data set.

[0014] Optionally, the tunnel surrounding rock deformation sample data set is subjected to data fitting according to the initial surrounding rock deformation prediction function to obtain the surrounding rock deformation prediction function corresponding to each bench segmentation, including:

[0015] Based on the cumulative deformation data time sequence of each bench segmentation in the tunnel surrounding rock deformation sample data set and the target influencing factor, the parameters in the initial surrounding rock deformation prediction function are preliminarily estimated to obtain a parameter initial estimation value;

[0016] The parameter initial estimation value is substituted into the initial surrounding rock deformation prediction function to obtain a preliminary surrounding rock deformation prediction function;

[0017] The preliminary surrounding rock deformation prediction function is fitted and optimized by using the tunnel surrounding rock deformation sample data set, and the parameters of the preliminary surrounding rock deformation prediction function are adjusted until the error between the cumulative deformation data calculated by the adjusted preliminary surrounding rock deformation prediction function and the cumulative deformation data in the tunnel surrounding rock deformation sample data set is less than a preset threshold, and the surrounding rock deformation prediction function corresponding to each bench segmentation is obtained.

[0018] Optionally, the tunnel surrounding rock deformation sample data set is used to fit and optimize the preliminary surrounding rock deformation prediction function, parameters of the preliminary surrounding rock deformation prediction function are adjusted until the error of the cumulative deformation data calculated by the adjusted preliminary surrounding rock deformation prediction function and the cumulative deformation data in the tunnel surrounding rock deformation sample data set is less than a preset threshold, and the surrounding rock deformation prediction function corresponding to each step segment is obtained, including:

[0019] The cumulative deformation data time series of each step segment in the tunnel surrounding rock deformation sample data set is divided into a plurality of data subsegments according to construction stages, and each data subsegment contains cumulative deformation data of a continuous time series;

[0020] For each data subsegment, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm to obtain an adjusted parameter;

[0021] The predicted cumulative deformation data of the preliminary surrounding rock deformation prediction function on the data subsegment is calculated according to the adjusted parameter;

[0022] The error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data subsegment is calculated, and the error value is taken as a feedback index for parameter adjustment;

[0023] If the error value is greater than or equal to the preset threshold, the adjusted parameter is continuously adjusted based on the feedback index until the error value is less than the preset threshold, and the adjusted adjusted parameter is determined as the target parameter corresponding to the data subsegment;

[0024] The target parameters corresponding to all data subsegments are integrated to generate a parameter set of the surrounding rock deformation prediction function corresponding to the step segment, and the surrounding rock deformation prediction function corresponding to each step segment is obtained according to the parameter set.

[0025] Optionally, for each data subsegment, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm to obtain an adjusted parameter, including:

[0026] The parameter search space of the nonlinear optimization algorithm is determined, and the parameter search space contains a parameter range related to the surrounding rock deformation result of the previous step segment, a parameter range related to the surrounding rock deformation potential of the current step segment, a data range related to the surrounding rock deformation rate of the current step segment, and a parameter range related to the construction time of the current step segment;

[0027] Within the parameter search space, a grid search method is used to preliminarily screen the parameter initial estimation value, generate a plurality of candidate parameter combinations, and calculate the goodness-of-fit index of the preliminary surrounding rock deformation prediction function corresponding to each candidate parameter combination on the data sub-section, the goodness-of-fit index being used to reflect the closeness of the predicted cumulative deformation data and the actual cumulative deformation data.

[0028] The candidate parameter combination with the highest goodness-of-fit index is selected as the initial iteration parameter, and the initial iteration parameter is input into the gradient descent module of the nonlinear optimization algorithm to calculate the parameter gradient of the preliminary surrounding rock deformation prediction function with respect to the initial iteration parameter.

[0029] The initial iteration parameter is updated and adjusted according to the direction and size of the parameter gradient until the adjustment number reaches a preset adjustment number threshold or the modulus of the parameter gradient is less than a preset gradient threshold, and the adjustment parameter is obtained.

[0030] Optionally, the error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data sub-section is calculated, and the error value is taken as a feedback index for parameter adjustment, comprising:

[0031] The predicted cumulative deformation data and the actual cumulative deformation data in the data sub-section are one-to-one corresponding matched according to the time stamp to generate a predicted actual data pair sequence.

[0032] For each predicted actual data pair in the predicted actual data pair sequence, the absolute difference value of the predicted cumulative deformation data and the actual cumulative deformation data is calculated, and the absolute difference value is squared to obtain a squared error value.

[0033] The sum of the squared error values of all predicted actual data pairs is calculated, and the sum of the squared error values is divided by the number of data pairs in the predicted actual data pair sequence to obtain the error value.

[0034] Optionally, the parameter initial estimation value is obtained by preliminarily estimating the parameters in the initial surrounding rock deformation prediction function based on the cumulative deformation data time sequence of each step section in the tunnel surrounding rock deformation sample data set and the target influencing factor, comprising:

[0035] A parameter type set corresponding to the target influencing factor in the initial surrounding rock deformation prediction function is identified, the parameter type set including a first parameter related to the surrounding rock deformation result of the previous step section, a second parameter related to the surrounding rock deformation potential of the current step section, a third parameter related to the surrounding rock deformation rate of the current step section, and a fourth parameter related to the construction time of the current step section.

[0036] extracting, from the tunnel surrounding rock deformation sample data set, an end point data value of a cumulative deformation data time sequence of the last bench section as an initial estimated value of the first parameter;

[0037] determining an initial estimated value of the second parameter according to a deformation data range value in the cumulative deformation data time sequence of the current bench section in the tunnel surrounding rock deformation sample data set, the initial estimated value of the second parameter being in positive correlation with the deformation data range value;

[0038] determining an initial estimated value of the third parameter according to a deformation data change amount of the cumulative deformation data time sequence of the current bench section in the tunnel surrounding rock deformation sample data set, the initial estimated value of the third parameter being in positive correlation with the deformation data change amount;

[0039] determining an initial estimated value of the fourth parameter according to a construction duration data of the current bench section in the tunnel surrounding rock deformation sample data set.

[0040] Optionally, the step of substituting the initial estimated value of the parameter into the initial surrounding rock deformation prediction function to obtain a preliminary surrounding rock deformation prediction function comprises:

[0041] substituting the initial estimated value of the parameter into the initial surrounding rock deformation prediction function to obtain a to-be-verified surrounding rock deformation prediction function;

[0042] obtaining, according to a verification data subset in the tunnel surrounding rock deformation sample data set and the to-be-verified surrounding rock deformation prediction function, a verification error value of a function output value of the to-be-verified surrounding rock deformation prediction function and corresponding verification data in the verification data subset;

[0043] if the verification error value is greater than or equal to a verification error threshold value, adjusting a fitting strategy of the to-be-verified surrounding rock deformation prediction function and / or the initial estimated value of the parameter based on the verification error value to perform re-fitting until the verification error value is less than the verification error threshold value, and obtaining the preliminary surrounding rock deformation prediction function.

[0044] Optionally, the step of obtaining a surrounding rock deformation prediction result of the to-be-constructed tunnel according to a to-be-constructed parameter of the to-be-constructed tunnel and the surrounding rock deformation prediction function comprises:

[0045] selecting, from the surrounding rock deformation prediction functions of the bench sections, a target surrounding rock deformation prediction function matched with the to-be-constructed bench section according to the to-be-constructed bench section;

[0046] substituting a construction time point in the to-be-constructed parameter into the target surrounding rock deformation prediction function to calculate a surrounding rock cumulative deformation prediction value corresponding to the construction time point of the to-be-constructed tunnel.

[0047] The accumulated deformation prediction value of the surrounding rock is taken as the surrounding rock deformation prediction result of the tunnel to be constructed.

[0048] In a second aspect, the present application provides a surrounding rock deformation prediction system based on tunnel bench method construction, which comprises a machine readable storage medium and a processor, the machine readable storage medium stores machine executable instructions, and the processor, when executing the machine executable instructions, realizes the surrounding rock deformation prediction method based on tunnel bench method construction.

[0049] The surrounding rock deformation prediction method and system based on tunnel bench method construction provided by the present application can accurately establish the mapping relationship between the surrounding rock deformation prediction result and the multi-dimensional influencing factors by obtaining the tunnel surrounding rock deformation sample data set containing the time series of the accumulated deformation data of each bench segment, and performing data fitting on the sample data set according to the initial surrounding rock deformation prediction function, and then obtaining the surrounding rock deformation prediction function corresponding to each bench segment, which comprehensively considers the surrounding rock deformation result of the previous bench segment, the surrounding rock deformation potential of the current bench segment, the surrounding rock deformation rate of the current bench segment, and the construction time of the current bench segment and other target influencing factors. With these scientific and reasonable technical means, the accuracy and reliability of the surrounding rock deformation prediction in the tunnel bench method construction can be effectively improved, and a strong guarantee for the safety and stability of the tunnel construction is provided. BRIEF DESCRIPTION OF DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0051] Figure 1 A flowchart of a surrounding rock deformation prediction method based on tunnel bench method construction provided by an embodiment of the present application;

[0052] Figure 2 A fitting curve diagram of crown settlement deformation data provided by an embodiment of the present application;

[0053] Figure 3 A fitting curve diagram of side wall horizontal deformation data provided by an embodiment of the present application;

[0054] Figure 4 A data diagram of subsequent construction surrounding rock deformation provided by an embodiment of the present application;

[0055] Figure 5A structural schematic diagram of a surrounding rock deformation prediction system based on tunnel step method construction is provided for an embodiment of the present application.

[0056] The specific embodiments of the present application have been shown through the above-described drawings, and will be described in more detail hereinafter. The drawings and the written description are not intended to limit the scope of the present application in any way, but to explain the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0057] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all. All other embodiments obtained by those skilled in the art without creative labor on the basis of the embodiments in the present application shall fall within the scope of protection of the present application.

[0058] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or end including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or end.

[0059] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily all refer to the same embodiment, nor is it necessarily independent or alternative embodiments to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0060] Figure 1 A flowchart of a surrounding rock deformation prediction method based on tunnel step method construction is provided for an embodiment of the present application. It should be understood that in other embodiments, the order of some steps of the surrounding rock deformation prediction method based on tunnel step method construction of the present embodiment can be shared according to actual needs, or some steps can be omitted or maintained. As shown in the figure, the method can include the following steps: Figure 1

[0061] ​Step S110: Obtain a tunnel surrounding rock deformation sample dataset, wherein the tunnel surrounding rock deformation sample dataset comprises accumulated deformation data time series of each bench segment, and the accumulated deformation data time series is obtained based on surrounding rock deformation caused by bench method construction of a reference tunnel.

[0062] In actual operation, in order to obtain the tunnel surrounding rock deformation sample dataset, first, bench method construction is carried out in the reference tunnel. During construction, a variety of professional monitoring equipment is used to monitor the deformation of the surrounding rock. These monitoring equipment can be high-precision displacement sensors, strain gauges, etc. Displacement sensors can accurately measure the displacement changes of the surrounding rock in different directions, and strain gauges can monitor the strain inside the surrounding rock. Through these devices, accumulated deformation data of different bench segments at different construction times is collected.

[0063] Suppose the reference tunnel is divided into multiple bench segments, such as bench segment A, bench segment B, etc., by bench method construction. For each bench segment, monitoring is continuously carried out from the start of construction. At each time point of construction, the accumulated deformation data of the bench segment is recorded. These data are continuously accumulated over time, reflecting the surrounding rock deformation of the bench segment at different construction stages.

[0064] The accumulated deformation data of different bench segments at different construction times collected is integrated together to form an original monitoring data set. This set contains a large amount of data, but there may be some abnormal data due to equipment errors, external interference, etc. Therefore, data screening processing needs to be performed on the original monitoring data set.

[0065] Data screening processing can use statistical analysis methods. First, calculate the statistical characteristics of the accumulated deformation data of each bench segment, such as mean, standard deviation, etc. Then, set screening rules according to these statistical characteristics. For example, data points deviating from the mean by a certain multiple of the standard deviation are considered abnormal data and are excluded from the original monitoring data set. After such screening processing, a screened monitoring data set is obtained.

[0066] Next, the screened monitoring data set is classified according to the bench segments. The accumulated deformation data belonging to the same bench segment is extracted and then arranged in chronological order. For example, for bench segment A, the accumulated deformation data at different construction times is arranged in chronological order to form the accumulated deformation data time series of bench segment A. Similarly, other bench segments are processed to obtain the accumulated deformation data time series of each bench segment.

[0067] Finally, the cumulative deformation data time series of each bench section are integrated together, and the tunnel surrounding rock deformation sample data set is obtained. This data set contains the cumulative deformation data time series of multiple bench sections, and each time series is a multi-dimensional data set, where each data point corresponds to the cumulative deformation of the bench section at a specific construction time.

[0068] Step S111: Monitor the surrounding rock deformation of the reference tunnel during the bench method construction process, collect the cumulative deformation data of different bench sections at different construction times, and form the original monitoring data set.

[0069] During the bench method construction of the reference tunnel, monitoring equipment is reasonably arranged at different positions of the tunnel. For displacement sensors, multiple displacement sensors are installed at key positions of each bench section, such as the top, side wall, etc., to comprehensively monitor the deformation of the bench section in all directions. Strain gauges can be installed inside the surrounding rock to monitor the stress changes inside the surrounding rock.

[0070] During the construction process, the surrounding rock deformation is monitored at certain time intervals. For example, the cumulative deformation data of each bench section is recorded once every certain period of time. This time interval can be adjusted according to the actual situation of the construction. As the construction progresses, the cumulative deformation data of different bench sections at different construction times is continuously collected.

[0071] These collected data are preliminarily sorted, and the cumulative deformation data of each bench section at different construction times are stored together to form the original monitoring data set. The data in this set contains rich information, but there may be some inaccurate or abnormal data points that need to be processed later.

[0072] Step S112: Data screening processing is performed on the original monitoring data set to obtain the screened monitoring data set.

[0073] When performing data screening processing on the original monitoring data set, first, statistical analysis is performed on the cumulative deformation data of each bench section. The mean, median, standard deviation, and other statistical characteristics of the cumulative deformation data of each bench section are calculated. These statistical characteristics can reflect the overall distribution of the cumulative deformation data of the bench section.

[0074] According to the calculated statistical characteristics, the screening rules are set. For example, for the cumulative deformation data of each bench section, if a data point deviates from the mean of the cumulative deformation data of the bench section by a certain multiple of the standard deviation, the data point is considered abnormal data. The multiple can be adjusted according to the actual situation to ensure that the screened abnormal data is truly not in line with the normal data distribution.

[0075] In the screening process, each data point in the original monitoring data set is checked one by one. If a data point meets the determination condition of abnormal data, it is excluded from the original monitoring data set. After such screening processing, the screened monitoring data set is obtained. The data in the screened monitoring data set is more accurate and reliable, and can better reflect the actual deformation of the tunnel surrounding rock.

[0076] Step S113: According to the step segmentation, the screened monitoring data set is classified, and the cumulative deformation data belonging to the same step segmentation is arranged in time sequence to form the cumulative deformation data time sequence of each step segmentation.

[0077] After obtaining the screened monitoring data set, it is classified according to the step segmentation. The data in the screened monitoring data set is divided according to the step segmentation, and the cumulative deformation data belonging to the same step segmentation is extracted to form the data set corresponding to each step segmentation.

[0078] For each data set corresponding to a step segmentation, the cumulative deformation data therein is sorted according to the construction time. The construction time can be calculated from the time point when the construction of the step segmentation starts, and the cumulative deformation data is arranged in time sequence. In this way, the cumulative deformation data time sequence of each step segmentation is formed. Each time sequence is a multi-dimensional data set, and each data point corresponds to the cumulative deformation of the step segmentation at a specific construction time.

[0079] Step S114: The cumulative deformation data time sequences of each step segmentation are integrated to obtain the tunnel surrounding rock deformation sample data set.

[0080] The cumulative deformation data time sequences of each step segmentation are integrated to form a unified data set, i.e. the tunnel surrounding rock deformation sample data set. This integration process can be to store the cumulative deformation data time sequences of each step segmentation according to certain rules, for example, to store each step segmentation time sequence as an independent subset in a larger data structure.

[0081] The tunnel surrounding rock deformation sample data set contains the cumulative deformation data time sequences of multiple step segmentations, which reflect the surrounding rock deformation of different step segmentations at different construction stages. Through analysis and processing of this data set, data basis can be provided for subsequent surrounding rock deformation prediction.

[0082] Step S120: data fitting is performed on the tunnel surrounding rock deformation sample dataset according to the initial surrounding rock deformation prediction function, to obtain the surrounding rock deformation prediction functions corresponding to each step segment, the initial surrounding rock deformation prediction function representing the mapping relationship between the surrounding rock deformation prediction result and the target influencing factor, the target influencing factor including the surrounding rock deformation result of the previous step segment, the surrounding rock deformation potential of the current step segment, the surrounding rock deformation rate of the current step segment, and the construction time of the current step segment.

[0083] The initial surrounding rock deformation prediction function is a pre-set function, which describes the mapping relationship between the surrounding rock deformation prediction result and the target influencing factor. In order to obtain the surrounding rock deformation prediction functions corresponding to each step segment, data fitting needs to be performed on the initial surrounding rock deformation prediction function. For example, the initial surrounding rock deformation prediction function is where y is the surrounding rock deformation prediction result, c is a parameter related to the surrounding rock deformation result of the previous step segment, a is a parameter related to the surrounding rock deformation potential of the current step segment (for example, it can be valued in the range of 100-200), b is a parameter related to the surrounding rock deformation rate of the current step segment (for example, it can be valued in the range of 0.03-0.08 according to engineering experience), t is a parameter related to the construction time of the current step segment, and x is the time point to be predicted.

[0084] When performing data fitting, the parameters in the initial surrounding rock deformation prediction function need to be preliminarily estimated first. This is because the parameters in the initial surrounding rock deformation prediction function are unknown and need to be determined according to the tunnel surrounding rock deformation sample dataset. The process of preliminary estimation of parameters can be performed by analyzing the relationship between the target influencing factors and the surrounding rock deformation.

[0085] Then, the preliminarily estimated parameters are substituted into the initial surrounding rock deformation prediction function to obtain the preliminary surrounding rock deformation prediction function. This preliminary surrounding rock deformation prediction function may not be able to well fit the tunnel surrounding rock deformation sample dataset, so further optimization is needed.

[0086] The optimization process can use methods such as nonlinear optimization algorithm. By continuously adjusting the parameters in the preliminary surrounding rock deformation prediction function, the fitting effect of the function on the tunnel surrounding rock deformation sample dataset becomes better and better. When the fitting effect reaches a certain standard, the surrounding rock deformation prediction functions corresponding to each step segment are obtained. These functions can accurately predict the surrounding rock deformation of different step segments at different construction stages.

[0087] Step S121: based on the cumulative deformation data time series of each step segment in the tunnel surrounding rock deformation sample data set and the target influencing factor, the parameters in the initial surrounding rock deformation prediction function are preliminarily estimated to obtain parameter initial estimation values.

[0088] When preliminarily estimating the parameters in the initial surrounding rock deformation prediction function, first, the parameter type set corresponding to the target influencing factor in the initial surrounding rock deformation prediction function is identified. The target influencing factor includes the surrounding rock deformation result of the previous step segment, the surrounding rock deformation potential of the current step segment, the surrounding rock deformation rate of the current step segment, and the construction time of the current step segment, so the parameter type set includes a first parameter related to the surrounding rock deformation result of the previous step segment, a second parameter related to the surrounding rock deformation potential of the current step segment, a third parameter related to the surrounding rock deformation rate of the current step segment, and a fourth parameter related to the construction time of the current step segment.

[0089] For the first parameter related to the surrounding rock deformation result of the previous step segment, the endpoint data value of the cumulative deformation data time series of the previous step segment in the tunnel surrounding rock deformation sample data set can be extracted as the initial estimation value of the parameter. This is because the final deformation result of the previous step segment will affect the surrounding rock deformation of the current step segment, so the endpoint data value thereof can be used to preliminarily estimate the parameter.

[0090] For the second parameter related to the surrounding rock deformation potential of the current step segment, the initial estimation value of the parameter is determined according to the deformation data range value in the cumulative deformation data time series of the current step segment in the tunnel surrounding rock deformation sample data set. The deformation data range value reflects the maximum deformation range that the current step segment can reach during construction, and the initial estimation value of the parameter is positively correlated with the deformation data range value. That is, the larger the deformation data range value, the larger the initial estimation value of the parameter.

[0091] For the third parameter related to the surrounding rock deformation rate of the current step segment, the initial estimation value of the parameter is determined according to the deformation data change amount of the cumulative deformation data time series of the current step segment in the tunnel surrounding rock deformation sample data set. The deformation data change amount reflects the deformation degree of the current step segment per unit time, and the initial estimation value of the parameter is positively correlated with the deformation data change amount. That is, the larger the deformation data change amount, the larger the initial estimation value of the parameter.

[0092] For the fourth parameter related to the construction time of the current step segment, the initial estimation value of the parameter is determined according to the construction duration data of the current step segment in the tunnel surrounding rock deformation sample data set. The construction duration data reflects the construction time of the current step segment, which also has an important influence on the surrounding rock deformation, so the construction duration data can be used to preliminarily estimate the parameter.

[0093] Step S122: Substitute the parameter initial estimate value into the initial surrounding rock deformation prediction function to obtain a preliminary surrounding rock deformation prediction function.

[0094] Substitute the parameter initial estimate value obtained by preliminary estimation into the initial surrounding rock deformation prediction function to obtain a to-be-verified surrounding rock deformation prediction function. This to-be-verified surrounding rock deformation prediction function is obtained based on the preliminary estimated parameters, and whether it can accurately describe the surrounding rock deformation needs to be verified.

[0095] In order to verify the accuracy of the to-be-verified surrounding rock deformation prediction function, calculation needs to be performed according to the verification data subset in the tunnel surrounding rock deformation sample data set and the to-be-verified surrounding rock deformation prediction function. The verification data subset is a part of data selected from the tunnel surrounding rock deformation sample data set, which is used to verify the performance of the to-be-verified surrounding rock deformation prediction function.

[0096] The data in the verification data subset is input into the to-be-verified surrounding rock deformation prediction function to obtain the function output value of the function. Then the verification error value of the function output value and the corresponding verification data in the verification data subset is calculated. The verification error value reflects the difference between the prediction result of the to-be-verified surrounding rock deformation prediction function and the actual data.

[0097] If the verification error value is greater than or equal to the verification error threshold, it means that the performance of the to-be-verified surrounding rock deformation prediction function is not good enough and needs to be adjusted. The fitting strategy of the to-be-verified surrounding rock deformation prediction function can be adjusted based on the verification error value, such as changing the fitting method or adjusting the parameter range of fitting. The parameter initial estimate value can also be adjusted to make it more accurate. Through continuous adjustment and re-fitting, until the verification error value is less than the verification error threshold, at this time the preliminary surrounding rock deformation prediction function is obtained.

[0098] Step S123: Perform fitting optimization on the preliminary surrounding rock deformation prediction function using the tunnel surrounding rock deformation sample data set, adjust the parameters of the preliminary surrounding rock deformation prediction function, until the cumulative deformation data calculated by the adjusted preliminary surrounding rock deformation prediction function and the cumulative deformation data in the tunnel surrounding rock deformation sample data set have an error less than a preset threshold, to obtain the surrounding rock deformation prediction function corresponding to each step section.

[0099] In order to perform fitting optimization on the preliminary surrounding rock deformation prediction function using the tunnel surrounding rock deformation sample data set, first the cumulative deformation data time series of each step section in the tunnel surrounding rock deformation sample data set is divided into multiple data sub-sections according to the construction stage. Each data sub-section contains continuous time series of cumulative deformation data, and the purpose of such division is to more carefully analyze the surrounding rock deformation in different construction stages.

[0100] For each data sub-section, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm. In the iterative adjustment process, better parameter values are constantly sought, so that the fitting effect of the preliminary surrounding rock deformation prediction function on the data sub-section is better.

[0101] The predicted cumulative deformation data of the preliminary surrounding rock deformation prediction function on the data sub-section is calculated according to the adjusted parameters. Then the error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data sub-section is calculated, and the error value is taken as the feedback index of parameter adjustment.

[0102] If the error value is greater than or equal to the preset threshold, it means that the current parameter cannot well fit the data sub-section, and the adjustment parameter needs to be adjusted based on the feedback index. Through continuous iterative adjustment, until the error value is less than the preset threshold. At this time, the adjusted adjustment parameter is determined as the target parameter corresponding to the data sub-section.

[0103] The target parameters corresponding to all data sub-sections are integrated to generate a parameter set of the surrounding rock deformation prediction function corresponding to the step section. According to this parameter set, the surrounding rock deformation prediction function corresponding to each step section can be obtained. These functions can accurately predict the surrounding rock deformation of different step sections in different construction stages.

[0104] Step S1231: The cumulative deformation data time series of each step section in the tunnel surrounding rock deformation sample data set is divided into multiple data sub-sections according to the construction stage, and each data sub-section contains continuous time series of cumulative deformation data.

[0105] When dividing the cumulative deformation data time series of each step section in the tunnel surrounding rock deformation sample data set into multiple data sub-sections, the basis for division needs to be determined first. The construction stage can be divided according to the characteristics of the step method construction, for example, the construction process of each step section can be divided into excavation stage, support stage, etc.

[0106] According to the division of the construction stage, the cumulative deformation data time series of each step section is divided. For example, for the cumulative deformation data time series of a step section, the data in the excavation stage is extracted to form a data sub-section. Similarly, the data in the support stage is extracted to form another data sub-section.

[0107] Each data sub-section contains continuous time series of cumulative deformation data, which reflects the surrounding rock deformation of the step section in a specific construction stage. By dividing the cumulative deformation data time series into multiple data sub-sections, the surrounding rock deformation characteristics of different construction stages can be analyzed more carefully, providing a more accurate data basis for subsequent parameter adjustment and fitting optimization.

[0108] Step S1232: For each data sub-segment, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm to obtain an adjusted parameter.

[0109] For each data sub-segment, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the iterative adjustment process of the nonlinear optimization algorithm is started.

[0110] First, the parameter search space of the nonlinear optimization algorithm is determined. The parameter search space includes the parameter range related to the surrounding rock deformation result of the last step segment, the parameter range related to the surrounding rock deformation potential of the current step segment, the data range related to the surrounding rock deformation rate of the current step segment, and the parameter range related to the construction time of the current step segment. These parameter ranges are determined according to actual engineering experience and analysis of target influencing factors, ensuring that suitable parameter values can be found within this range.

[0111] Within the parameter search space, a grid search method is used to preliminarily screen the initial estimated value of the parameter. The grid search method divides the parameter search space into multiple grid points and calculates the parameter combination corresponding to each grid point. The fitting goodness index of the preliminary surrounding rock deformation prediction function corresponding to each candidate parameter combination is calculated. The fitting goodness index is used to reflect the closeness of the predicted cumulative deformation data and the actual cumulative deformation data, for example, the coefficient of determination can be used to measure it.

[0112] The candidate parameter combination with the highest fitting goodness index is selected as the initial iteration parameter. The initial iteration parameter is input into the gradient descent module of the nonlinear optimization algorithm, and the parameter gradient of the preliminary surrounding rock deformation prediction function with respect to the initial iteration parameter is calculated. The parameter gradient reflects the rate and direction of change of the function at that point.

[0113] The initial iteration parameter is updated and adjusted according to the direction and size of the parameter gradient. If the direction of the parameter gradient points to the direction of increasing function value, the parameter value is decreased; if the direction of the parameter gradient points to the direction of decreasing function value, the parameter value is increased. Iterative adjustment is continuously performed until the adjustment number reaches the preset adjustment number threshold or the modulus of the parameter gradient is less than the preset gradient threshold. At this time, the adjusted parameter is obtained.

[0114] Step S1233: The predicted cumulative deformation data of the preliminary surrounding rock deformation prediction function on the data sub-segment is calculated according to the adjusted parameter.

[0115] The adjustment parameter is substituted into the preliminary surrounding rock deformation prediction function to calculate each time point in the data subsegment. For each time point in the data subsegment, the target influencing factor data corresponding to the time point is input into the preliminary surrounding rock deformation prediction function, and the adjustment parameter is combined to calculate the predicted cumulative deformation data of the time point.

[0116] The calculation is sequentially performed on all time points in the data subsegment, and the predicted cumulative deformation data of the preliminary surrounding rock deformation prediction function on the data subsegment is obtained. These predicted cumulative deformation data reflect the prediction results of the preliminary surrounding rock deformation prediction function on the data subsegment under the adjustment parameter.

[0117] Step S1234: Calculate the error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data subsegment, and use the error value as a feedback indicator for parameter adjustment.

[0118] The predicted cumulative deformation data and the actual cumulative deformation data in the data subsegment are matched one by one according to the time stamp to generate a sequence of predicted actual data pairs. For each predicted actual data pair in the sequence of predicted actual data pairs, the absolute difference between the predicted cumulative deformation data and the actual cumulative deformation data is calculated.

[0119] The absolute difference is squared to obtain the squared error value. This is because squaring can amplify the effect of the error, making the error more obvious.

[0120] The squared error values of all predicted actual data pairs are added to obtain the sum of squared error values. Then, the sum of squared error values is divided by the number of data pairs in the sequence of predicted actual data pairs to obtain the error value. This error value reflects the average difference between the predicted cumulative deformation data and the actual cumulative deformation data, and can be used as a feedback indicator for parameter adjustment.

[0121] Step S1235: If the error value is greater than or equal to the preset threshold, continue to adjust the adjustment parameter based on the feedback indicator until the error value is less than the preset threshold, and determine the adjusted adjustment parameter as the target parameter corresponding to the data subsegment.

[0122] If the calculated error value is greater than or equal to the preset threshold, it indicates that the fitting effect of the preliminary surrounding rock deformation prediction function is not good enough, and the adjustment parameter needs to be further adjusted.

[0123] Based on the error value as a feedback indicator, the adjustment parameter can be further adjusted using methods such as gradient descent. According to the size and direction of the error value, the adjustment direction and adjustment amplitude of the adjustment parameter are determined.

[0124] The error value of the predicted cumulative deformation data and the actual cumulative deformation data is recalculated after each adjustment. Until the error value is less than the preset threshold, which means that the fitting effect of the preliminary surrounding rock deformation prediction function on the data subsegment has reached the requirement. At this time, the adjusted adjustment parameter is determined as the target parameter corresponding to the data subsegment. The target parameter is obtained after multiple iterations and adjustment, which can make the preliminary surrounding rock deformation prediction function as accurate as possible to predict the cumulative deformation data of the surrounding rock in the data subsegment.

[0125] Step S1236: integrating the target parameters corresponding to all data subsegments to generate a parameter set of the surrounding rock deformation prediction function corresponding to the step section, and obtaining the surrounding rock deformation prediction function corresponding to each step section according to the parameter set.

[0126] After obtaining the target parameter corresponding to each data subsegment, the target parameters need to be integrated. Since each data subsegment corresponds to the situation of the step section in different construction stages, integrating these target parameters can comprehensively consider the surrounding rock deformation characteristics of the entire step section in different construction stages.

[0127] The integration method can be determined according to the nature and characteristics of the target parameters. For example, if the target parameters are independent of each other, and the influence of the target parameters of each data subsegment on the final function is the same, the target parameters can be simply combined to form a parameter vector. If the influence of the target parameters of different data subsegments on the final function is different, a corresponding weight can be assigned to each target parameter, and then they are integrated by weighted splicing.

[0128] After integration, a parameter set of the surrounding rock deformation prediction function corresponding to the step section is generated. This parameter set contains the target parameter information of all data subsegments, reflecting the comprehensive influence of the step section in different construction stages on the surrounding rock deformation prediction.

[0129] Substituting this parameter set into the preliminary surrounding rock deformation prediction function can obtain the surrounding rock deformation prediction function corresponding to the step section. This function can more accurately describe the relationship between the surrounding rock deformation and the target influencing factors of the step section in the entire construction process, and is of great significance for predicting the surrounding rock deformation of the step section at different construction time.

[0130] Step S130: obtaining the surrounding rock deformation prediction result of the tunnel to be constructed according to the to-be-constructed parameters of the tunnel to be constructed and the surrounding rock deformation prediction function, wherein the to-be-constructed parameters include the to-be-constructed step section and the construction time.

[0131] In practical applications, for a tunnel to be constructed, the parameters to be constructed, i.e., the step section to be constructed and the construction time, are first determined. The step section to be constructed indicates the specific step section position to be constructed, and the construction time indicates which time point in the construction process of the step section.

[0132] According to the step section to be constructed, a target surrounding rock deformation prediction function that matches the step section to be constructed is selected from the surrounding rock deformation prediction functions corresponding to the step sections obtained before. Because different step sections may differ in geological conditions, construction technology, etc., the corresponding surrounding rock deformation prediction functions are also different. Selecting a function that matches the step section to be constructed can more accurately predict the surrounding rock deformation.

[0133] The construction time in the parameters to be constructed is substituted into the target surrounding rock deformation prediction function. During the substitution process, the data format and unit of the construction time are ensured to be consistent with the requirements of the target surrounding rock deformation prediction function to ensure the accuracy of the calculation. Through the calculation of the function, the surrounding rock cumulative deformation prediction value of the tunnel to be constructed corresponding to the construction time can be obtained.

[0134] This surrounding rock cumulative deformation prediction value reflects the cumulative deformation of the surrounding rock of the tunnel to be constructed under the given step section to be constructed and construction time. The surrounding rock cumulative deformation prediction value is taken as the surrounding rock deformation prediction result of the tunnel to be constructed. This result can provide an important reference for construction personnel, helping them to understand the deformation trend of the tunnel surrounding rock in advance, so as to take corresponding measures to ensure construction safety and tunnel stability.

[0135] Step S131: According to the step section to be constructed, a target surrounding rock deformation prediction function that matches the step section to be constructed is selected from the surrounding rock deformation prediction functions corresponding to the step sections.

[0136] After the step section to be constructed is determined, a step section identifier corresponds to the surrounding rock deformation prediction functions corresponding to the step sections obtained before. An index table or mapping relationship can be established to associate the identifier of each step section with the corresponding surrounding rock deformation prediction function.

[0137] When the identifier of the step section to be constructed is obtained, the index table or mapping relationship is looked up to quickly locate the target surrounding rock deformation prediction function that matches the step section to be constructed. This method can improve the efficiency of selecting the target function and avoid unnecessary calculation and comparison.

[0138] For example, if the bench section is identified by a number, a dictionary structure can be constructed with the bench section number as the key and the corresponding surrounding rock deformation prediction function as the value. When the number of the bench section to be constructed is known, the corresponding function can be directly obtained from the dictionary.

[0139] Step S132: Substitute the construction time in the to-be-constructed parameter into the target surrounding rock deformation prediction function to calculate the surrounding rock cumulative deformation prediction value of the to-be-constructed tunnel at the construction time.

[0140] Before substituting the construction time into the target surrounding rock deformation prediction function, necessary preprocessing of the construction time is required. Ensure that the data type and format of the construction time meet the input requirements of the target surrounding rock deformation prediction function. For example, if the target surrounding rock deformation prediction function requires the construction time to be represented as a time difference from the start of the bench section construction, the actual construction time needs to be converted into this form of time difference.

[0141] Then, the preprocessed construction time is input into the target surrounding rock deformation prediction function as input. The target surrounding rock deformation prediction function will process and calculate the input construction time according to its internal mapping relationship and parameter set. During the calculation process, the function will comprehensively consider the surrounding rock deformation results of the previous bench section, the surrounding rock deformation potential of the current bench section, the surrounding rock deformation rate of the current bench section, and the construction time of the current bench section, etc. Target influencing factors, and finally output the surrounding rock cumulative deformation prediction value of the to-be-constructed tunnel at the construction time.

[0142] This calculation process is based on the function and parameter set obtained by fitting and optimizing the tunnel surrounding rock deformation sample data set, so it can accurately predict the cumulative deformation of the surrounding rock to a certain extent.

[0143] Step S133: Take the surrounding rock cumulative deformation prediction value as the surrounding rock deformation prediction result of the to-be-constructed tunnel.

[0144] After obtaining the surrounding rock cumulative deformation prediction value of the to-be-constructed tunnel at the construction time, it is taken as the surrounding rock deformation prediction result of the to-be-constructed tunnel. This result can be presented in various forms, for example, a report can be generated to record in detail the to-be-constructed bench section of the to-be-constructed tunnel, the construction time, and the corresponding surrounding rock cumulative deformation prediction value. The result can also be displayed in the form of a visual chart, such as a line chart, to intuitively show the trend of the surrounding rock cumulative deformation prediction value with the construction time.

[0145] Based on the surrounding rock deformation prediction result, the construction personnel can formulate corresponding construction strategies and safety measures in advance. If the prediction result shows that the cumulative deformation value of the surrounding rock is large, it may be necessary to strengthen the support measures, adjust the construction schedule, and the like, to ensure the construction safety and the long-term stability of the tunnel.

[0146] The method provided by the embodiments of the present application comprises the following steps: obtaining a tunnel surrounding rock deformation sample data set comprising a time sequence of cumulative deformation data of each step segment; and performing data fitting on the sample data set according to an initial surrounding rock deformation prediction function, to obtain a surrounding rock deformation prediction function corresponding to each step segment. The function comprehensively considers the surrounding rock deformation result of the previous step segment, the surrounding rock deformation potential of the current step segment, the surrounding rock deformation rate of the current step segment, and the construction time of the current step segment and the like target influencing factors. With these scientific and reasonable technical means, the mapping relationship between the surrounding rock deformation prediction result and the multi-dimensional influencing factors can be accurately established, and then the surrounding rock deformation prediction result of the tunnel to be constructed can be accurately obtained according to the to-be-constructed parameter of the tunnel to be constructed and the surrounding rock deformation prediction function, thereby effectively improving the accuracy and reliability of the surrounding rock deformation prediction in the tunnel step method construction, and providing a strong guarantee for the safety and stability of the tunnel construction.

[0147] Optionally, in order to further improve the accuracy and reliability of the prediction, more influencing factors can be considered. For example, the geological condition factor. Different geological structures and rock properties have great differences in the influence on the surrounding rock deformation. Detailed geological information can be obtained through geological exploration and be taken as a new influencing factor and incorporated into the initial surrounding rock deformation prediction function. When performing data fitting and optimization, the parameters related to the geological conditions also need to be reasonably estimated and adjusted.

[0148] In addition, the construction process factor also has an influence on the surrounding rock deformation. Different excavation methods, support time and support strength and the like construction process parameters will lead to different surrounding rock deformation conditions. The construction process parameters can be taken as additional target influencing factors, and together with the original surrounding rock deformation result of the previous step segment, the surrounding rock deformation potential of the current step segment, the surrounding rock deformation rate of the current step segment, the construction time of the current step segment and the like factors, a more complex initial surrounding rock deformation prediction function is constructed. When processing these new influencing factors, attention should be paid to the unity of the dimension and the matching of the characteristic dimension, to avoid calculation errors.

[0149] Optionally, in actual application, the obtained surrounding rock deformation prediction function can also be dynamically adjusted according to different construction stages and actual situations. As the construction proceeds, some new rules or problems can be found, at which time new data can be collected, data fitting and optimization can be performed again, and the parameter set of the surrounding rock deformation prediction function can be updated, to improve the accuracy and adaptability of the prediction.

[0150] For example, in the early stages of construction, due to the relatively small amount of data, the accuracy of the obtained rock deformation prediction function may be limited. As construction progresses and more rock deformation data is collected, this new data can be added to the tunnel rock deformation sample dataset for further fitting and optimization, resulting in a more accurate rock deformation prediction function. Furthermore, if geological conditions change or construction techniques are adjusted during construction, the prediction function can be adjusted and updated accordingly based on the actual situation.

[0151] For example, the following uses the actual crown settlement deformation data and actual sidewall horizontal deformation data of each step (including the first step, the second step, and the third step) in the tunnel excavation process using the step method as an example to fit the above-mentioned initial surrounding rock deformation prediction function to obtain the corresponding surrounding rock deformation prediction function, which can be used to predict the surrounding rock deformation during subsequent construction.

[0152] Figure 2 This is a schematic diagram of a fitting curve for the settlement deformation data of the arch provided in an embodiment of this application. Figure 2 As shown, it includes fitting diagrams of the top arch settlement deformation for the first stage, the second stage, and the third stage. Each fitting diagram includes monitoring data and a fitting curve for the top arch settlement deformation. The fitting curve is generated by fitting the corresponding monitoring data to the initial surrounding rock deformation prediction function.

[0153] from Figure 2 As can be seen, when the tunnel is excavated using the bench method, the settlement and deformation pattern of the crown arch conforms to the initial surrounding rock deformation prediction function, and the layered fitting effect is good. Correspondingly, the relationship between the settlement and deformation of the crown arch and time after one bench excavation is: One step has no upper step, so the c value is 0; the relationship between the settlement deformation of the top arch after the excavation of the second step and time is as follows: The step above the second step is the first step, and the c value is related to the y value of the first step (e.g., directly substituting the y value of the first step); the relationship between the settlement deformation of the crown arch after the excavation of the third step and time is: The step above the third step is the second step, and the value of c is related to the y-value of the second step (for example, by directly substituting the y-value of the second step).

[0154] Error analysis was performed on the monitoring data of the settlement and deformation of the top arch of the above three steps and the corresponding fitting curves. The calculation showed that the errors were all within 5%, which shows that the fitting accuracy of the target surrounding rock deformation prediction function obtained in this application is high.

[0155] Figure 3 This is a schematic diagram of a fitting curve for the horizontal deformation data of a sidewall provided in an embodiment of this application. Figure 3As shown, it includes a one-step side wall horizontal deformation fitting graph, a two-step side wall horizontal deformation fitting graph, and a three-step side wall horizontal deformation fitting graph. Among them, each fitting graph includes monitoring data of side wall horizontal deformation and a fitting curve, which is generated based on the initial surrounding rock deformation prediction function after fitting the corresponding monitoring data.

[0156] From Figure 3 It can be seen that when the tunnel is excavated by the bench method, the side wall horizontal deformation law conforms to the initial surrounding rock deformation prediction function, and the layer fitting effect is good. Correspondingly, the relationship between the side wall horizontal deformation after one-step excavation and time is , where the one-step has no previous step, and the value of c is 0; the relationship between the side wall horizontal deformation after two-step excavation and time is , where the previous step of the two-step is the one-step, and the value of c is related to the value of y of the one-step (for example, directly substituting the value of y of the one-step); the relationship between the side wall horizontal deformation after three-step excavation and time is , where the previous step of the three-step is the two-step, and the value of c is related to the value of y of the two-step (for example, directly substituting the value of y of the two-step).

[0157] Error analysis is performed on the side wall horizontal deformation monitoring data of the above three steps and the corresponding fitting curves, and it is calculated that the error is within 5%, so it can be seen that the fitting accuracy of the target surrounding rock deformation prediction function obtained by the present application is high.

[0158] After obtaining the target surrounding rock deformation prediction function by fitting, the surrounding rock deformation in the subsequent construction process can be predicted according to the target surrounding rock deformation prediction function obtained by fitting, and the predicted surrounding rock deformation can be verified according to the actual surrounding rock deformation data in the subsequent actual construction process.

[0159] Exemplarily, Figure 4 a data diagram of surrounding rock deformation in subsequent construction is provided for the embodiments of the present application. As Figure 4 shown, the actual surrounding rock deformation data during subsequent construction includes data values of crown settlement deformation and side wall horizontal deformation.

[0160] For the target surrounding rock deformation prediction function obtained by fitting, the predicted values and actual values of the crown settlement deformation, and the error between the two are shown in Table 1 after substituting t=10, 20, 40, 60, 80, and 100 into the target surrounding rock deformation prediction function:

[0161] Table 1

[0162] Construction time t Predicted value Actual value Error 10 149.4863 138 7.7% 20 168.2778 157 6.7% 40 290.9671 291 0.01% 60 338.847 351 3.46% 80 363.0453 380 4.46% 100 368.9674 369 0%

[0163] As shown in Table 1, the error between the predicted value and the actual value output by the target surrounding rock deformation prediction function is within 10%, wherein t=20, 40, 100 are the convergence values of the crown settlement deformation after the first, second and third step excavations, and the predicted value and the actual value are relatively small, so it can be seen that the accuracy of the crown settlement deformation predicted by the target surrounding rock deformation prediction function is high.

[0164] For the target surrounding rock deformation prediction function obtained by fitting, the predicted value and the actual value of the side wall horizontal deformation, and the error between the two are shown in Table 2 after t=10, 20, 40, 60, 80, 100 are respectively substituted into the target surrounding rock deformation prediction function:

[0165] Table 2

[0166] Construction time t Predicted value Actual value Error 10 208.9987 210.65 0.78% 20 251.8026 245.30 2.65% 40 428.2525 417.53 3.53% 60 480.5402 466.85 2.93% 80 503.7955 479.52 5.06% 100 507.2821 485.16 4.56%

[0167] As shown in Table 2, the error between the predicted value and the actual value output by the target surrounding rock deformation prediction function is within 10%, wherein t=20, 40, 100 are the convergence values of the side wall horizontal deformation after the first, second and third step excavations, and the predicted value and the actual value are relatively small, so it can be seen that the accuracy of the side wall horizontal deformation predicted by the target surrounding rock deformation prediction function is high.

[0168] Figure 5 A structure diagram of a surrounding rock deformation prediction system 100 based on tunnel step method construction provided by the embodiment is shown in FIG. 1. Figure 5 As shown in FIG. 1, the processor 120 can be used in the surrounding rock deformation prediction system 100 based on tunnel step method construction, and is used to execute the functions in the present application.

[0169] The surrounding rock deformation prediction system 100 based on tunnel step method construction can be a general server or a special-purpose server, both of which can be used to implement the surrounding rock deformation prediction method based on tunnel step method construction of the present application. Although only one server is shown in the present application, for the sake of convenience, the functions described in the present application can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0170] For example, the surrounding rock deformation prediction system 100 based on the tunnel bench method construction can include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as a disk, a ROM, or a RAM, or any combination thereof. Exemplarily, the surrounding rock deformation prediction system 100 based on the tunnel bench method construction can also include program instructions stored in a ROM, a RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application can be implemented according to these program instructions. The surrounding rock deformation prediction system 100 based on the tunnel bench method construction also includes an input / output (I / O) interface 150 between the computer and other input / output devices.

[0171] For ease of illustration, only one processor is described in the surrounding rock deformation prediction system 100 based on the tunnel bench method construction. However, it should be noted that the surrounding rock deformation prediction system 100 based on the tunnel bench method construction in the present application can also include multiple processors, so the steps performed by one processor described in the present application can also be jointly performed or separately performed by multiple processors. For example, if the processor of the surrounding rock deformation prediction system 100 based on the tunnel bench method construction performs steps A and B, it should be understood that steps A and B can also be jointly performed by two different processors or separately performed in one processor. For example, a first processor performs step A, a second processor performs step B, or the first processor and the second processor jointly perform steps A and B.

[0172] The embodiment of the present application discloses a computer readable storage medium storing a computer program for electronic data exchange, wherein the computer program causes a computer to perform the steps in the surrounding rock deformation prediction method based on the tunnel bench method construction of the preceding embodiments.

[0173] The embodiment of the present application discloses a computer program product comprising a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the surrounding rock deformation prediction method based on the tunnel bench method construction of the preceding embodiments.

[0174] The device embodiments described above are only illustrative, wherein the modules illustrated as separate components can or can not be physically separated, and the components shown as modules can or can not be physical modules, i.e., can be located in one place, or can be distributed to multiple network modules. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0175] Those skilled in the art can clearly understand the implementation of the various embodiments by means of software and necessary general hardware platforms through the above specific description of the embodiments, and of course, the embodiments can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magnetic disk storage, a magnetic tape storage, or any other medium that can be used to store data in a computer readable manner.

[0176] Finally, it should be noted that: the above disclosed is only the preferred embodiment of the present application, only for the description of the technical solutions of the present application, not to limit; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand; it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A surrounding rock deformation prediction method based on tunnel bench construction, characterized by, The method comprises: obtaining a tunnel surrounding rock deformation sample data set, wherein the tunnel surrounding rock deformation sample data set comprises accumulated deformation data time series of each bench section, and the accumulated deformation data time series is obtained based on surrounding rock deformation caused by bench method construction of a reference tunnel; data fitting is performed on the tunnel surrounding rock deformation sample data set according to an initial surrounding rock deformation prediction function, to obtain a surrounding rock deformation prediction function corresponding to each bench section, wherein the initial surrounding rock deformation prediction function represents a mapping relationship between a surrounding rock deformation prediction result and a to-be-predicted time point, and the initial surrounding rock deformation prediction function satisfies the following formula: wherein y is the surrounding rock deformation prediction result, c is a surrounding rock deformation result of a previous bench section, a is a surrounding rock deformation potential of a current bench section, b is a surrounding rock deformation rate of the current bench section, t is a construction time length of the current bench section, and x is the to-be-predicted time point, and the surrounding rock deformation potential is determined based on a range value of surrounding rock deformation data in the tunnel surrounding rock deformation sample data set; according to a to-be-constructed bench section, a target surrounding rock deformation prediction function matched with the to-be-constructed bench section is selected from the surrounding rock deformation prediction functions corresponding to each bench section; the to-be-predicted time point is substituted into the target surrounding rock deformation prediction function, to obtain a surrounding rock accumulated deformation prediction value of the to-be-constructed tunnel at the to-be-predicted time point; the surrounding rock accumulated deformation prediction value is taken as a surrounding rock deformation prediction result of the to-be-constructed tunnel.

2. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 1, characterized in that, The tunnel surrounding rock deformation sample data set is obtained by: monitoring surrounding rock deformation during the bench method construction of the reference tunnel, collecting accumulated deformation data of different bench sections at different construction times, and forming an original monitoring data set; performing data screening processing on the original monitoring data set, to obtain a screened monitoring data set; classifying the screened monitoring data set according to the bench sections, arranging accumulated deformation data belonging to the same bench section in chronological order, and forming accumulated deformation data time series of each bench section; integrating the accumulated deformation data time series of each bench section, to obtain the tunnel surrounding rock deformation sample data set.

3. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 1, characterized in that, The data fitting on the tunnel surrounding rock deformation sample data set according to the initial surrounding rock deformation prediction function, to obtain the surrounding rock deformation prediction function corresponding to each bench section, comprises: preliminarily estimating parameters in the initial surrounding rock deformation prediction function based on the accumulated deformation data time series of each bench section in the tunnel surrounding rock deformation sample data set and a target influencing factor, to obtain a parameter initial estimation value, wherein the target influencing factor comprises a surrounding rock deformation result of a previous bench section, a surrounding rock deformation potential of a current bench section, a surrounding rock deformation rate of the current bench section, and a construction time length of the current bench section; substituting the parameter initial estimation value into the initial surrounding rock deformation prediction function, to obtain a preliminary surrounding rock deformation prediction function; The preliminary surrounding rock deformation prediction function is fitted and optimized by using the tunnel surrounding rock deformation sample data set, parameters of the preliminary surrounding rock deformation prediction function are adjusted, and the preliminary surrounding rock deformation prediction function is adjusted until the error of the cumulative deformation data calculated by the adjusted preliminary surrounding rock deformation prediction function and the cumulative deformation data in the tunnel surrounding rock deformation sample data set is less than a preset threshold value, so as to obtain the surrounding rock deformation prediction function corresponding to each step segment.

4. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 3, characterized in that, The preliminary surrounding rock deformation prediction function is fitted and optimized by using the tunnel surrounding rock deformation sample data set, parameters of the preliminary surrounding rock deformation prediction function are adjusted, and the preliminary surrounding rock deformation prediction function is adjusted until the error of the cumulative deformation data calculated by the adjusted preliminary surrounding rock deformation prediction function and the cumulative deformation data in the tunnel surrounding rock deformation sample data set is less than a preset threshold value, so as to obtain the surrounding rock deformation prediction function corresponding to each step segment. The cumulative deformation data time series of each step segment in the tunnel surrounding rock deformation sample data set is divided into a plurality of data subsegments according to construction stages, and each data subsegment contains cumulative deformation data of a continuous time series; For each data subsegment, the initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm to obtain an adjusted parameter; The predicted cumulative deformation data of the preliminary surrounding rock deformation prediction function on the data subsegment is calculated according to the adjusted parameter; The error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data subsegment is calculated, and the error value is taken as a feedback index for parameter adjustment; If the error value is greater than or equal to the preset threshold value, the adjusted parameter is continuously adjusted based on the feedback index until the error value is less than the preset threshold value, and the adjusted adjusted parameter is determined as the target parameter corresponding to the data subsegment; The target parameters corresponding to all data subsegments are integrated to generate a parameter set of the surrounding rock deformation prediction function corresponding to the step segment, and the surrounding rock deformation prediction function corresponding to each step segment is obtained according to the parameter set.

5. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 4, characterized in that, The initial estimated value of the parameter of the preliminary surrounding rock deformation prediction function is taken as input, and the initial estimated value of the parameter is iteratively adjusted by a nonlinear optimization algorithm to obtain an adjusted parameter, including: The parameter search space of the nonlinear optimization algorithm is determined, and the parameter search space contains a parameter range related to the surrounding rock deformation result of the previous step segment, a parameter range related to the surrounding rock deformation potential of the current step segment, a data range related to the surrounding rock deformation rate of the current step segment, and a parameter range related to the construction time of the current step segment; In the parameter search space, a grid search method is used to preliminarily screen the initial estimated value of the parameter, a plurality of candidate parameter combinations are generated, and a fitting goodness index of the preliminary surrounding rock deformation prediction function corresponding to each candidate parameter combination on the data subsegment is calculated, the fitting goodness index is used to reflect the closeness of the predicted cumulative deformation data and the actual cumulative deformation data; selecting the candidate parameter combination with the highest goodness-of-fit index as an initial iteration parameter, inputting the initial iteration parameter into a gradient descent module of the nonlinear optimization algorithm, and calculating a parameter gradient of the preliminary surrounding rock deformation prediction function with respect to the initial iteration parameter; updating and adjusting the initial iteration parameter according to the direction and size of the parameter gradient until the number of adjustments reaches a preset adjustment number threshold or the modulus of the parameter gradient is less than a preset gradient threshold, to obtain an adjustment parameter.

6. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 4, characterized in that, The calculation of the error value of the predicted cumulative deformation data and the actual cumulative deformation data in the data sub-section, and the error value as a feedback index for parameter adjustment, includes: corresponding matching of the predicted cumulative deformation data and the actual cumulative deformation data in the data sub-section according to the time stamp, to generate a predicted actual data pair sequence; for each predicted actual data pair in the predicted actual data pair sequence, calculating an absolute difference value of the predicted cumulative deformation data and the actual cumulative deformation data, and performing a square operation on the absolute difference value to obtain a square error value; calculating the sum of the square error values of all predicted actual data pairs, and dividing the sum of the square error values by the number of data pairs in the predicted actual data pair sequence to obtain the error value.

7. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 3, characterized in that, The preliminary estimation of the parameters in the initial surrounding rock deformation prediction function based on the cumulative deformation data time series of each step segment in the tunnel surrounding rock deformation sample data set and the target influencing factor, to obtain a parameter initial estimation value, includes: identifying a parameter type set corresponding to the target influencing factor in the initial surrounding rock deformation prediction function, the parameter type set including a first parameter related to the surrounding rock deformation result of the previous step segment, a second parameter related to the surrounding rock deformation potential of the current step segment, a third parameter related to the surrounding rock deformation rate of the current step segment, and a fourth parameter related to the construction time of the current step segment; extracting the endpoint data value of the cumulative deformation data time series of the previous step segment in the tunnel surrounding rock deformation sample data set as the initial estimation value of the first parameter; determining the initial estimation value of the second parameter according to the deformation data range value in the cumulative deformation data time series of the current step segment in the tunnel surrounding rock deformation sample data set, the initial estimation value of the second parameter being in a positive correlation with the deformation data range value; determining the initial estimation value of the third parameter according to the deformation data change amount of the cumulative deformation data time series of the current step segment in the tunnel surrounding rock deformation sample data set, the initial estimation value of the third parameter being in a positive correlation with the deformation data change amount; determining the initial estimation value of the fourth parameter according to the construction duration data of the current step segment in the tunnel surrounding rock deformation sample data set.

8. The surrounding rock deformation prediction method based on the tunnel bench method construction according to claim 3, characterized in that, The initial estimation value of the parameter is substituted into the initial surrounding rock deformation prediction function to obtain a preliminary surrounding rock deformation prediction function, which includes: substituting the parameter initial estimation value into the initial surrounding rock deformation prediction function to obtain a to-be-verified surrounding rock deformation prediction function; According to the validation data subset in the tunnel surrounding rock deformation sample data set, the to-be-validated surrounding rock deformation prediction function, a function output value of the to-be-validated surrounding rock deformation prediction function and a validation error value of corresponding validation data in the validation data subset are obtained; If the validation error value is greater than or equal to a validation error threshold value, a fitting strategy of the to-be-validated surrounding rock deformation prediction function and / or the parameter initial estimated value are adjusted based on the validation error value for re-fitting until the validation error value is less than the validation error threshold value, and the preliminary surrounding rock deformation prediction function is obtained.

9. A surrounding rock deformation prediction system based on tunnel bench construction, characterized by The method comprises a processor and a computer readable storage medium, wherein the computer readable storage medium stores machine executable instructions, and the machine executable instructions are executed by the computer to implement the surrounding rock deformation prediction method based on the tunnel bench method construction according to any one of claims 1-8.

Images