A method and system for predicting the impact of tunnel blasting construction on existing buildings

Abstract

The present invention relates to the technical field of tunnel blasting control, and in particular to a method and system for predicting the impact of tunnel blasting construction on existing buildings. The method of the present invention comprises the following steps: obtaining a tunnel blasting data sample set, and normalizing the tunnel blasting data sample set, wherein the tunnel blasting data sample includes tunnel blasting impact parameters and corresponding blasting vibration speeds; improving the fish swarm search algorithm using an optimization strategy, and combining the back propagation neural network and the improved fish swarm search algorithm to construct a tunnel blasting vibration speed prediction model; training and verifying the tunnel blasting vibration speed prediction model through the tunnel blasting data sample set, and then obtaining a predicted blasting vibration speed based on the blasting impact parameters of the tunnel to be blasted; and completing the prediction of the impact of tunnel blasting construction on existing buildings based on the predicted blasting vibration speed. The present invention solves the problem of quickly and accurately predicting the impact of tunnel blasting construction on existing buildings in different scenarios.

Description

Technical Field

[0001] The invention relates to the technical field of tunnel blasting control, and in particular to a method and system for predicting the impact of tunnel blasting construction on existing buildings. Background Art

[0002] There are many negative problems in the application of blasting in engineering. The shock wave generated by blasting will cause ground vibration, the flying gravel will cause harm to personnel and equipment, the floating dust will pollute the surrounding environment, etc., but the most prominent is the harm of vibration, which directly threatens the safety of personnel and buildings. Therefore, in the blasting design technical plan, the prediction of the impact on the surrounding existing buildings is one of the necessary factors to ensure the feasibility of the technical plan.

[0003] The common prediction method is to explore the law of fitting blasting vibration velocity based on empirical methods such as the degree of attenuation of ground seismic waves, so as to predict the impact of blasting construction on existing buildings through blasting vibration velocity. However, the seismic waves generated during tunnel blasting construction will produce different effects according to different vibration source characteristics, geological environment, and structural characteristics of existing buildings. So far, there is no accurate method to study the law of vibration velocity amplitude attenuation. At present, the law of blasting vibration changes in a certain project or a small range at home and abroad is mainly based on linear fitting of on-site monitoring data and then adjusted and improved, and the prediction significance has great limitations. Therefore, finding a method that can be applied to different types and scenarios to quickly and accurately predict the impact of tunnel blasting construction on existing buildings is an urgent problem to be solved. Summary of the invention

[0004] In view of the shortcomings of existing methods and the needs of practical applications, in order to solve the problem of quickly and accurately predicting the impact of tunnel blasting construction on existing buildings in different types and scenarios. On the one hand, the present invention provides a prediction method for the impact of tunnel blasting construction on existing buildings, including the following steps: obtaining a tunnel blasting data sample set, and normalizing the tunnel blasting data sample set, the tunnel blasting data sample includes tunnel blasting impact parameters and corresponding blasting vibration speed; using an optimization strategy to improve the fish swarm search algorithm, and combining the back propagation neural network and the improved fish swarm search algorithm to construct a tunnel blasting vibration speed prediction model; through the tunnel blasting data sample set, training and verifying the tunnel blasting vibration speed prediction model, and then obtaining the predicted blasting vibration speed according to the blasting impact parameters of the tunnel to be blasted; based on the predicted blasting vibration speed, completing the prediction of the impact of tunnel blasting construction on existing buildings. The present invention optimizes the back propagation neural network by improving the fish swarm search algorithm, and then uses the optimized back propagation neural network to extract multiple tunnel blasting impact parameter features, thereby solving the problem of quickly and accurately predicting the impact of tunnel blasting construction on existing buildings in different types and scenarios.

[0005] Optionally, the tunnel blasting influencing parameters include the maximum section charge, number of holes, hole diameter, hole spacing, row spacing, blasthole depth, filling height, explosive type, hole arrangement method, slot eye resistance line, slot eye charge, slot eye arrangement method, charge method, blast center distance, landform type, rock property, crack development, rock weathering degree, soil compactness. The tunnel blasting influencing parameters of the present invention integrate the blasting scheme technical parameters and geological parameters, which is conducive to improving the prediction accuracy and adaptability of the present invention.

[0006] Optionally, the method of improving the fish school search algorithm by using an optimization strategy and combining the back propagation neural network and the improved fish school search algorithm to construct a tunnel blasting vibration velocity prediction model comprises the following steps:

[0007] Based on the chaos algorithm, the initial fish school is obtained according to the hyperparameters of the back propagation neural network as individuals; the adaptive inertia weight is introduced to improve the fish school search algorithm to obtain the dynamic adaptive fish school search algorithm; the dynamic adaptive fish school search algorithm is optimized by the first optimization strategy to obtain the first dynamic adaptive fish school search algorithm; the dynamic adaptive fish school search algorithm is optimized by the second optimization strategy to obtain the second dynamic adaptive fish school search algorithm; the dynamic adaptive fish school search algorithm is optimized by the third optimization strategy to obtain the third dynamic adaptive fish school search algorithm; the optimal hyperparameters of the back propagation neural network are obtained through the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm, and the tunnel blasting vibration velocity prediction model is constructed using the optimal hyperparameters. The present invention obtains the hyperparameters of the back propagation neural network through the improved fish school search algorithm, and then obtains the optimal tunnel blasting vibration velocity prediction model, which further improves the prediction speed and accuracy of the present invention.

[0008] Optionally, based on a chaos algorithm, an initial fish swarm is obtained according to the hyperparameters of the back propagation neural network as individuals, comprising the following steps:

[0009] The initial value and parameter of the first chaotic map are set to obtain a first layer chaotic sequence; each data of the first layer chaotic sequence is used as the initial value of the second chaotic map to obtain a second layer chaotic sequence; according to the first layer chaotic sequence and the second layer chaotic sequence, the parameters of the first chaotic map and the parameters of the second chaotic map are adjusted to obtain a chaotic mapping model; the initial fish school is obtained by combining the hyperparameters and the chaotic mapping model. The present invention obtains the initial fish school by using two chaotic maps, which is conducive to improving the diversity of the population, and further helps to improve the prediction accuracy and prediction speed of the present invention.

[0010] Optionally, the introduced adaptive inertia weight satisfies the following formula:

[0011]

[0012] in, represents the adaptive inertia weight of the i-th individual at the t-th iteration, T represents the maximum number of iterations, ω max represents the maximum weight coefficient, ω min represents the minimum weight coefficient, represents the maximum fitness value at the tth iteration, The adaptive inertia weight provided by the present invention speeds up the search efficiency for the optimal value and further improves the prediction ability of the present invention.

[0013] Optionally, the dynamic adaptive fish school search algorithm is optimized by using a first optimization strategy to obtain a first dynamic adaptive fish school search algorithm that satisfies the following formula:

[0014]

[0015] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, λ represents a random number that follows the Levy flight distribution, T represents the maximum number of iterations, represents the historical optimal position of the ith individual in the first t iterations.

[0016] Optionally, the dynamic adaptive fish school search algorithm is optimized by using a second optimization strategy to obtain a second dynamic adaptive fish school search algorithm that satisfies the following formula:

[0017]

[0018] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, α represents the first learning factor, G t represents the optimal position at the tth iteration, β represents the second learning factor, H t represents the optimal position of the previous t iteration history, and L represents a random number that obeys the Levy flight distribution.

[0019] Optionally, the dynamic adaptive fish school search algorithm is optimized by using a third optimization strategy to obtain a third dynamic adaptive fish school search algorithm, which satisfies the following formula:

[0020]

[0021] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, T represents the maximum number of iterations, R1 and R2 represent random vectors in [0, 1], represents the position of the random individual at the tth iteration, b represents a random number that satisfies the Fibonacci spiral, and τ represents a random number of [-1, 1]. The dynamic adaptive fish school search algorithm obtained by the present invention through three optimization strategies effectively improves the global search ability and local search ability of the algorithm of the present invention, and further improves the prediction efficiency of the present invention.

[0022] Optionally, obtaining the optimal hyperparameters of the back propagation neural network through the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm comprises the following steps:

[0023] A fitness evaluation function is set, and the initial fish school is evaluated by using the fitness evaluation function; the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm, and the third dynamic adaptive fish school search algorithm are respectively executed according to the evaluation results, and the execution results are screened to obtain a new fish school; the new fish school is used as a new initial fish school, and iterative optimization is performed until the iteration termination condition is met to obtain the optimal hyperparameters of the back propagation neural network. The present invention obtains the optimal hyperparameters through multiple iterations, which is conducive to obtaining the optimal tunnel blasting vibration velocity prediction model, thereby improving the prediction accuracy and prediction speed of the present invention.

[0024] In the second aspect, in order to be able to efficiently execute the prediction method of the impact of tunnel blasting construction on existing buildings provided by the present invention, the present invention also provides a prediction system for the impact of tunnel blasting construction on existing buildings, including a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the prediction method of the impact of tunnel blasting construction on existing buildings as described in the first aspect of the present invention. The prediction system for the impact of tunnel blasting construction on existing buildings of the present invention has a compact structure and stable performance, and can stably execute the prediction method of the impact of tunnel blasting construction on existing buildings provided by the present invention, further improving the overall applicability and practical application capabilities of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flow chart of a method for predicting the impact of tunnel blasting construction on existing buildings provided by an embodiment of the present invention;

[0026] Figure 2 A framework diagram of a prediction system for the impact of tunnel blasting construction on existing buildings provided by an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of the structure of a device for predicting the impact of tunnel blasting construction on existing buildings provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0029] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0030] See also Figure 1 In order to solve the problem of quickly and accurately predicting the impact of tunnel blasting construction on existing buildings in different types and scenarios, the present invention provides a prediction method for the impact of tunnel blasting construction on existing buildings, such as Figure 1 As shown, the method comprises the following steps:

[0031] S1. Obtain a tunnel blasting data sample set, and perform normalization processing on the tunnel blasting data sample set, wherein the tunnel blasting data sample includes tunnel blasting influencing parameters and corresponding blasting vibration speeds.

[0032] Specifically, a sample set of tunnel blasting data is obtained from public databases and relevant literature. The influencing parameters of tunnel blasting include the maximum section charge, number of holes, hole diameter, hole spacing, row spacing, blasthole depth, filling height, explosive type, hole layout method, slot eye resistance line, slot eye charge, slot eye layout method, charge method, blast center distance, landform type, rock properties, crack development, rock weathering degree, and soil compactness.

[0033] Furthermore, the normalization process satisfies the following formula:

[0034]

[0035] Among them, x' i represents the normalized value, x i Represents the original value, min(x i ) represents the minimum value in the data sample, max(x i ) represents the maximum value in the data sample.

[0036] S2. The fish swarm search algorithm is improved by using the optimization strategy, and a tunnel blasting vibration velocity prediction model is constructed by combining the back propagation neural network and the improved fish swarm search algorithm.

[0037] In the embodiment, the step S2 uses the optimization strategy to improve the fish school search algorithm, and combines the back propagation neural network and the improved fish school search algorithm to construct a tunnel blasting vibration velocity prediction model, including the following steps:

[0038] S21. Based on the chaos algorithm, an initial fish school is obtained according to the hyperparameters of the back propagation neural network as individuals.

[0039] Specifically, based on the chaos algorithm, the initial fish school is obtained according to the hyperparameters of the back propagation neural network as individuals, including the following steps:

[0040] S211, setting the initial value and parameter of the first chaotic map to obtain a first-layer chaotic sequence; the first chaotic map is one of Sine mapping, Logistic mapping, Kent mapping or Tent mapping;

[0041] S212. Use each data of the first-layer chaotic sequence as the initial value of the second chaotic map to obtain the second-layer chaotic sequence; the second chaotic map is one of Sine mapping, Logistic mapping, Kent mapping or Tent mapping. In some other embodiments, the first chaotic map and the second chaotic map can also be set to other chaotic maps.

[0042] S213, adjusting the parameters of the first chaotic mapping and the parameters of the second chaotic mapping according to the first chaotic sequence and the second chaotic sequence to obtain a chaotic mapping model; in one embodiment, adjusting the parameters of the first chaotic mapping and the parameters of the second chaotic mapping to satisfy the following formula:

[0043] Among them, A represents the parameters of the Logistic mapping, a represents the parameters of the Sine mapping, and X km Indicates that X k0 is the value obtained by iterating the initial value through Sine mapping, X k0 Indicates the kth value of the initial value after Logistic mapping.

[0044] S214. Combining the hyperparameters and the chaotic mapping model, obtaining the initial fish school.

[0045] The hyperparameters are mapped to chaotic variables CX, and the initial position is in [a, b]. a and b can be selected according to the actual situation. The value range of chaotic variable CX is (0, 1). Then the chaotic variable CX is subjected to chaotic operation through the chaotic mapping model to obtain chaotic variable CX1. The chaotic variable CX1 is mapped to the ordinary variable CX0 again, that is, the chaotic variable CX1 is mapped from the interval (0, 1) to the interval [a, b]. The ordinary variable CX0 is the initial fish school.

[0046] S22. Introduce adaptive inertia weight to improve the fish school search algorithm and obtain a dynamic adaptive fish school search algorithm.

[0047] In the embodiment, the introduced adaptive inertia weight satisfies the following formula:

[0048]

[0049] in, represents the adaptive inertia weight of the i-th individual at the t-th iteration, T represents the maximum number of iterations, ω max represents the maximum weight coefficient, ωmin represents the minimum weight coefficient, represents the maximum fitness value at the tth iteration, Represents the minimum fitness value at the tth iteration.

[0050] S23. Optimize the dynamic adaptive fish school search algorithm using a first optimization strategy to obtain a first dynamic adaptive fish school search algorithm.

[0051] Specifically, the dynamic adaptive fish school search algorithm is optimized using the first optimization strategy to obtain a first dynamic adaptive fish school search algorithm that satisfies the following formula:

[0052]

[0053] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, λ represents a random number that follows the Levy flight distribution, T represents the maximum number of iterations, represents the historical optimal position of the ith individual in the first t iterations.

[0054] S24. Optimizing the dynamic adaptive fish school search algorithm using a second optimization strategy to obtain a second dynamic adaptive fish school search algorithm.

[0055] Specifically, the dynamic adaptive fish school search algorithm is optimized using the second optimization strategy to obtain a second dynamic adaptive fish school search algorithm that satisfies the following formula:

[0056]

[0057] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, α represents the first learning factor, G t represents the optimal position at the tth iteration, β represents the second learning factor, H t represents the optimal position of the previous t iteration history, and L represents a random number that obeys the Levy flight distribution.

[0058] It should be understood that the first learning factor and the second learning factor can affect the convergence speed and search performance. When the first learning factor is larger, it is more inclined to rely on the optimal position of this iteration for search; and when the second learning factor is larger, it is more inclined to rely on the historical optimal position of the group. The value range of the first learning factor and the second learning factor is [1.4, 3].

[0059] S25. Optimize the dynamic adaptive fish school search algorithm using a third optimization strategy to obtain a third dynamic adaptive fish school search algorithm.

[0060] Specifically, the dynamic adaptive fish school search algorithm is optimized using the third optimization strategy to obtain a third dynamic adaptive fish school search algorithm that satisfies the following formula:

[0061]

[0062] in, represents the position of the i-th individual at the t+1th iteration, represents the adaptive inertia weight of the i-th individual at the t+1-th iteration, The position of the i-th individual at the t-th iteration, T represents the maximum number of iterations, R1 and R2 represent random vectors in [0, 1], represents the position of the random individual at the tth iteration, b represents a random number that satisfies the Fibonacci spiral, and τ represents a random number in [-1, 1].

[0063] S26. Obtain the optimal hyperparameters of the back propagation neural network through the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm, and use the optimal hyperparameters to construct the tunnel blasting vibration velocity prediction model.

[0064] Specifically, step S26 obtains the optimal hyperparameters of the back propagation neural network through the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm, and constructs the tunnel blasting vibration velocity prediction model using the optimal hyperparameters, including the following steps:

[0065] S261, setting a fitness evaluation function, and using the fitness evaluation function to evaluate the initial fish school;

[0066] In the embodiment, the fitness evaluation function can be set to a Griewank function, a Rastrigin function, an Ackley function, or a root mean square error function.

[0067] S262, respectively executing the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm according to the evaluation results, and filtering the execution results to obtain a new fish school;

[0068] The first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm are respectively executed on the initial fish school, and then sorted and screened according to fitness, and the top one third is used as the new fish school.

[0069] S263, taking the new fish school as a new initial fish school, performing iterative optimization until an iteration termination condition is met, and obtaining the optimal hyperparameters of the back propagation neural network.

[0070] Specifically, the iteration termination conditions include reaching the maximum number of iterations or satisfying that the fitness difference of the optimal individuals obtained in the three previous iterations is less than a preset threshold.

[0071] S3. Training and verifying the tunnel blasting vibration velocity prediction model through the tunnel blasting data sample set, and then obtaining the predicted blasting vibration velocity according to the blasting influencing parameters of the tunnel to be blasted.

[0072] The tunnel blasting data sample set is randomly divided into a training set and a validation set. The tunnel blasting vibration velocity prediction model is trained using the training set, and then the validation set is used to obtain the validation prediction value through the tunnel blasting vibration velocity prediction model. The mean square error (MSE) and correlation coefficient (R) between the validation prediction value of the tunnel blasting vibration velocity prediction model and the validation set are calculated. 2 ), set the error threshold of the tunnel blasting vibration velocity prediction model to verify the accuracy of the tunnel blasting vibration velocity prediction model; if the error of the tunnel blasting vibration velocity prediction model is greater than the error threshold, collect more data sets to train the tunnel blasting vibration velocity prediction model until the error of the tunnel blasting vibration velocity prediction model is less than the set error threshold.

[0073] Obtain the technical solution and geological parameters related to the tunnel to be blasted, input them into the trained tunnel blasting vibration velocity prediction model, and obtain the blasting vibration velocity of the particle where the existing building is located.

[0074] S4. Based on the predicted blasting vibration speed, the impact of tunnel blasting construction on existing buildings is predicted.

[0075] Specifically, the impact on existing buildings is predicted based on the predicted blasting vibration velocity in combination with the "Regulations on Blasting Safety".

[0076] See also Figure 2In an embodiment, in order to efficiently execute the prediction method of the impact of tunnel blasting construction on existing buildings provided by the present invention, the present invention also provides a prediction system of the impact of tunnel blasting construction on existing buildings, including: an input device, an output device, a processor, and a memory, wherein the input device, the output device, the processor, and the memory are interconnected, and the memory contains program instructions, and the program instructions are used for the steps of the prediction method of the impact of tunnel blasting construction on existing buildings. The prediction system of the impact of tunnel blasting construction on existing buildings of the present invention has a compact structure and stable performance, and can stably execute the prediction method of the impact of tunnel blasting construction on existing buildings of the present invention, further improving the overall applicability and practical application ability of the present invention.

[0077] In an embodiment, the processor may be a central processing unit (CPU), which may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. An input device may be used to obtain data information. An output device may be used to output the results obtained by storing program instructions contained in a computer program in a memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the type of device.

[0078] In yet another alternative embodiment, see Figure 3 In order to efficiently execute the prediction method of the impact of tunnel blasting construction on existing buildings provided by the present invention, this embodiment also provides a prediction device for the impact of tunnel blasting construction on existing buildings, such as Figure 3 As shown, including:

[0079] The memory 10 is used to store computer programs; the processor 20 is used to execute the computer programs to implement the above-mentioned method for predicting the impact of tunnel blasting construction on existing buildings.

[0080] Memory 10 , processor 20 , communication interface 31 and communication bus 32 . Memory 10 , processor 20 , communication interface 31 all communicate with each other via communication bus 32 .

[0081] In an embodiment, the memory 10 is used to store one or more program instructions, and the memory 10 may store program instructions for implementing the following functions: obtaining a tunnel blasting data sample set and normalizing the tunnel blasting data sample set, wherein the tunnel blasting data sample includes tunnel blasting influencing parameters and corresponding blasting vibration speeds; improving the fish school search algorithm using an optimization strategy, and combining the back propagation neural network and the improved fish school search algorithm to construct a tunnel blasting vibration speed prediction model; training and verifying the tunnel blasting vibration speed prediction model through the tunnel blasting data sample set, and then obtaining a predicted blasting vibration speed based on the blasting influencing parameters of the tunnel to be blasted; and completing the prediction of the impact of tunnel blasting construction on existing buildings based on the predicted blasting vibration speed.

[0082] In a possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function, etc.; the data storage area may store data created during use.

[0083] In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include NVRAM. The memory stores an operating system and operating instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks. The processor 20 may be a central processing unit (CPU), a specific application integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, and the processor 20 may be a microprocessor or may also be any conventional processor, etc. The processor 20 may call a program stored in the memory 10. The communication interface 31 may be an interface of a communication module for connecting to other devices or systems.

[0084] Of course, it should be noted that Figure 3 The structure shown does not constitute a limitation on the prediction device for the impact of tunnel blasting construction on existing buildings in this embodiment. In actual applications, the prediction device for the impact of tunnel blasting construction on existing buildings may include Figure 3 More or fewer components than shown, or combinations of certain components.

[0085] The embodiment further provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for predicting the impact of tunnel blasting construction on existing buildings are implemented.

[0086] The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0087] In summary, the present invention optimizes the back propagation neural network by improving the fish school search algorithm, and then uses the optimized back propagation neural network to extract multiple tunnel blasting influence parameter characteristics, thereby solving the problem of quickly and accurately predicting the impact of tunnel blasting construction on existing buildings in different types and scenarios. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for predicting the impact of tunnel blasting construction on existing buildings, characterized in that: The method for predicting the impact of tunnel blasting construction on existing buildings comprises the following steps: Acquire a tunnel blasting data sample set, and perform normalization processing on the tunnel blasting data sample set, wherein the tunnel blasting data sample includes tunnel blasting influencing parameters and corresponding blasting vibration speed; The fish swarm search algorithm was improved by using optimization strategy, and a tunnel blasting vibration velocity prediction model was constructed by combining the back propagation neural network and the improved fish swarm search algorithm. The tunnel blasting vibration velocity prediction model is trained and verified by the tunnel blasting data sample set, and then the predicted blasting vibration velocity is obtained according to the blasting influencing parameters of the tunnel to be blasted; Based on the predicted blasting vibration speed, the impact of tunnel blasting construction on existing buildings is predicted; The method of improving the fish swarm search algorithm by using an optimization strategy and combining the back propagation neural network with the improved fish swarm search algorithm to construct a tunnel blasting vibration velocity prediction model includes the following steps: Based on the chaos algorithm, the initial fish school is obtained according to the hyperparameters of the back propagation neural network as individuals; Adaptive inertia weight is introduced to improve the fish school search algorithm and obtain a dynamic adaptive fish school search algorithm; Optimizing the dynamic adaptive fish school search algorithm using a first optimization strategy to obtain a first dynamic adaptive fish school search algorithm; Optimizing the dynamic adaptive fish school search algorithm using a second optimization strategy to obtain a second dynamic adaptive fish school search algorithm; Optimizing the dynamic adaptive fish school search algorithm using a third optimization strategy to obtain a third dynamic adaptive fish school search algorithm; Obtaining optimal hyperparameters of the back propagation neural network through the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm, and constructing the tunnel blasting vibration velocity prediction model using the optimal hyperparameters; Based on the chaos algorithm, the initial fish group is obtained according to the hyperparameters of the back propagation neural network as individuals, including the following steps: Setting the initial value and parameters of the first chaotic map to obtain the first layer chaotic sequence; Using each data of the first-layer chaotic sequence as the initial value of the second chaotic map to obtain a second-layer chaotic sequence; According to the first layer chaotic sequence and the second layer chaotic sequence, adjusting the parameters of the first chaotic mapping and the parameters of the second chaotic mapping to obtain a chaotic mapping model; Combining the hyperparameters with the chaotic mapping model to obtain the initial fish school; The introduced adaptive inertia weight satisfies the following formula: in, Indicates The individual in The adaptive inertia weight at the iteration, represents the maximum number of iterations, represents the maximum weight coefficient, represents the minimum weight coefficient, Indicates The maximum fitness value at the iteration, Indicates The minimum fitness value at the iteration; The first optimization strategy is used to optimize the dynamic adaptive fish school search algorithm to obtain a first dynamic adaptive fish school search algorithm, which satisfies the following formula: in, Indicates The first iteration The location of an individual, Indicates The individual in The adaptive inertia weight at the iteration, No. The first iteration The location of an individual, represents random numbers that follow the Levy flight distribution, represents the maximum number of iterations, Before In the iteration The historical optimal position of each individual; The second optimization strategy is used to optimize the dynamic adaptive fish school search algorithm to obtain a second dynamic adaptive fish school search algorithm, which satisfies the following formula: in, Indicates The iteration The location of an individual, Indicates The individual in The adaptive inertia weight at the iteration, No. The iteration The location of an individual, represents the first learning factor, Indicates The optimal position at the iteration, represents the second learning factor, Before The best position in the iteration history, represents random numbers that follow the Levy flight distribution; The third optimization strategy is used to optimize the dynamic adaptive fish school search algorithm to obtain a third dynamic adaptive fish school search algorithm, which satisfies the following formula: in, Indicates The iteration The location of an individual, Indicates The individual in The adaptive inertia weight at the iteration, No. The iteration The location of an individual, represents the maximum number of iterations, and represents a random vector in [0,1], Indicates The position of the random individual at the iteration, represents a random number that satisfies the Fibonacci spiral, Represents a random number in [-1,1]; The method of obtaining the optimal hyperparameters of the back propagation neural network by using the initial fish school, the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm comprises the following steps: Setting a fitness evaluation function, and using the fitness evaluation function to evaluate the initial fish school; Execute the first dynamic adaptive fish school search algorithm, the second dynamic adaptive fish school search algorithm and the third dynamic adaptive fish school search algorithm respectively according to the evaluation results, and filter the execution results to obtain a new fish school; The new fish group is used as a new initial fish group, and iterative optimization is performed until an iteration termination condition is met, thereby obtaining the optimal hyperparameters of the back propagation neural network.

2. A method for predicting the impact of tunnel blasting construction on existing buildings according to claim 1, characterized in that: The tunnel blasting influencing parameters include the maximum section charge, number of holes, hole diameter, hole spacing, row spacing, blasthole depth, filling height, type of explosives, hole layout method, groove eye resistance line, groove eye charge, groove eye layout method, charging method, blast center distance, landform type, rock properties, crack development, rock weathering degree, and soil compactness.

3. A prediction system for the impact of tunnel blasting construction on existing buildings, characterized in that: The prediction system for the impact of tunnel blasting construction on existing buildings includes: an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. The memory includes program instructions, and the program instructions are used to execute the prediction method for the impact of tunnel blasting construction on existing buildings described in any one of claims 1-2.

Images