Multi-source monitoring data analysis method and system based on active earthquake source

Abstract

The present invention provides a multi-source monitoring data analysis method and system based on an active earthquake source, which relates to the field of artificial intelligence technology. By introducing a pre-trained seismic wave enhancement analysis network, the seismic wave enhancement analysis network is composed of multiple cascaded model function layers, each layer is embedded with a multi-stage knowledge transfer unit, and deep learning and efficient analysis of seismic wave data are achieved. The multi-stage knowledge transfer learning mechanism can not only effectively extract the key features in the waveform data of the time period, but also gradually enhance the understanding and recognition ability of the seismic wave sequence through layer-by-layer knowledge transfer and node fusion. This method not only improves the accuracy and robustness of seismic wave data analysis, but also significantly enhances the sensitivity and prediction accuracy of earthquake monitoring. By generating enhanced waveform data and strengthening learning seismic wave sequences, more reliable and data support is provided for earthquake early warning, seismological research and earthquake disaster assessment.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a multi-source monitoring data analysis method and system based on active seismic sources. Background Art

[0002] As a natural disaster, earthquake prediction and monitoring are crucial for reducing casualties and property losses. Traditional earthquake monitoring methods rely primarily on hardware devices such as seismometers to directly record and analyze seismic waves. However, these methods are often limited by equipment accuracy, deployment density, and data processing capabilities, making it difficult to achieve comprehensive and accurate monitoring of seismic activity.

[0003] With advances in science and technology, particularly the rapid development of big data and artificial intelligence, earthquake monitoring data analysis methods have significantly improved. Multi-source monitoring technology, based on active seismic sources, has become a key earthquake monitoring tool. This technology actively excites seismic waves and simultaneously records their propagation at multiple monitoring stations, thereby obtaining richer and more comprehensive earthquake information.

[0004] However, efficient and accurate analysis of multi-source monitoring data remains a pressing issue. Traditional data analysis methods often struggle to effectively process large-scale, multi-dimensional seismic wave data, leading to omissions or misinterpretations of key information. Furthermore, the complexity and nonlinear nature of seismic wave data pose significant challenges to data analysis. Summary of the Invention

[0005] In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides a multi-source monitoring data analysis method based on active seismic sources, the method comprising:

[0006] Acquire waveform data of multiple time periods of candidate seismic wave sequences of multi-source monitoring data of an active seismic source, and for each time period of waveform data, load the waveform data into a pre-trained seismic wave enhancement analysis network, wherein the seismic wave enhancement analysis network includes multiple model function layers, each model function layer is cascaded, and each model function layer includes a multi-stage knowledge transfer unit, and the multi-stage knowledge transfer unit includes more than two stages of knowledge transfer learning;

[0007] In the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loading data of the target model functional layer, and the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of each stage is node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the target model functional layer, and the multi-stage knowledge transfer learning vector is used to generate the reinforcement learning knowledge vector data corresponding to the target model functional layer; wherein the first stage knowledge transfer learning is used to learn the loading data of the target model functional layer, and the knowledge transfer learning of each remaining stage is used to learn the remaining node knowledge vector data except the key node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous stage; wherein, when the target model functional layer is the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the basic knowledge vector data of the time period waveform data; when the target model functional layer is the model functional layer after the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer;

[0008] Based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, the reinforcement waveform data corresponding to the waveform data of the time period is generated; based on the reinforcement waveform data corresponding to each of the waveform data of the time period, the reinforcement learning seismic wave sequence corresponding to the candidate seismic wave sequence is generated.

[0009] On the other hand, an embodiment of the present invention also provides a multi-source monitoring data analysis system based on active seismic sources, including a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0010] Based on the above aspects, the embodiment of the present application introduces a pre-trained seismic wave enhancement analysis network, which is composed of multiple cascaded model function layers, each layer of which is embedded with a multi-stage knowledge transfer unit, to achieve deep learning and efficient analysis of seismic wave data. The multi-stage knowledge transfer learning mechanism can not only effectively extract the key features in the waveform data of the time period, but also gradually enhance the understanding and recognition ability of the seismic wave sequence through layer-by-layer knowledge transfer and node fusion. This method not only improves the accuracy and robustness of seismic wave data analysis, but also significantly enhances the sensitivity and prediction accuracy of earthquake monitoring. By generating enhanced waveform data and strengthening learning seismic wave sequences, more reliable and data support is provided for earthquake early warning, seismological research and earthquake disaster assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a schematic diagram of the execution flow of the multi-source monitoring data analysis method based on active seismic sources provided by an embodiment of the present invention.

[0012] Figure 2 Schematic diagram of the hardware architecture of a multi-source monitoring data analysis system based on active seismic sources provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The present invention will be described in detail below with reference to the accompanying drawings. Figure 1 This is a flow chart of a multi-source monitoring data analysis method based on active seismic sources provided by an embodiment of the present invention. The multi-source monitoring data analysis method based on active seismic sources is introduced in detail below.

[0014] Step S110, obtain waveform data of multiple time periods of candidate seismic wave sequences of multi-source monitoring data of active seismic sources, and for each time period of waveform data, load the waveform data of the time period into a pre-trained seismic wave enhancement analysis network, wherein the seismic wave enhancement analysis network includes multiple model function layers, each model function layer is cascaded, and each of the model function layers includes a multi-stage knowledge transfer unit, and the multi-stage knowledge transfer unit includes more than two stages of knowledge transfer learning.

[0015] In this embodiment, in the context of earthquake monitoring, an active seismic source refers to a device that artificially generates seismic waves to obtain underground structural information. This device can generate seismic waves at different locations and times, thereby generating multi-source monitoring data. This source monitoring data includes a large number of candidate seismic wave sequences, each of which is composed of waveform data from multiple time periods.

[0016] For example, in a specific earthquake monitoring area, multiple active seismic source transmitters are deployed. These devices transmit seismic wave signals underground at predetermined intervals and according to a predetermined excitation method. After these seismic waves propagate and reflect underground, they are received by multiple seismic monitoring sensors distributed throughout the area. The data collected by the server from these sensors is the multi-source monitoring data of the active seismic source.

[0017] Within this multi-source monitoring data, there are many candidate seismic wave sequences. For example, a single candidate seismic wave sequence might contain waveform data from different time periods throughout the entire process of the seismic wave, from excitation to propagation, reflection, and reception. This waveform data reflects the characteristics of the seismic waves at different moments, such as amplitude, frequency, and phase.

[0018] After the server obtains the waveform data for these time periods, it loads each period into a pre-trained seismic wave intensification analysis network. This network is a complex neural network structure composed of multiple model function layers. Each model function layer is cascaded. Furthermore, each model function layer includes a multi-stage knowledge transfer unit, which has knowledge transfer learning capabilities of more than two stages. This multi-stage knowledge transfer learning capability allows the network to better leverage existing knowledge and experience when processing period waveform data, thereby improving its analysis and processing capabilities.

[0019] For example, assume the seismic wave intensification analysis network has three model functional layers: functional layer A, functional layer B, and functional layer C. When the server obtains waveform data for a period of time, it first loads this waveform data into functional layer A. The multi-stage knowledge transfer unit in functional layer A then begins preliminary knowledge transfer learning on this waveform data. Utilizing its internal multi-stage knowledge transfer learning mechanism, it analyzes the characteristics of this waveform data from different perspectives and levels, laying the foundation for subsequent processing and analysis.

[0020] Step S120: In the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loaded data of the target model functional layer. The key node knowledge vector data in the knowledge transfer learning vector generated by each stage of knowledge transfer learning is node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the target model functional layer. The multi-stage knowledge transfer learning vector is used to generate reinforcement learning knowledge vector data corresponding to the target model functional layer. The first stage of knowledge transfer learning is used to learn the loaded data of the target model functional layer, and the remaining stages of knowledge transfer learning are used to learn the remaining node knowledge vector data, excluding the key node knowledge vector data, generated by node-splitting the knowledge transfer learning vector generated by the previous stage of knowledge transfer learning. When the target model functional layer is the first of the multiple model functional layers, the loaded data of the target model functional layer is the basic knowledge vector data of the waveform data for the time period. When the target model functional layer is a model functional layer subsequent to the first of the multiple model functional layers, the loaded data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to the previous model functional layer.

[0021] In this embodiment, when operating in the multi-stage knowledge migration unit of the target model functional layer, it is assumed that the target model functional layer is functional layer B (in the aforementioned seismic wave enhancement analysis network composed of functional layer A, functional layer B and functional layer C).

[0022] First, because functional layer B is not the first model functional layer (the first is functional layer A), its loaded data is the reinforcement learning knowledge vector data corresponding to functional layer A. After processing the time period waveform data, functional layer A generates reinforcement learning knowledge vector data. This data contains various feature information and learned knowledge about the time period waveform data after functional layer A has processed it.

[0023] In the multi-stage knowledge transfer unit of functional layer B, multi-stage knowledge transfer learning begins. Assume that this multi-stage knowledge transfer unit has three stages of knowledge transfer learning, namely stage 1, stage 2, and stage 3.

[0024] In phase 1, functional layer B performs first-order knowledge transfer learning on the reinforcement learning knowledge vector data obtained from functional layer A. For example, this first-order knowledge transfer learning in functional layer B may further refine certain characteristics of the seismic waves output by functional layer A (such as the energy distribution within a specific frequency range) and generate a first-order knowledge transfer learning vector. This first-order knowledge transfer learning vector contains the new feature information and knowledge learned in this phase.

[0025] Next, we enter Phase 2, the transitional knowledge transfer unit. In this phase, we first need to obtain the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the previous level (i.e., Phase 1). Node splitting involves decomposing the knowledge transfer learning vector generated by Phase 1 into different node knowledge vector data according to certain rules. For example, assuming that the knowledge transfer learning vector in Phase 1 contains comprehensive information about seismic wave frequency, amplitude, and phase, node splitting can be used to split it into node knowledge vector data about frequency, node knowledge vector data about amplitude, and node knowledge vector data about phase. Knowledge transfer learning is then performed on these remaining node knowledge vector data (here, assuming that the node knowledge vector data about amplitude and phase, excluding certain key nodes, is included). This process is similar to in-depth learning of a portion of a knowledge system during the learning process. Through learning in this phase, a transitional knowledge transfer learning vector is generated. This transitional knowledge transfer learning vector contains the deeper knowledge about seismic wave amplitude and phase learned during this transitional phase.

[0026] Finally, in stage three, which is the last knowledge transfer unit, similarly, the remaining node knowledge vector data is first obtained by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level (stage two). Then, the final stage of knowledge transfer learning is performed on these remaining node knowledge vector data (assuming that they are more detailed partial node knowledge vector data about phase) to generate the key node knowledge vector data generated by the final stage of knowledge transfer learning. This key node knowledge vector data may contain the most critical feature information about the seismic wave phase learned at this stage.

[0027] Next, the key node knowledge vector data within the knowledge transfer learning vector generated by knowledge transfer learning in each stage (stage 1, stage 2, and stage 3) undergoes node fusion. Node fusion can be understood as combining the most important knowledge learned in each stage to form a complete, more valuable knowledge set. In this way, a multi-stage knowledge transfer learning vector corresponding to functional layer B is generated. This multi-stage knowledge transfer learning vector represents the comprehensive knowledge output of functional layer B after multi-stage knowledge transfer learning and will be used to generate reinforcement learning knowledge vector data corresponding to functional layer B.

[0028] Assume that the target model functional layer is functional layer A (i.e., the first model functional layer). In this case, the loading data for the target model functional layer is the basic knowledge vector data of the time period waveform data. This basic knowledge vector data is obtained through the regularization conversion unit and node expansion unit in the seismic wave enhancement analysis network.

[0029] For example, the server first performs regular conversion on the period waveform data through the regular conversion unit. Assume that the period waveform data is an irregular sequence of seismic wave amplitude and frequency data containing various noises and interferences. The regular conversion unit will generate regular waveform data according to predetermined rules, such as adjusting the amplitude value range to a specific interval, normalizing the frequency data, etc. Then, the regular waveform data is expanded through the node expansion unit. For example, the regular waveform data originally only contains the amplitude and frequency values ​​at a few simple time points. Through the node expansion unit, more nodes can be added between these time points according to a certain algorithm (such as an interpolation algorithm), thereby generating basic knowledge vector data containing more detailed information.

[0030] In the multi-stage knowledge transfer unit of functional layer A, it is assumed that there are two stages of knowledge transfer learning. In the first stage of knowledge transfer learning, functional layer A directly learns the basic knowledge vector data, learning the basic characteristics of the waveform data in this period, such as the initial amplitude change trend and frequency distribution range, to generate the first stage of knowledge transfer learning vector. Then, in the second stage of knowledge transfer learning, the first stage of knowledge transfer learning vector is node-split to obtain the remaining node knowledge vector data except for the key node knowledge vector data. This remaining node knowledge vector data is then learned. Finally, the key node knowledge vector data of the two stages are node-fused to generate the multi-stage knowledge transfer learning vector corresponding to functional layer A. This multi-stage knowledge transfer learning vector is used to generate the reinforcement learning knowledge vector data corresponding to functional layer A.

[0031] Step S130, based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, generate the reinforcement waveform data corresponding to the waveform data of the time period, and based on the reinforcement waveform data corresponding to each of the waveform data of the time period, generate the reinforcement learning seismic wave sequence corresponding to the candidate seismic wave sequence.

[0032] In the process of generating enhanced waveform data corresponding to the period waveform data based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, it is assumed that the last model functional layer is functional layer C (in the aforementioned seismic wave enhancement analysis network composed of functional layer A, functional layer B and functional layer C).

[0033] After the previous multi-stage knowledge transfer learning and related processing, the functional layer C generates the corresponding reinforcement learning knowledge vector data. This data contains rich information about the waveform data of the time period after being processed by the entire seismic wave reinforcement analysis network.

[0034] For example, the server needs to generate enhanced waveform data corresponding to the time period waveform data. First, the server combines the basic knowledge vector data of the time period waveform data with the reinforcement learning knowledge vector data corresponding to functional layer C. Assume that the basic knowledge vector data of the time period waveform data contains some basic characteristic information of the initial seismic wave (such as the original amplitude and frequency information), and the reinforcement learning knowledge vector data corresponding to functional layer C contains more accurate and detailed characteristic information about the seismic wave obtained after processing by the entire network (such as corrected amplitude and frequency information, more precise phase information, etc.). These two vector data are combined to generate the target reinforcement learning knowledge vector data.

[0035] Then, based on the target reinforcement learning knowledge vector data, the enhanced waveform data corresponding to the time period waveform data is determined. In this process, the seismic wave enhancement analysis network also includes an inverse regularization conversion unit and a node compression unit.

[0036] The server first uses the node compression unit to perform node compression on the target reinforcement learning knowledge vector data. Assuming the target reinforcement learning knowledge vector data is a vector containing a large amount of detailed node information after multi-layer processing, the node compression unit compresses the number of nodes in this vector to the same number as the period waveform data according to a specific algorithm, based on factors such as the original number of nodes in the period waveform data. For example, if the original period waveform data has 100 nodes, and the target reinforcement learning knowledge vector data has 200 nodes after processing, the node compression unit selects the 100 most important nodes or merges and compresses the 200 nodes into 100 nodes to generate the target knowledge vector data.

[0037] Next, the target knowledge vector data is subjected to an inverse regularization conversion by the inverse regularization conversion unit. Inverse regularization conversion is the inverse process of regularization conversion. If, in step S110, the regularization conversion unit has performed operations such as adjusting the amplitude value range to a specific interval and normalizing the frequency data on the original period waveform data, the inverse regularization conversion unit will restore the amplitude and frequency data in the target knowledge vector data to a representation similar to the original period waveform data, thereby generating enhanced waveform data corresponding to the period waveform data.

[0038] Finally, based on the enhanced waveform data corresponding to the waveform data for each time period, a reinforcement learning seismic wave sequence corresponding to the candidate seismic wave sequence is generated. For example, for a candidate seismic wave sequence containing waveform data for 10 time periods, the server generates corresponding enhanced waveform data for each time period according to the above process. These 10 enhanced waveform data are then combined together in their original order to generate the reinforcement learning seismic wave sequence corresponding to this candidate seismic wave sequence. Compared to the original candidate seismic wave sequence, this reinforcement learning seismic wave sequence has more accurate and reliable seismic wave characteristic information and can be better used for earthquake-related analysis and research, such as the study of seismic wave propagation characteristics and the inference of underground geological structures.

[0039] Based on the above steps, the embodiment of the present application introduces a pre-trained seismic wave enhancement analysis network, which is composed of multiple cascaded model function layers, each layer of which is embedded with a multi-stage knowledge transfer unit, to achieve deep learning and efficient analysis of seismic wave data. The multi-stage knowledge transfer learning mechanism can not only effectively extract the key features in the waveform data of the time period, but also gradually enhance the understanding and recognition ability of the seismic wave sequence through layer-by-layer knowledge transfer and node fusion. This method not only improves the accuracy and robustness of seismic wave data analysis, but also significantly enhances the sensitivity and prediction accuracy of earthquake monitoring. By generating enhanced waveform data and strengthening learning seismic wave sequences, more reliable and data support is provided for earthquake early warning, seismological research and earthquake disaster assessment.

[0040] In a possible implementation, the seismic wave intensification analysis network further includes a regularization conversion unit and a node expansion unit, and the step of determining the basic knowledge vector data of the time period waveform data includes:

[0041] The regularization conversion unit performs regularization conversion on the period waveform data to generate regularized waveform data.

[0042] The node expansion unit performs node expansion on the regularized waveform data to generate basic knowledge vector data of the time period waveform data.

[0043] In this embodiment, the regularization conversion unit first performs regularization conversion on the period waveform data to generate regularized waveform data. For example, after obtaining the period waveform data of a candidate seismic wave sequence from the multi-source monitoring data of an active seismic source, these period waveform data may have various different representation forms and numerical ranges. The regularization conversion unit processes this data according to predefined rules. If the amplitude data in the period waveform data ranges from -100 to 100, and the amplitude value ranges of waveform data from different periods vary significantly, the regularization conversion unit converts it to a uniform value range of 0 to 1, making the amplitude data of waveform data from different periods comparable. For frequency data, if it is in Hertz and the values ​​are relatively dispersed, the regularization conversion unit normalizes it according to a certain proportional relationship, so that the frequency data also follows specific rules. After these operations, the original period waveform data is converted into regularized waveform data. This regularized waveform data is more regular in data format and numerical range, facilitating subsequent processing.

[0044] Next, the node expansion unit performs node expansion on the regularized waveform data to generate basic knowledge vector data for the period waveform data. Since the regularized waveform data may not have dense data points after regularization, the node expansion unit supplements this information. For example, regularized waveform data records a data point every 1 second on the time axis, which may not be detailed enough for seismic wave analysis. The node expansion unit inserts new data points between adjacent data points based on a specific algorithm, such as linear interpolation or an interpolation algorithm based on a seismic wave propagation model. If the original regularized waveform data has 10 data points in a certain period, after node expansion, this number may increase to 20. These additional data points, together with the original data points, form a richer set of information. This regularized waveform data, containing more data points, is further processed and organized to form the basic knowledge vector data for the period waveform data. This basic knowledge vector data contains more detailed and comprehensive information about the period waveform data and serves as the foundation for the subsequent multi-stage knowledge transfer learning in the first model function layer of the seismic wave enhancement analysis network. During this process, the server processes the data strictly according to the functional logic of the regularized conversion unit and the node expansion unit to ensure the accuracy and validity of the generated basic knowledge vector data, so as to provide a reliable data foundation for the entire seismic wave enhancement analysis process.

[0045] In a possible implementation, when the target model function layer is a model function layer subsequent to the first model function layer among the multiple model function layers, step S120 includes:

[0046] Step S121, in the first-order knowledge transfer unit of the multi-stage knowledge transfer unit, first-order knowledge transfer learning is performed on the reinforcement learning knowledge vector data corresponding to the previous model function layer of the target model function layer to generate a first-order knowledge transfer learning vector.

[0047] Step S122: In the transition knowledge transfer unit of the multi-stage knowledge transfer unit, the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level is obtained, and knowledge transfer learning is performed on the remaining node knowledge vector data to generate a transition knowledge transfer learning vector.

[0048] Step S123: In the last knowledge transfer unit of the multi-stage knowledge transfer unit, the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level is obtained, and the last stage of knowledge transfer learning is performed on the remaining node knowledge vector data to generate the key node knowledge vector data generated by the last stage of knowledge transfer learning.

[0049] Step S124 , performing node fusion on the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning in each stage, and generating a multi-stage knowledge transfer learning vector corresponding to the functional layer of the target model.

[0050] In a possible implementation, based on the above steps S121 and S124, the method further includes:

[0051] Step A110 , performing node splitting on the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, and generating node knowledge vector data of each feature node.

[0052] Step A120, based on the arrangement order of each feature node, starting from the first feature node, the node knowledge vector data of the target number of nodes with a later node order are used as the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, and the node knowledge vector data of the remaining nodes are used as the remaining node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level.

[0053] In this embodiment, first, in the first-order knowledge transfer unit of the multi-stage knowledge transfer unit, the server performs first-order knowledge transfer learning on the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer, thereby generating a first-order knowledge transfer learning vector. Taking functional layers A, B, and C in the previously mentioned seismic wave enhancement analysis network as an example, when the target model functional layer is functional layer B, its previous model functional layer is functional layer A. After a series of processing, functional layer A generates reinforcement learning knowledge vector data. This data contains various information obtained from functional layer A's processing of the time period waveform data, such as the adjusted and analyzed characteristic information of the seismic wave amplitude, frequency, phase, etc. The first-order knowledge transfer unit in functional layer B receives this reinforcement learning knowledge vector data from functional layer A. First-order knowledge transfer learning is a process of in-depth exploration and analysis of this data. It analyzes, extracts, and recombines the various elements in the reinforcement learning knowledge vector data of functional layer A based on the learning mechanism and parameter settings of functional layer B itself. For example, if the reinforcement learning knowledge vector data of functional layer A contains information such as the amplitude variation trend of seismic waves in different time periods and the frequency distribution range, the first-order knowledge transfer unit may focus on the new representation and potential relationship of this information in functional layer B, thereby generating a first-order knowledge transfer learning vector. This first-order knowledge transfer learning vector may contain the characteristic information about seismic waves after preliminary adjustment and reorganization, providing a new starting point for subsequent learning.

[0054] Next, in the transitional knowledge transfer unit of the multi-stage knowledge transfer unit, the server obtains the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the previous level's knowledge transfer learning. Knowledge transfer learning is then performed on this remaining node knowledge vector data to generate a transitional knowledge transfer learning vector. Taking functional layer B as an example, the previous level is the first-order knowledge transfer learning vector generated by the first-order knowledge transfer unit. This vector contains various characteristic information about seismic waves, and the server performs node splitting on it. Node splitting decomposes the data in this vector according to different characteristics or attributes according to pre-defined rules. For example, the first-order knowledge transfer learning vector contains information about the amplitude, frequency, phase, and propagation time of the seismic wave. Node splitting may separate this information into different node knowledge vector data, such as amplitude-related node knowledge vector data and frequency-related node knowledge vector data. Then, based on a specific algorithm or rule, a portion of the node knowledge vector data is identified as key node knowledge vector data, and the remainder is the remaining node knowledge vector data. The transitional knowledge transfer unit performs knowledge transfer learning on this remaining node knowledge vector data. This learning process utilizes the specific learning algorithms and parameters of the transitional knowledge transfer unit in functional layer B to further analyze and process the information contained in the remaining node knowledge vector data. For example, the remaining frequency-related node knowledge vector data may be adjusted and optimized based on the general frequency patterns previously learned in the entire seismic wave enhancement analysis network and the characteristics of the waveform data in the current period, thereby generating a transitional knowledge transfer learning vector. This vector incorporates the results of learning the remaining node knowledge vector data in the transitional knowledge transfer unit and, to a certain extent, focuses more on certain specific seismic wave characteristic information.

[0055] Then, in the last knowledge transfer unit of the multi-stage knowledge transfer unit, the server will also obtain the remaining node knowledge vector data generated by node splitting of the knowledge transfer learning vector generated by the previous level of knowledge transfer learning, and then perform the final stage of knowledge transfer learning on these remaining node knowledge vector data, thereby generating the key node knowledge vector data generated by the final stage of knowledge transfer learning. Continuing with functional layer B as an example, the previous level is the transition knowledge transfer learning vector generated by the transition knowledge transfer unit. This vector will be node-split again to obtain different remaining node knowledge vector data. This data will be received by the last knowledge transfer unit, and then the final stage of knowledge transfer learning will be performed. This learning process will conduct a final in-depth mining and adjustment of these remaining node knowledge vector data based on the special algorithm of the last knowledge transfer unit in functional layer B and the previously accumulated knowledge about seismic waves. For example, if the previous transitional knowledge transfer learning vector obtained residual node knowledge vector data about the change of seismic wave phase within a specific time period after node splitting, the final knowledge transfer unit may accurately analyze and adjust this residual node knowledge vector data based on the overall understanding of the entire seismic wave enhancement analysis network on phase changes and the special requirements of the waveform data in the current period, thereby generating the key node knowledge vector data generated by the final stage of knowledge transfer learning. This key node knowledge vector data contains the very critical information about seismic wave characteristics learned in the final stage.

[0056] Finally, the server performs node fusion on the key node knowledge vector data within the knowledge transfer learning vector generated by each stage of knowledge transfer learning, thereby generating a multi-stage knowledge transfer learning vector corresponding to the target model's functional layer. For example, for functional layer B, the key node knowledge vector data from the first-stage knowledge transfer learning vector, the key node knowledge vector data from the transitional knowledge transfer learning vector, and the key node knowledge vector data generated by the final stage of knowledge transfer learning are fused. This fusion process is not a simple addition, but rather an organic combination based on factors such as the importance and relevance of each key node knowledge vector data. Through a specific algorithm, the information contained in these key node knowledge vector data is integrated to form a complete, multi-stage knowledge transfer learning vector that integrates the key information from each stage. This vector is a key outcome of the multi-stage knowledge transfer learning process for functional layer B and serves as the basis for further processing to generate the reinforcement learning knowledge vector data corresponding to functional layer B.

[0057] This process also includes node splitting of the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level to generate node knowledge vector data for each feature node, and based on the order of arrangement of each feature node, starting from the first feature node, the node knowledge vector data of the target number of nodes with a later node order are used as the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, and the node knowledge vector data of the remaining nodes are used as the remaining node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level. Taking the first-order knowledge transfer learning vector in functional layer B as an example, this vector contains information on multiple feature nodes, such as the amplitude feature nodes, frequency feature nodes, phase feature nodes, etc. of the seismic wave, arranged in a certain order. Assume that according to a predetermined rule, the node knowledge vector data of a certain number (target number) of nodes with a later node order are used as key node knowledge vector data. If there are 10 characteristic nodes in the first-order knowledge transfer learning vector, the node knowledge vector data of the three nodes with the lowest node order (such as characteristic nodes related to phase, propagation time, and energy attenuation) is defined as the key node knowledge vector data, and the node knowledge vector data of the remaining seven nodes (such as characteristic nodes related to amplitude, frequency, etc.) is the remaining node knowledge vector data. This method of determining the key node knowledge vector data and the remaining node knowledge vector data helps to process node information of different importance in a targeted manner during the different stages of knowledge transfer learning, thereby improving the efficiency and effectiveness of the entire multi-stage knowledge transfer learning process, ensuring that the multi-stage knowledge transfer learning vector corresponding to the target model functional layer can accurately reflect the characteristic information of the seismic wave, and providing strong support for subsequent seismic wave analysis and processing.

[0058] In one possible embodiment, each model functional layer also includes a node nonlinear conversion unit based on a feature self-focusing strategy, and the method also includes: loading the multi-stage knowledge transfer learning vector generated by the multi-stage knowledge transfer unit in the target model functional layer into the node nonlinear conversion unit in the target model functional layer, performing node nonlinear conversion processing on the multi-stage knowledge transfer learning vector according to the feature self-focusing strategy, and generating a multi-stage knowledge transfer learning vector for node nonlinear conversion.

[0059] The step of generating the reinforcement learning knowledge vector data corresponding to the target model functional layer includes: in the target model functional layer, based on the multi-stage knowledge transfer learning vector corresponding to the target model functional layer, the multi-stage knowledge transfer learning vector of the node nonlinear conversion and the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer, generating the reinforcement learning knowledge vector data corresponding to the target model functional layer.

[0060] In one possible implementation, the multi-stage knowledge transfer learning vector corresponding to the target model functional layer includes a node knowledge transfer learning vector of each feature node, and performing node nonlinear conversion processing on the multi-stage knowledge transfer learning vector according to the feature self-focusing strategy to generate a node nonlinear conversion multi-stage knowledge transfer learning vector includes:

[0061] A focusing process of a feature self-focusing strategy is performed on the multi-stage knowledge transfer learning vector to determine the attention weight corresponding to each node in the multi-stage knowledge transfer learning vector.

[0062] The attention weights and node knowledge transfer learning vectors corresponding to each feature node are respectively multiplied to generate a multi-stage knowledge transfer learning vector for node nonlinear transformation.

[0063] In one possible implementation, the generating of the reinforcement learning knowledge vector data corresponding to the target model functional layer based on the multi-stage knowledge transfer learning vector corresponding to the target model functional layer, the multi-stage knowledge transfer learning vector of the node nonlinear conversion, and the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer includes:

[0064] The multi-stage knowledge transfer learning vector corresponding to the target model functional layer and the multi-stage knowledge transfer learning vector of the node nonlinear conversion are combined to generate combined knowledge vector data.

[0065] Perform node conversion on the combined knowledge vector data to generate converted knowledge vector data, and combine the converted knowledge vector data with the reinforcement learning knowledge vector data corresponding to the previous model function layer of the target model function layer to generate reinforcement learning knowledge vector data corresponding to the target model function layer.

[0066] In this embodiment, first, the server loads the multi-stage knowledge transfer learning vector generated by the multi-stage knowledge transfer unit in the target model functional layer into the node nonlinear conversion unit in the target model functional layer, and then performs node nonlinear conversion processing on the multi-stage knowledge transfer learning vector according to the feature self-focusing strategy to generate a multi-stage knowledge transfer learning vector for node nonlinear conversion. Taking the functional layer in the seismic wave enhancement analysis network mentioned above as an example, it is assumed that the target model functional layer is functional layer B. After a series of knowledge transfer learning, the multi-stage knowledge transfer unit in functional layer B generates a multi-stage knowledge transfer learning vector. This multi-stage knowledge transfer learning vector contains rich characteristic information about seismic waves, which exists in the form of node knowledge transfer learning vectors of each characteristic node. When this multi-stage knowledge transfer learning vector is loaded into the node nonlinear conversion unit of functional layer B, it begins to be processed according to the feature self-focusing strategy.

[0067] During this process, the multi-stage knowledge transfer learning vector is focused using a feature self-focusing strategy to determine the attention weights corresponding to each node in the multi-stage knowledge transfer learning vector. For example, feature nodes in the multi-stage knowledge transfer learning vector may include seismic wave amplitude feature nodes, frequency feature nodes, phase feature nodes, and propagation time feature nodes. For amplitude feature nodes, the server analyzes their importance in the overall seismic wave analysis using the feature self-focusing strategy. If the amplitude change in the waveform data for the current period is highly important for seismic wave identification and analysis, then this amplitude feature node is assigned a higher attention weight. This attention weight is determined based on the knowledge of seismic wave characteristics accumulated by the seismic wave enhancement analysis network during previous learning and processing, as well as the characteristics of the waveform data for the current period. Similarly, feature nodes such as frequency, phase, and propagation time are assigned corresponding attention weights based on their respective importance.

[0068] Next, the server performs a dot product of the attention weight corresponding to each feature node with the node knowledge transfer learning vector, generating a multi-stage knowledge transfer learning vector for node nonlinear transformation. For example, for the amplitude feature node, assuming its attention weight is 0.8, its node knowledge transfer learning vector contains a series of numerical information about the amplitude, such as the amplitude values ​​at different time periods. This attention weight of 0.8 is dot-producted with each value in the node knowledge transfer learning vector for the amplitude feature node to obtain the attention-adjusted numerical information about the amplitude. The same operation is performed for other feature nodes such as frequency, phase, and propagation time, dot-producting their respective attention weights with the corresponding node knowledge transfer learning vector. Finally, these dot-producted feature node information is combined to generate a multi-stage knowledge transfer learning vector for node nonlinear transformation. Compared to the original multi-stage knowledge transfer learning vector, this vector more prominently highlights the information of important feature nodes and better reflects the key requirements of the current analysis.

[0069] Afterwards, in the target model functional layer, reinforcement learning knowledge vector data corresponding to the target model functional layer is generated based on the multi-stage knowledge transfer learning vector corresponding to the target model functional layer, the multi-stage knowledge transfer learning vector for node nonlinear conversion, and the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer. Still taking functional layer B as an example, the multi-stage knowledge transfer learning vector corresponding to functional layer B contains the characteristic information of the seismic wave after multi-stage knowledge transfer learning, while the multi-stage knowledge transfer learning vector for node nonlinear conversion is information that is more focused on important features after node nonlinear conversion processing. The reinforcement learning knowledge vector data corresponding to functional layer A, the previous model functional layer of functional layer B, contains the result information of functional layer A's seismic wave processing.

[0070] The server first combines the multi-stage knowledge transfer learning vector corresponding to the target model functional layer and the multi-stage knowledge transfer learning vector of the node nonlinear conversion to generate the combined knowledge vector data. For example, the information about the amplitude, frequency, phase, etc. of the seismic wave in the multi-stage knowledge transfer learning vector corresponding to functional layer B and the amplitude, frequency, phase, etc. information after attention adjustment in the multi-stage knowledge transfer learning vector of the node nonlinear conversion are combined according to the corresponding feature nodes. If the amplitude information in the multi-stage knowledge transfer learning vector is [1, 2, 3], and the amplitude information in the multi-stage knowledge transfer learning vector of the node nonlinear conversion is [0.8, 1.6, 2.4] after attention adjustment, then the combined amplitude information may be [1.8, 3.6, 5.4] (this is just a simple example, the actual combination method may be more complex). The same combination operation is performed on other feature nodes such as frequency and phase to obtain the combined knowledge vector data.

[0071] Then, the server performs node conversion on the combined knowledge vector data to generate converted knowledge vector data. Node conversion is to adjust each node in the combined knowledge vector data according to the conversion rules within functional layer B. For example, for the amplitude information [1.8, 3.6, 5.4] in the combined knowledge vector data, it may be converted to [2, 4, 6] according to specific conversion rules (this is just an example). Corresponding conversion operations are also performed on other feature nodes such as frequency and phase to obtain the converted knowledge vector data. Finally, the server combines the converted knowledge vector data with the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer to generate the reinforcement learning knowledge vector data corresponding to the target model functional layer. For example, the amplitude, frequency, phase and other information in the converted knowledge vector data of functional layer B and the relevant information in the reinforcement learning knowledge vector data corresponding to functional layer A are combined according to certain rules. If the amplitude information in the reinforcement learning knowledge vector data corresponding to functional layer A is [0.5, 1, 1.5], and the amplitude information in the knowledge vector data after conversion of functional layer B is [2, 4, 6], then the amplitude information after combination may be [2.5, 5, 7.5] (this is just an example). The same combination operation is performed on other feature nodes such as frequency and phase, and finally the reinforcement learning knowledge vector data corresponding to functional layer B is obtained. This reinforcement learning knowledge vector data contains the complete analysis information about seismic waves after comprehensively considering the multi-stage knowledge transfer learning, node nonlinear conversion and the processing results of functional layer A of functional layer B, providing a more accurate and comprehensive basis for the subsequent seismic wave analysis process. As a result, in each model functional layer of the seismic wave reinforcement analysis network, the understanding and analysis capabilities of seismic wave characteristics are gradually improved, so that seismic wave data can be processed more effectively, providing strong support for related work such as earthquake research and monitoring.

[0072] In one possible embodiment, step S130 includes: combining the basic knowledge vector data of the period waveform data and the reinforcement learning knowledge vector data corresponding to the last model function layer to generate target reinforcement learning knowledge vector data, and determining the reinforcement waveform data corresponding to the period waveform data based on the target reinforcement learning knowledge vector data.

[0073] The seismic wave enhancement analysis network further includes an inverse regularization conversion unit and a node compression unit. The determination of the enhanced waveform data corresponding to the waveform data of the time period based on the target reinforcement learning knowledge vector data includes:

[0074] The target reinforcement learning knowledge vector data is node compressed by the node compression unit to generate target knowledge vector data, and the number of nodes in the target knowledge vector data is the same as the number of nodes in the period waveform data.

[0075] The target knowledge vector data is subjected to an inverse regularization conversion by the inverse regularization conversion unit to generate enhanced waveform data corresponding to the period waveform data.

[0076] In this embodiment, the basic knowledge vector data of the period waveform data and the reinforcement learning knowledge vector data corresponding to the last model functional layer are first combined to generate the target reinforcement learning knowledge vector data. Taking the functional layers A, B, and C in the previous seismic wave enhancement analysis network as an example, it is assumed that functional layer C is the last model functional layer. The basic knowledge vector data of the period waveform data is obtained after the period waveform data is initially regularized and expanded. It contains relatively basic but comprehensive information about the period waveform data, such as a detailed vector representation of the amplitude, frequency, phase, and other information of the seismic wave in the initial state. The reinforcement learning knowledge vector data corresponding to functional layer C is the high-level feature information about the seismic wave obtained after comprehensive processing of multi-stage knowledge transfer learning within functional layer C, nonlinear node conversion, and related information of the previous functional layer (functional layer B). The combination of these two vector data is not a simple addition, but the corresponding seismic wave feature information in the two vector data is integrated according to the rules pre-set by the seismic wave enhancement analysis network. For example, the amplitude information in the basic knowledge vector data may be the amplitude values ​​of the original, uncooked seismic waves at different times, while the amplitude information in the reinforcement learning knowledge vector data may be the result of adjustment, supplementation, or re-evaluation of these amplitude values ​​after analysis by functional layer C. When combined, these two types of amplitude information are fused according to a certain algorithm, such as weighted summation based on their respective weights or merging according to specific logical rules. Similarly, similar operations are performed on other seismic wave characteristic information such as frequency and phase, and finally the target reinforcement learning knowledge vector data is generated. This target reinforcement learning knowledge vector data combines the basic information of the time period waveform data with the advanced information deeply processed by the last functional layer of the entire seismic wave enhancement analysis network, providing a comprehensive data foundation for determining the enhanced waveform data corresponding to the time period waveform data.

[0077] Next, the server determines the enhanced waveform data corresponding to the time period waveform data based on the target reinforcement learning knowledge vector data. This process involves the inverse regularization conversion unit and node compression unit in the seismic wave enhancement analysis network.

[0078] The server first uses a node compression unit to compress the target reinforcement learning knowledge vector data to generate the target knowledge vector data. After the previous combination step, the target reinforcement learning knowledge vector data may contain more node information than the original period waveform data. This node information is generated for detailed analysis and processing within the seismic wave enhancement analysis network. However, to generate enhanced waveform data corresponding to the period waveform data, the number of nodes must be adjusted to match the number of nodes in the period waveform data. For example, the target reinforcement learning knowledge vector data may contain 100 nodes of information about various seismic wave characteristics, while the original period waveform data only has 50 nodes. The node compression unit compresses the target reinforcement learning knowledge vector data according to specific algorithms and rules. This algorithm may merge or discard relatively unimportant node information based on an assessment of the importance of seismic wave characteristics. For example, nodes corresponding to subtle fluctuations that have little impact on the overall seismic wave characteristics may be merged or removed. After node compression, the number of nodes in the generated target knowledge vector data is the same as the number of nodes in the period waveform data, for example, it becomes 50 nodes, and this target knowledge vector data still retains the key characteristic information of the seismic wave, but exists in a more compact form that matches the number of nodes in the original period waveform data.

[0079] Then, the server performs an inverse regularization conversion on the target knowledge vector data through the inverse regularization conversion unit to generate enhanced waveform data corresponding to the time period waveform data. In the previous step, the time period waveform data was subjected to a regularization conversion by the regularization conversion unit, such as normalizing the amplitude and frequency, adjusting the data to a specific value range, and other operations. Now, the inverse regularization conversion unit is to perform the opposite operation. If the amplitude value range is normalized from the original -100 to 100 to 0 to 1 during the regularization conversion, then during the inverse regularization conversion, it will be restored to the original value range or the corresponding reasonable value range based on the amplitude information in the target knowledge vector data and the previous regularization conversion algorithm. Similarly, corresponding inverse regularization conversion operations are also performed for other seismic wave characteristic information such as frequency and phase. After the inverse regularization conversion, the target knowledge vector data is converted into enhanced waveform data corresponding to the time period waveform data. This enhanced waveform data is based on the waveform data of the original period and is obtained after deep processing by the seismic wave enhancement analysis network. It is more accurate and can better reflect the true characteristics of seismic waves. It has more important value in the analysis and research of seismic waves and possible subsequent applications (such as earthquake early warning, underground structure detection, etc.) because it integrates the basic information of the original data and the advanced feature information after network learning analysis, and can provide more reliable data support for related technologies and research.

[0080] In one possible implementation, the method includes:

[0081] Step S101 , obtaining sample learning data, each of the sample learning data includes waveform data of a first template period and waveform data of a second template period, wherein the data value of the waveform data of the first template period is greater than the data value of the waveform data of the second template period.

[0082] Step S102: Load the waveform data of the second template period into the initialized seismic wave enhancement analysis network, wherein the seismic wave enhancement analysis network includes multiple model function layers, each model function layer is cascaded, and each of the model function layers includes a multi-stage knowledge transfer unit, and the multi-stage knowledge transfer unit includes more than two orders of knowledge transfer learning.

[0083] In step S103, in the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loaded data of the target model functional layer. The key node knowledge vector data in the knowledge transfer learning vector generated by each stage of knowledge transfer learning is node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the target model functional layer. The multi-stage knowledge transfer learning vector is used to generate reinforcement learning knowledge vector data corresponding to the target model functional layer. The first stage of knowledge transfer learning is used to learn the loaded data of the target model functional layer, and the remaining stages of knowledge transfer learning are used to learn the remaining node knowledge vector data, excluding the key node knowledge vector data, generated by node-splitting the knowledge transfer learning vector generated by the previous stage of knowledge transfer learning. When the target model functional layer is the first of the multiple model functional layers, the loaded data of the target model functional layer is the basic knowledge vector data of the waveform data of the second template period. When the target model functional layer is a model functional layer subsequent to the first of the multiple model functional layers, the loaded data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to the previous model functional layer.

[0084] Step S104: Based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, generate template enhanced waveform data corresponding to the second template period waveform data, generate a target training cost based on the error between the template enhanced waveform data and the first template period waveform data, and optimize the network parameters of the seismic wave enhancement analysis network based on the target training cost.

[0085] Wherein, step S101 includes:

[0086] Step S1011: Acquire a first template seismic wave sequence, where the first template seismic wave sequence includes a plurality of first template time period waveform data.

[0087] Step S1012: performing noise disturbance processing on the first template seismic wave sequence to generate a noise-perturbed seismic wave sequence.

[0088] Step S1013: performing feature compression on the noise-perturbed seismic wave sequence to generate a second template seismic wave sequence.

[0089] Step S1014 : generating sample learning data based on first template period waveform data and second template period waveform data in the same period identifier obtained from the first template seismic wave sequence and the second template seismic wave sequence.

[0090] In this embodiment, during the training process of the seismic wave analysis and seismic wave enhancement analysis network, the first step is to obtain sample learning data. For example, a first template seismic wave sequence can be obtained first, and this first template seismic wave sequence contains multiple first template time period waveform data. For example, in a specific seismic monitoring area, the relatively pure, accurate and representative seismic wave data collected by high-precision seismic monitoring equipment constitutes a first template seismic wave sequence. These first template time period waveform data have high data value, and they can more accurately reflect the true characteristics of the seismic waves in the area at different time periods, including accurate amplitude, frequency, phase and other information.

[0091] Then, the server performs noise perturbation processing on the first template seismic wave sequence. This operation is to simulate various interference factors that may exist in actual seismic wave monitoring. In actual environments, seismic waves will be interfered with by many factors during propagation, such as the unevenness of underground geological structures, weak vibrations caused by human activities, etc. The server adds noise to the first template seismic wave sequence through a specific algorithm to generate a seismic wave sequence after noise perturbation processing. For example, according to a predetermined noise model, random fluctuations within a certain range may be added to the amplitude of the original seismic wave, or a small offset may be introduced in the frequency, so that this seismic wave sequence is closer to the complex seismic wave conditions that may actually be received.

[0092] Next, the server performs feature compression on the seismic wave sequence after noise disturbance processing. Since the seismic wave sequence after noise disturbance processing may contain a lot of redundant information or overly complex features, feature compression is required to facilitate subsequent learning and analysis. The feature compression process is to extract the most representative and key feature information in the seismic wave sequence through a specific algorithm, thereby generating a second template seismic wave sequence. For example, when processing the frequency characteristics of seismic waves, the frequency information that fluctuates frequently within a small range but has little effect on the overall seismic wave characteristics may be merged or simplified, retaining those frequency characteristics that have an important impact on the overall shape and propagation characteristics of the seismic wave.

[0093] Finally, the server generates sample learning data based on the waveform data of the first template period and the waveform data of the second template period in the same time period from the first template seismic wave sequence and the second template seismic wave sequence. This means that for each specific time period, there is a corresponding set of waveform data of the first template period (with higher data value) and waveform data of the second template period (relatively low-value data after noise perturbation and feature compression). Such a pair of data constitutes a sample learning data. These sample learning data will be used to train the seismic wave enhancement analysis network.

[0094] After acquiring the sample learning data, the server begins training the initialized seismic wave enhancement analysis network using this data. The server loads the waveform data from the second template period into the initialized seismic wave enhancement analysis network. This network consists of multiple model function layers, each of which is cascaded and includes multi-stage knowledge transfer units. The multi-stage knowledge transfer units include two or more stages of knowledge transfer learning.

[0095] When the waveform data of the second template period is loaded into the network, in the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loaded data of the target model functional layer. Take a seismic wave enhancement analysis network containing three model functional layers (functional layer A, functional layer B, and functional layer C) as an example. If the target model functional layer is functional layer A (i.e., the first model functional layer), the loaded data of the target model functional layer is the basic knowledge vector data of the waveform data of the second template period. This basic knowledge vector data is obtained by performing the aforementioned operations such as regularization transformation and node expansion on the waveform data of the second template period. In the multi-stage knowledge transfer unit of functional layer A, it is assumed that this unit has three stages of knowledge transfer learning.

[0096] In the first stage of knowledge transfer learning, functional layer A directly learns the basic knowledge vector data of the waveform data from the second template period. This learning process analyzes various seismic wave characteristics contained in this vector data, such as the initial amplitude distribution and frequency range. Through internal calculation and analysis mechanisms, it generates the first stage of knowledge transfer learning vector. This vector contains the results of learning the basic knowledge vector data in this stage, such as the initial understanding and analysis of amplitude and frequency characteristics.

[0097] Then, in the subsequent stage of knowledge transfer learning, for the second stage, the knowledge transfer learning vector generated by the knowledge transfer learning in the first stage must first be node-split to obtain the remaining node knowledge vector data besides the key node knowledge vector data. Assuming that the knowledge transfer learning vector in the first stage contains comprehensive information about the amplitude, frequency, phase, etc. of the seismic wave, through node splitting, it can be split into node knowledge vector data about amplitude, node knowledge vector data about frequency, etc. according to different characteristics. Then, these remaining node knowledge vector data are learned. This learning process will further mine and analyze the information in these remaining node knowledge vector data according to the algorithm and parameter settings within functional layer A. For example, a more in-depth study of the specific frequency range in the frequency node knowledge vector data will be conducted to generate the knowledge transfer learning vector for the second stage.

[0098] In the third phase, the knowledge transfer learning vectors generated in the second phase are similarly split into nodes to obtain the remaining node knowledge vector data. This data is then used to learn and generate the key node knowledge vector data generated in the third phase of knowledge transfer learning. This key node knowledge vector data may contain important information learned in this phase about a key feature of seismic waves (such as the phase variation pattern under specific conditions).

[0099] Finally, the key node knowledge vector data within the knowledge transfer learning vector generated by knowledge transfer learning in each stage (three stages) is fused. Node fusion combines the most important knowledge learned in each stage according to specific rules. For example, the key information about amplitude in the first stage, frequency in the second stage, and phase in the third stage are organically combined to form a multi-stage knowledge transfer learning vector corresponding to functional layer A. This vector is used to generate the reinforcement learning knowledge vector data corresponding to functional layer A.

[0100] If the target model functional layer is functional layer B (i.e., the model functional layer after the first model functional layer), the loaded data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to functional layer A. In the multi-stage knowledge transfer unit of functional layer B, multi-stage knowledge transfer learning is also performed. In the first-order knowledge transfer unit of the multi-stage knowledge transfer unit, first-order knowledge transfer learning is performed on the reinforcement learning knowledge vector data corresponding to functional layer A to generate a first-order knowledge transfer learning vector. This process is similar to the knowledge transfer learning in functional layer A, except that the input data becomes the reinforcement learning knowledge vector data of functional layer A. Functional layer B will analyze and process this data according to its own learning mechanism, extract new feature information, and generate a first-order knowledge transfer learning vector.

[0101] In the transition knowledge transfer unit of the multi-stage knowledge transfer unit, the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level (the first-order knowledge transfer unit) is obtained, and knowledge transfer learning is performed on this remaining node knowledge vector data to generate a transition knowledge transfer learning vector. For example, the first-order knowledge transfer learning vector contains more complex feature information about the seismic wave after processing by functional layer A. Node knowledge vector data related to different features is obtained through node splitting. Then, in-depth learning is performed on some of the node knowledge vector data (the remaining node knowledge vector data), such as a more detailed analysis of the node knowledge vector data related to the energy attenuation characteristics of the seismic wave after processing by functional layer A, to generate a transition knowledge transfer learning vector.

[0102] In the last knowledge transfer unit of the multi-stage knowledge transfer unit, the remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level (transition knowledge transfer unit) is obtained, and the final stage of knowledge transfer learning is performed on these remaining node knowledge vector data to generate the key node knowledge vector data generated by the final stage of knowledge transfer learning. For example, the final stage of learning is performed on the remaining node knowledge vector data related to the change in the seismic wave phase within a specific time period in the transition knowledge transfer learning vector to obtain key information about this phase change at this stage, forming the key node knowledge vector data generated by the final stage of knowledge transfer learning.

[0103] Then, the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning in each stage (first stage, transition, and last stage) are node-fused to generate a multi-stage knowledge transfer learning vector corresponding to functional layer B. This vector is used to generate the reinforcement learning knowledge vector data corresponding to functional layer B.

[0104] The same process will also be carried out in functional layer C (assuming it is the last model functional layer), and finally the reinforcement learning knowledge vector data corresponding to functional layer C will be generated.

[0105] Based on the reinforcement learning knowledge vector data corresponding to the last model functional layer (functional layer C), the server generates template reinforcement waveform data corresponding to the waveform data of the second template period. This process is similar to the process of generating reinforcement waveform data when processing normal period waveform data. It includes combining the basic knowledge vector data of the waveform data of the second template period with the reinforcement learning knowledge vector data corresponding to functional layer C to generate target reinforcement learning knowledge vector data, then performing node compression on the target reinforcement learning knowledge vector data through a node compression unit to generate target knowledge vector data, and finally performing inverse regularization conversion on the target knowledge vector data through an inverse regularization conversion unit to obtain template reinforcement waveform data corresponding to the waveform data of the second template period.

[0106] Next, the server generates a target training cost based on the error between the template-enhanced waveform data and the waveform data from the first template period. Since the waveform data from the first template period has higher data value and is more accurate, while the template-enhanced waveform data is the result of the seismic wave enhancement analysis network processing the relatively low-value waveform data from the second template period, the difference between the two can reflect the current processing effect of the seismic wave enhancement analysis network. Through specific error calculation methods, such as calculating the mean square error between the two in terms of amplitude, frequency, phase, etc., a quantized error value is obtained, which is the target training cost.

[0107] Finally, the server optimizes the network parameters of the seismic wave enhancement analysis network based on this target training cost. According to the size and direction of the target training cost (whether it is too large or too small, and on which features the error is large), the parameters of each model function layer in the seismic wave enhancement analysis network are adjusted. For example, if it is found that the amplitude error between the template enhancement waveform data and the waveform data of the first template period is large when calculating the target training cost, then the parameters in the model function layer related to amplitude analysis (such as the parameter part involving amplitude analysis in function layer A, function layer B, and function layer C) will be adjusted. This may be to adjust the weights in knowledge transfer learning, the coefficients in the node nonlinear conversion, etc., so that the network can reduce this error when processing similar data next time, thereby improving the analysis and processing capabilities of seismic wave data. After multiple such training processes, the seismic wave enhancement analysis network can continuously optimize its own parameters to process seismic wave data more accurately.

[0108] Figure 2 The hardware structure of the multi-source monitoring data analysis system 100 based on active seismic sources for implementing the multi-source monitoring data analysis method based on active seismic sources provided by an embodiment of the present invention is shown as follows: Figure 2As shown, the multi-source monitoring data analysis system 100 based on seismic active sources may include a processor 110 , a machine-readable storage medium 120 , a bus 130 , and a communication unit 140 .

[0109] The machine-readable storage medium 120 may store data and / or instructions. In some embodiments, the machine-readable storage medium 120 may store data acquired from an external terminal. In some embodiments, the machine-readable storage medium 120 may store data and / or instructions used by the seismic active source-based multi-source monitoring data analysis system 100 to execute or use to implement the exemplary methods described herein.

[0110] During the specific implementation process, one or more processors 110 execute computer executable instructions stored in the machine-readable storage medium 120, so that the processor 110 can execute the multi-source monitoring data analysis method based on seismic active source as described in the above method embodiment. The processor 110, the machine-readable storage medium 120 and the communication unit 140 are connected through the bus 130, and the processor 110 can be used to control the sending and receiving actions of the communication unit 140.

[0111] The specific implementation process of the processor 110 can refer to the various method embodiments executed by the multi-source monitoring data analysis system 100 based on the above-mentioned active earthquake source. The implementation principles and technical effects are similar and will not be repeated here in this embodiment.

[0112] In addition, an embodiment of the present invention further provides a readable storage medium, in which computer-executable instructions are preset. When a processor executes the computer-executable instructions, the above-mentioned multi-source monitoring data analysis method based on active seismic sources is implemented.

[0113] It should be noted that in order to simplify the description of the present invention and thus help understand one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, multiple features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A multi-source monitoring data analysis method based on active seismic sources, characterized in that: The method comprises: Acquire waveform data of multiple time periods of candidate seismic wave sequences of multi-source monitoring data of an active seismic source, and for each time period of waveform data, load the waveform data into a pre-trained seismic wave enhancement analysis network, wherein the seismic wave enhancement analysis network includes multiple model function layers, each model function layer is cascaded, and each model function layer includes a multi-stage knowledge transfer unit, and the multi-stage knowledge transfer unit includes more than two stages of knowledge transfer learning; In the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loading data of the target model functional layer, and the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of each stage is node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the target model functional layer, and the multi-stage knowledge transfer learning vector is used to generate the reinforcement learning knowledge vector data corresponding to the target model functional layer; wherein the first stage knowledge transfer learning is used to learn the loading data of the target model functional layer, and the knowledge transfer learning of each remaining stage is used to learn the remaining node knowledge vector data except the key node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous stage; wherein, when the target model functional layer is the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the basic knowledge vector data of the time period waveform data; when the target model functional layer is the model functional layer after the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer; Based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, the reinforcement waveform data corresponding to the waveform data of the time period is generated; based on the reinforcement waveform data corresponding to each of the waveform data of the time period, the reinforcement learning seismic wave sequence corresponding to the candidate seismic wave sequence is generated.

2. The multi-source monitoring data analysis method based on active seismic sources according to claim 1 is characterized in that: The seismic wave enhancement analysis network further includes a regularization conversion unit and a node expansion unit. The step of determining the basic knowledge vector data of the time period waveform data includes: Performing regularization conversion on the period waveform data by the regularization conversion unit to generate regularized waveform data; The node expansion unit performs node expansion on the regularized waveform data to generate basic knowledge vector data of the time period waveform data.

3. The multi-source monitoring data analysis method based on active seismic sources according to claim 1 is characterized in that: When the target model function layer is a model function layer subsequent to the first model function layer among the multiple model function layers, performing multi-stage knowledge transfer learning on the loaded data of the target model function layer, performing node fusion on key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning in each stage, and generating a multi-stage knowledge transfer learning vector corresponding to the target model function layer, including: In the first-order knowledge transfer unit of the multi-stage knowledge transfer unit, first-order knowledge transfer learning is performed on the reinforcement learning knowledge vector data corresponding to the previous model function layer of the target model function layer to generate a first-order knowledge transfer learning vector; In the transition knowledge transfer unit of the multi-stage knowledge transfer unit, obtaining remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, performing knowledge transfer learning on the remaining node knowledge vector data, and generating a transition knowledge transfer learning vector; In the last knowledge transfer unit of the multi-stage knowledge transfer unit, obtaining remaining node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, performing the last stage of knowledge transfer learning on the remaining node knowledge vector data, and generating key node knowledge vector data generated by the last stage of knowledge transfer learning; The key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning in each stage are node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the functional layer of the target model.

4. The multi-source monitoring data analysis method based on active seismic sources according to claim 3 is characterized in that: The method further comprises: Perform node splitting on the knowledge transfer learning vector generated by the previous level of knowledge transfer learning to generate node knowledge vector data for each feature node; According to the arrangement order of each feature node, starting from the first feature node, the node knowledge vector data of the target number of nodes with a later node order are used as the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level, and the node knowledge vector data of the remaining nodes are used as the remaining node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of the previous level.

5. The multi-source monitoring data analysis method based on active seismic sources according to claim 1 is characterized in that: Each model function layer further includes a node nonlinear conversion unit based on a feature self-focusing strategy, and the method further includes: Loading the multi-stage knowledge transfer learning vector generated by the multi-stage knowledge transfer unit in the target model functional layer into the node nonlinear conversion unit in the target model functional layer, performing node nonlinear conversion processing on the multi-stage knowledge transfer learning vector according to the feature self-focusing strategy, and generating a multi-stage knowledge transfer learning vector for node nonlinear conversion; The step of generating the reinforcement learning knowledge vector data corresponding to the target model functional layer includes: In the target model functional layer, reinforcement learning knowledge vector data corresponding to the target model functional layer is generated based on the multi-stage knowledge transfer learning vector corresponding to the target model functional layer, the multi-stage knowledge transfer learning vector of the node nonlinear conversion, and the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer.

6. The multi-source monitoring data analysis method based on active seismic sources according to claim 5 is characterized in that: The multi-stage knowledge transfer learning vector corresponding to the target model functional layer includes a node knowledge transfer learning vector of each feature node, and performing node nonlinear conversion processing on the multi-stage knowledge transfer learning vector according to the feature self-focusing strategy to generate a node nonlinear conversion multi-stage knowledge transfer learning vector includes: Performing a focusing process of a feature self-focusing strategy on the multi-stage knowledge transfer learning vector to determine the attention weight corresponding to each node in the multi-stage knowledge transfer learning vector; The attention weights and node knowledge transfer learning vectors corresponding to each feature node are respectively multiplied to generate a multi-stage knowledge transfer learning vector for node nonlinear transformation.

7. The multi-source monitoring data analysis method based on active seismic sources according to claim 5 is characterized in that: The step of generating the reinforcement learning knowledge vector data corresponding to the target model functional layer based on the multi-stage knowledge transfer learning vector corresponding to the target model functional layer, the multi-stage knowledge transfer learning vector of the node nonlinear conversion, and the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer includes: Combining the multi-stage knowledge transfer learning vector corresponding to the target model functional layer and the multi-stage knowledge transfer learning vector of the node nonlinear conversion to generate combined knowledge vector data; Perform node conversion on the combined knowledge vector data to generate converted knowledge vector data, and combine the converted knowledge vector data with the reinforcement learning knowledge vector data corresponding to the previous model function layer of the target model function layer to generate reinforcement learning knowledge vector data corresponding to the target model function layer.

8. The multi-source monitoring data analysis method based on active seismic sources according to claim 1 is characterized in that: The step of generating the enhanced waveform data corresponding to the period waveform data based on the reinforcement learning knowledge vector data corresponding to the last model function layer includes: combining the basic knowledge vector data of the waveform data for the time period and the reinforcement learning knowledge vector data corresponding to the last model function layer to generate target reinforcement learning knowledge vector data, and determining the reinforcement waveform data corresponding to the waveform data for the time period based on the target reinforcement learning knowledge vector data; The seismic wave enhancement analysis network further includes an inverse regularization conversion unit and a node compression unit. The determination of the enhanced waveform data corresponding to the waveform data of the time period based on the target reinforcement learning knowledge vector data includes: Performing node compression on the target reinforcement learning knowledge vector data by the node compression unit to generate target knowledge vector data, wherein the number of nodes in the target knowledge vector data is the same as the number of nodes in the period waveform data; The target knowledge vector data is subjected to an inverse regularization conversion by the inverse regularization conversion unit to generate enhanced waveform data corresponding to the period waveform data.

9. The multi-source monitoring data analysis method based on active seismic sources according to claim 1 is characterized in that: The method comprises: Acquire sample learning data, each of the sample learning data includes waveform data of a first template period and waveform data of a second template period, wherein the data value of the waveform data of the first template period is greater than the data value of the waveform data of the second template period; Loading the waveform data of the second template period into an initialized seismic wave enhancement analysis network, wherein the seismic wave enhancement analysis network includes a plurality of model function layers, each of which is cascaded, and each of which includes a multi-stage knowledge transfer unit, and the multi-stage knowledge transfer unit includes two or more stages of knowledge transfer learning; In the multi-stage knowledge transfer unit of the target model functional layer, multi-stage knowledge transfer learning is performed on the loading data of the target model functional layer, and the key node knowledge vector data in the knowledge transfer learning vector generated by the knowledge transfer learning of each stage is node-fused to generate a multi-stage knowledge transfer learning vector corresponding to the target model functional layer, and the multi-stage knowledge transfer learning vector is used to generate the reinforcement learning knowledge vector data corresponding to the target model functional layer; wherein the first stage knowledge transfer learning is used to learn the loading data of the target model functional layer, and the knowledge transfer learning of each remaining stage is used to learn the remaining node knowledge vector data except the key node knowledge vector data generated by node splitting the knowledge transfer learning vector generated by the knowledge transfer learning of the previous stage; wherein, when the target model functional layer is the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the basic knowledge vector data of the waveform data of the second template period; when the target model functional layer is the model functional layer after the first model functional layer among the multiple model functional layers, the loading data of the target model functional layer is the reinforcement learning knowledge vector data corresponding to the previous model functional layer of the target model functional layer; Based on the reinforcement learning knowledge vector data corresponding to the last model functional layer, generating template reinforcement waveform data corresponding to the waveform data of the second template period, generating a target training cost based on the error between the template reinforcement waveform data and the waveform data of the first template period, and optimizing the network parameters of the seismic wave reinforcement analysis network based on the target training cost; The step of obtaining sample learning data includes: Acquire a first template seismic wave sequence, wherein the first template seismic wave sequence includes a plurality of first template time period waveform data; performing noise disturbance processing on the first template seismic wave sequence to generate a noise-perturbed seismic wave sequence; Performing feature compression on the noise-perturbed seismic wave sequence to generate a second template seismic wave sequence; Based on obtaining first template period waveform data and second template period waveform data in the same period identifier from the first template seismic wave sequence and the second template seismic wave sequence, sample learning data is generated.

10. A multi-source monitoring data analysis system based on active seismic sources, characterized in that: The multi-source monitoring data analysis system based on active seismic sources includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the multi-source monitoring data analysis method based on active seismic sources described in any one of claims 1 to 9.

Images