Karst cave environment data analysis method and system based on artificial intelligence

Abstract

The invention provides a karst cave environment data analysis method and system based on artificial intelligence, and the method achieves the high-precision prediction of a karst cave environment evolution mode and risk knowledge points through obtaining the karst cave environment monitoring data of a first template and a second template, and carrying out the multi-round parameter learning. By constructing the cave environment evolution mode prediction network and the cave environment risk prediction network, the evolution mode characteristics of the cave environment and the corresponding risk knowledge points can be accurately estimated and predicted. In the network parameter learning process, through continuous iterative optimization, the prediction network gradually approaches the real situation until the network convergence requirement is met, so that the cave environment risk prediction network and the cave environment evolution mode prediction network with high-precision prediction capability are generated, the intelligent level of karst cave environment monitoring is improved, and the prediction efficiency is improved. And accurate prediction of the environmental evolution mode of the karst cave and timely discovery and early warning of potential environmental risk points are realized.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a karst cave environment data analysis method and system based on artificial intelligence. Background Art

[0002] Karst caves, as unique natural landforms on Earth, have complex and changeable internal environments, involving interactions among geology, climate, hydrology, ecology, and other aspects. For a long time, the monitoring and analysis of karst cave environments have mainly relied on manual field surveys and traditional instrumental measurement methods. However, these methods have significant limitations, such as long monitoring cycles, difficulty in data acquisition, poor real-time performance, and the inability to fully reflect the dynamic changes in the cave environment.

[0003] In order to gain a deeper understanding of the evolution of karst cave environments, relevant technologies are constantly exploring new monitoring and analysis technologies. Traditional environmental monitoring methods often focus on the measurement of a single or a few environmental parameters, which makes it difficult to fully reveal the multidimensional characteristics of the cave environment and their interactions. In addition, the evolution of the cave environment is often accompanied by a series of complex risk factors, such as geological disasters and ecological imbalance. The timely identification and early warning of these risk factors are crucial for the protection of cave resources and the safety of personnel.

[0004] Therefore, there is an urgent need for a method that can efficiently and accurately analyze karst cave environmental data in order to achieve a comprehensive understanding of cave environmental evolution patterns and timely early warning of risk factors. 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 karst cave environment data analysis method based on artificial intelligence, the method comprising:

[0006] Acquire first template karst cave environment monitoring data and second template karst cave environment monitoring data, wherein the first template karst cave environment monitoring data carries marked cave environment evolution pattern characteristics and marked cave environment risk knowledge points corresponding to the marked cave environment evolution pattern characteristics, and the second template karst cave environment monitoring data carries marked cave environment risk knowledge points corresponding to the marked cave environment evolution pattern characteristics, and the first template karst cave environment monitoring data and the second template karst cave environment monitoring data both contain monitoring records arranged in chronological order, and each monitoring record contains multiple environmental parameter values, and the environmental parameter values ​​include one or more combinations of air temperature parameter values, air humidity parameter values, carbon dioxide concentration parameter values, air pressure parameter values, cave internal structure change parameter values, oxygen concentration parameter values, wind speed parameter values, light intensity parameter values, methane concentration parameter values, hydrogen sulfide concentration parameter values, cave water flow velocity parameter values, cave water pH parameter values, soil moisture parameter values, rock temperature parameter values, radon gas concentration parameter values, microbial community quantity parameter values, cave wall roughness parameter values, cave top dripping velocity parameter values, negative ion concentration parameter values, and cave dust concentration parameter values;

[0007] In the a+1th round of network parameter learning task, the cave environment evolution pattern prediction network generated in the ath round of network parameter learning task is used to predict the second template karst cave environment monitoring data to generate a first cave environment evolution pattern feature, wherein the cave environment evolution pattern prediction network is used to estimate the cave environment evolution pattern feature in the karst cave environment monitoring data, and a is a positive integer;

[0008] Using the first template karst cave environment monitoring data, parameter learning is performed on the cave environment risk prediction network generated in the a-th round of network parameter learning tasks to generate a cave environment risk prediction network in the a+1-th round of network parameter learning tasks, wherein the cave environment risk prediction network is used to predict cave environment risk knowledge points of cave environment evolution pattern characteristics based on the karst cave environment monitoring data;

[0009] Using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task, the first template karst cave environment monitoring data and the second template karst cave environment monitoring data are predicted to generate second cave environment evolution pattern features corresponding to the first template karst cave environment monitoring data and third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data;

[0010] According to the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, as well as the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, until the network convergence requirements are met to generate a cave environment risk prediction network and a cave environment evolution pattern prediction network that have completed parameter learning.

[0011] On the other hand, an embodiment of the present invention also provides a karst cave environment data analysis system based on artificial intelligence, 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.

[0012] Based on the above aspects, the embodiment of the present application obtains the karst cave environment monitoring data of the first template and the second template containing rich environmental parameter values, and uses these data to perform multiple rounds of parameter learning on the network, thereby achieving high-precision prediction of the karst cave environment evolution pattern and risk knowledge points. This method can accurately estimate and predict the evolution pattern characteristics of the cave environment and its corresponding risk knowledge points by constructing a cave environment evolution pattern prediction network and a cave environment risk prediction network. In the process of network parameter learning, through continuous iterative optimization, the prediction network gradually approaches the actual situation until the network convergence requirements are met, thereby generating a cave environment risk prediction network and a cave environment evolution pattern prediction network with high-precision prediction capabilities, which not only improves the intelligent level of karst cave environment monitoring, but also realizes the accurate prediction of the karst cave environment evolution pattern and the timely discovery and warning of potential environmental risk points, providing data support for the protection of cave resources and disaster warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the execution flow of the karst cave environment data analysis method based on artificial intelligence provided in an embodiment of the present invention.

[0014] Figure 2 It is a schematic diagram of the hardware architecture of the karst cave environment data analysis system based on artificial intelligence provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention will be described in detail below with reference to the accompanying drawings. Figure 1 It is a flow chart of a karst cave environment data analysis method based on artificial intelligence provided by an embodiment of the present invention. The karst cave environment data analysis method based on artificial intelligence is introduced in detail below.

[0016] Step S110, obtaining the first template karst cave environment monitoring data and the second template karst cave environment monitoring data, wherein the first template karst cave environment monitoring data carries the annotated cave environment evolution pattern characteristics and the annotated cave environment risk knowledge points corresponding to the annotated cave environment evolution pattern characteristics, and the second template karst cave environment monitoring data carries the annotated cave environment risk knowledge points corresponding to the annotated cave environment evolution pattern characteristics, and the first template karst cave environment monitoring data and the second template karst cave environment monitoring data both contain monitoring records arranged in chronological order, and each monitoring record contains multiple environment risk knowledge points. The environmental parameter values ​​include one or more combinations of air temperature parameter values, air humidity parameter values, carbon dioxide concentration parameter values, air pressure parameter values, cave internal structure change parameter values, oxygen concentration parameter values, wind speed parameter values, light intensity parameter values, methane concentration parameter values, hydrogen sulfide concentration parameter values, cave water flow velocity parameter values, cave water pH parameter values, soil moisture parameter values, rock temperature parameter values, radon gas concentration parameter values, microbial community quantity parameter values, cave wall roughness parameter values, cave top dripping velocity parameter values, negative ion concentration parameter values, and cave dust concentration parameter values.

[0017] In this embodiment, a karst cave area is considered, which is named "karst cave A". In this cave, in order to obtain the first template karst cave environment monitoring data and the second template karst cave environment monitoring data, multiple monitoring stations are pre-set in the cave.

[0018] For the first template karst cave environmental monitoring data, it is assumed that monitoring stations are set up near the entrance, in the middle and deep of cave A. Each monitoring station is equipped with advanced environmental monitoring instruments that can accurately measure multiple environmental parameter values. For example, at the station near the entrance, monitoring and data are recorded at fixed time intervals (such as once an hour) from a certain moment. The air temperature parameter value fluctuates within a day. The temperature is low in the morning, gradually rises as the sun rises, reaches a high value at noon, and begins to drop in the afternoon; the air humidity parameter value is affected by the climate outside the cave and the water vapor exchange inside the cave, and will increase significantly after rainy weather. The carbon dioxide concentration parameter value will change due to factors such as biological respiration and rock chemistry in the cave. For example, when the microbial community in the cave is active, the carbon dioxide concentration may increase. The air pressure parameter value will change with the exchange of airflow inside and outside the cave. When there is strong wind outside the cave, the air pressure in the cave will be affected accordingly. The parameter value of the internal structure change of the cave may be obtained through regular laser scanning or high-precision measurement equipment, such as the discovery of tiny crack expansion in the rocks in a certain area. The oxygen concentration parameter value is related to the carbon dioxide concentration value, because biological respiration consumes oxygen and produces carbon dioxide. The wind speed parameter value varies in different areas of the cave, and the wind speed may be higher in places where the cave passage is narrow. The light intensity parameter value is relatively high in areas close to the cave entrance, and very low deep inside the cave. The methane concentration parameter value may vary due to reasons such as the decomposition of organic matter in the cave. The hydrogen sulfide concentration parameter value is also related to chemical processes and microbial activity in the cave. The cave water flow rate parameter value depends on the underground river or seepage in the cave. For example, the water flow rate may increase during the rainy season. The cave water pH parameter value is affected by factors such as the rock composition in the cave and the mineral content in the water flow. The soil moisture parameter value is also monitored in some areas of the cave covered with soil, which will affect the growth of plants (if there are a small number of plants adapted to the cave environment). The rock temperature parameter value is related to the heat conduction and air temperature in the cave. The radon gas concentration parameter value may vary due to the radioactive properties of the rock. The microbial community quantity parameter value is obtained by laboratory analysis of samples collected from different areas in the cave. The cave wall roughness parameter value is obtained through a special roughness measuring instrument. The dripping rate parameter value on the top of the cave is related to the rock seepage at the top of the cave. The negative ion concentration parameter value is affected by the air ionization in the cave. The dust concentration parameter value in the cave is related to factors such as air flow and rock weathering in the cave.

[0019] Each monitoring record contains the above-mentioned multiple environmental parameter values. In addition, these monitoring records are arranged in chronological order, thus forming the first template karst cave environment monitoring data. At the same time, based on years of research experience and in-depth understanding of the cave, the researchers have annotated these monitoring data, annotated the cave environment evolution pattern characteristics and the corresponding annotated cave environment risk knowledge points. For example, when the carbon dioxide concentration in the cave continues to rise, the air humidity increases and the rock temperature drops, the annotated cave environment evolution pattern characteristics are "the ecological balance inside the cave begins to develop towards a biological community that is conducive to moisture-loving and resistant to high carbon dioxide environments, and rock weathering may accelerate", and the corresponding annotated cave environment risk knowledge points are "some fragile stalagmites and stalactites in the cave may be subject to increased risk of erosion by acidic substances (carbonic acid formed by high concentrations of carbon dioxide dissolved in water), and biological communities adapted to low carbon dioxide environments may face survival pressure".

[0020] For the second template karst cave environment monitoring data, it is also obtained in karst cave A, but it may be from different monitoring sites or using some supplementary monitoring methods. These data also contain monitoring records arranged in chronological order. Each record also contains multiple environmental parameter values ​​and carries the same annotated cave environment evolution pattern characteristics as the first template karst cave environment monitoring data. The annotated cave environment risk knowledge points corresponding to the annotated cave environment evolution pattern characteristics. For example, at another monitoring site close to the underground river in the cave, the changes in the monitored water flow velocity parameter values ​​and other related environmental parameter values ​​are recorded. Although these data may differ from the first template karst cave environment monitoring data in terms of the specific values ​​and change trends of some parameters, they are all related to the same cave environment evolution pattern characteristics and related environmental risk knowledge points.

[0021] Step S120, in the a+1th round of network parameter learning task, the cave environment evolution pattern prediction network generated in the ath round of network parameter learning task is used to predict the second template karst cave environment monitoring data to generate a first cave environment evolution pattern feature, and the cave environment evolution pattern prediction network is used to estimate the cave environment evolution pattern features in the karst cave environment monitoring data, and a is a positive integer.

[0022] Assuming that the value of a is 1, then in the second round of network parameter learning task, the cave environment evolution pattern prediction network has been generated in the first round of network parameter learning task. This network is built based on a large amount of previous data and algorithms to estimate the cave environment evolution pattern characteristics in the karst cave environment monitoring data.

[0023] The second template karst cave environment monitoring data of karst cave A is input into the cave environment evolution pattern prediction network. Taking the data of the monitoring station near the underground river in the cave as an example, the network will analyze the input monitoring records containing multiple environmental parameter values ​​such as air temperature, humidity, and water flow rate according to its internal complex algorithms and models.

[0024] Assuming that the algorithm model of the network contains a multi-layer neural network structure, in the first layer, it may extract features from the input environmental parameter values. For example, for the water flow velocity parameter value, it will analyze its change amplitude and change frequency in different time periods; for the air temperature parameter value, it will analyze its correlation characteristics with the water flow velocity and other parameters. Then, these initially extracted features will be passed to the next layer of the network structure for further analysis and integration.

[0025] After being processed by the multi-layer network, the first cave environment evolution pattern feature is finally generated. This feature may be described as "In the area near the underground river, due to the seasonal fluctuations in water flow velocity and the coordinated changes in air temperature and humidity, the cave environment presents a periodic humidity regulation pattern, and this pattern affects the distribution and activity of the surrounding microbial communities, causing the microbial communities to evolve towards species that adapt to higher humidity and a specific temperature range." This first cave environment evolution pattern feature is an estimate of the cave environment evolution pattern obtained by the network after complex calculations and analysis based on the input second template karst cave environment monitoring data.

[0026] Step S130, using the first template karst cave environment monitoring data to perform parameter learning on the cave environment risk prediction network generated in the a-th round of network parameter learning task, to generate a cave environment risk prediction network in the a+1-th round of network parameter learning task, wherein the cave environment risk prediction network is used to predict the cave environment risk knowledge points of the cave environment evolution pattern characteristics based on the karst cave environment monitoring data.

[0027] Continuing with karst cave A as an example, the first template karst cave environment monitoring data is used to perform parameter learning on the cave environment risk prediction network generated in the first round of network parameter learning tasks to generate the cave environment risk prediction network in the second round of network parameter learning tasks.

[0028] Assume that in the first template karst cave environment monitoring data, the data from the monitoring station in the middle of Cave A shows that the carbon dioxide concentration value continues to rise in a certain period of time, while the oxygen concentration value slightly decreases, and the parameter value of the internal structure change of the cave shows that some rocks have slight deformations. According to the annotated cave environment evolution pattern characteristics "the increase in carbon dioxide concentration leads to changes in the chemical environment in the cave, which may affect the biological community and rock stability" and the corresponding annotated cave environment risk knowledge point "biodiversity may decrease, and some cave structures are at risk of collapse."

[0029] The cave environment risk prediction network first extracts the local state change characteristics of the first local karst cave space (here is the monitoring area in the middle of cave A). For example, for the state change of rising carbon dioxide concentration, the network will analyze its rising rate, duration, and degree of correlation with other environmental parameter values ​​(such as air humidity, temperature, etc.), and generate the first local state change characteristics corresponding to the first local karst cave space. Then, this local space is spatially encoded, for example, based on its geographical location in the cave, spatial range and other factors, the first environmental block spatial data corresponding to the first local karst cave space is generated.

[0030] The first local state change characteristics and the first environmental block spatial data are loaded into the cave environment risk prediction network generated in the first round of network parameter learning tasks. The network processes these data according to its internal algorithm model, such as a risk assessment model based on probability statistics or a classification model in a machine learning algorithm. After calculation, the first estimated cave environment evolution pattern and estimated cave environment risk knowledge points corresponding to the first template karst cave environment monitoring data are obtained. Assume that the first estimated cave environment evolution pattern is "the increase in carbon dioxide concentration will cause changes in the structure of some biological communities in the cave, and the stability of the cave structure will decrease", and the estimated cave environment risk knowledge points are "some organisms in the cave that rely on a specific gas environment may become extinct, and the possibility of local collapse of the cave roof increases."

[0031] Then, the parameters of the cave environment risk prediction network generated in the first round of network parameter learning task are learned based on the third characteristic loss between the first estimated cave environment evolution pattern and the first candidate cave environment evolution pattern (i.e., the previously labeled "increased carbon dioxide concentration leads to changes in the chemical environment in the cave, which may affect the biological community and rock stability"), as well as the labeled cave environment risk knowledge points and the estimated cave environment risk knowledge points. The third characteristic loss here can be determined by calculating the degree of difference between the two in describing various aspects of the cave environment evolution pattern (such as biological community changes, rock stability changes, etc.). For example, if the description of biological community changes in the first estimated cave environment evolution pattern differs from the first candidate cave environment evolution pattern in terms of the degree of impact on biological species, then this difference will be quantified as part of the third characteristic loss. Similarly, for the labeled cave environment risk knowledge points and the estimated cave environment risk knowledge points, the differences in the specific types of risks (such as biological extinction risk, cave collapse risk) and risk levels (such as high, medium, and low) will also be compared. In this way, the parameters of the cave environment risk prediction network are adjusted to generate the cave environment risk prediction network in the second round of network parameter learning task.

[0032] Step S140, using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task, predict the first template karst cave environment monitoring data and the second template karst cave environment monitoring data, and generate the second cave environment evolution pattern characteristics corresponding to the first template karst cave environment monitoring data and the third cave environment evolution pattern characteristics corresponding to the second template karst cave environment monitoring data.

[0033] In this embodiment, the basic cave environment evolution pattern prediction network in the second round of network parameter learning tasks and the cave environment risk prediction network generated in the second round of network parameter learning tasks are used to predict the first template karst cave environment monitoring data and the second template karst cave environment monitoring data of karst cave A.

[0034] For the first template karst cave environment monitoring data, the basic cave environment evolution pattern prediction network in the second round of network parameter learning tasks is first used for prediction. This basic cave environment evolution pattern prediction network includes an environmental state change encoding subnetwork and a fully connected mapping subnetwork. Taking the monitoring data near the entrance of cave A as an example, the environmental state change encoding subnetwork will extract features for each monitoring record. For example, for the monitoring records at a certain moment, including parameter values ​​such as air temperature, humidity, and wind speed, it will extract the initial feature vector sequence that reflects the change in the cave environment state. This initial feature vector sequence may contain information such as the relative change amplitude of each parameter value and the proportional relationship with other parameters.

[0035] Then, based on this initial feature vector sequence, a clustering algorithm is used to cluster the matching environmental states into the same category. For example, the states with low air temperature, high humidity and low wind speed are clustered into one category, and each category represents an environmental state node. According to the time sequence and the similarity between states, the intra-node transfer relationship within the environmental state node and the inter-node transfer relationship between the environmental state nodes are constructed to generate a state transition graph. This state transition graph describes the dynamic change process of the environmental state near the cave entrance over time, such as the transition process from the cold and dry state in winter to the warm and humid state in spring.

[0036] Traverse this state transition diagram, starting from the initial state, and assign a unique code to the environmental state at each time point according to the time sequence and state transition relationship to generate a state sequence code. Arrange this state sequence code in chronological order to form an environmental state change trajectory. Use smoothing filtering to remove noise and jitter in the environmental state change trajectory, such as eliminating sudden fluctuations in temperature values ​​caused by small errors in monitoring instruments. At the same time, use time series analysis to extract periodic and trend characteristics in the environmental state change trajectory, such as finding that air temperature has seasonal periodic changes every year and has a tendency to rise slowly year by year. Use pattern recognition to identify target change patterns or abnormal points in the environmental state change trajectory, such as identifying the abnormal increase in wind speed caused by large-scale engineering construction outside the cave in a certain year. After obtaining the enhanced environmental state change trajectory features, load them into the fully connected mapping subnetwork, and finally generate the second cave environment evolution pattern features corresponding to the first template karst cave environment monitoring data. The characteristics of this second cave environment evolution pattern may be described as "near the cave entrance, over time, due to the combined effects of external environmental and internal cave factors, the cave environment exhibits seasonal temperature and humidity fluctuations, and this fluctuation has a periodic impact on the biological community at the cave entrance. At the same time, in the long run, due to the increase in external interference factors, the environmental stability of the cave entrance tends to decline."

[0037] For the second template karst cave environment monitoring data, the basic cave environment evolution pattern prediction network in the second round of network parameter learning tasks and the cave environment risk prediction network generated in the second round of network parameter learning tasks are used for prediction. First, the environmental state change encoding subnetwork encodes the environmental state change of the second template karst cave environment monitoring data, and takes the monitoring data near the underground river in the cave as an example to generate the environmental state change trajectory features corresponding to the second template karst cave environment monitoring data. Then, the cave environment risk prediction network generated in the second round of network parameter learning tasks extracts the local state change features of the environmental state change trajectory features corresponding to the second template karst cave environment monitoring data, and generates the second local state change features corresponding to the annotated cave environment evolution pattern features. For example, for the water flow velocity change part in the environmental state change trajectory features near the underground river, its local state change features related to the changes in the surrounding biological communities are extracted.

[0038] The trajectory characteristics of environmental state changes and the local state change characteristics are integrated to generate enhanced trajectory characteristics. For example, the trajectory characteristics related to the underground river flow velocity and the local state change characteristics of the biological community changes are integrated together to obtain enhanced trajectory characteristics containing more information. Finally, this enhanced trajectory feature is loaded into the fully connected mapping subnetwork to generate the third cave environment evolution pattern characteristics corresponding to the second template karst cave environment monitoring data. This third cave environment evolution pattern characteristic may be described as "In the area near the underground river, the change in water flow velocity is closely related to the surrounding biological community and the chemical environment inside the cave. The seasonal fluctuation of water flow velocity leads to the periodic succession of the biological community, and this succession affects the circulation of chemical substances in the cave. At the same time, the cave structure tends to change locally under the long-term erosion of water flow."

[0039] Step S150, based on the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, until the network convergence requirements are met to generate a cave environment risk prediction network and a cave environment evolution pattern prediction network that have completed parameter learning.

[0040] In this embodiment, according to the first feature loss between the marked cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate a cave environment evolution pattern prediction network in the second round of network parameter learning tasks, until the network convergence requirements are met to generate a cave environment risk prediction network and a cave environment evolution pattern prediction network that have completed parameter learning.

[0041] Taking the various cave environment evolution pattern characteristics obtained in karst cave A as an example, the annotated cave environment evolution pattern characteristics of the first template karst cave environment monitoring data include the first candidate cave environment evolution pattern and the first local karst cave space. Assume that the first candidate cave environment evolution pattern is characterized as "due to the combined effect of multiple environmental factors in the cave, the biological community and cave structure are in a dynamic equilibrium state, but are easily unbalanced when subjected to external interference", and the first local karst cave space is the middle area of ​​cave A. The second cave environment evolution pattern characteristics include the second estimated cave environment evolution pattern and the second estimated local karst cave space. Assume that the second estimated cave environment evolution pattern is "in the middle area of ​​the cave, due to local environmental changes, the stability of the biological community is affected to a certain extent, but the changes in the cave structure are not obvious", and the second estimated local karst cave space is also the middle area of ​​cave A.

[0042] Calculate the first sub-feature loss between the first candidate cave environment evolution pattern and the second estimated cave environment evolution pattern. For example, in terms of biological community stability, the annotation is that it is easy to be unbalanced, while the estimation is that it is affected to a certain extent. It is necessary to quantify this difference in description. For cave structure, the annotation is that no obvious changes are mentioned, while the estimation is that the changes are not obvious. This difference should also be quantified to generate the first sub-feature loss. At the same time, calculate the second sub-feature loss between the first local karst cave space and the second estimated local karst cave space. This may involve the quantification of differences in spatial range, environmental characteristics within the space, etc., and finally generate the first feature loss.

[0043] Similarly, for the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, the first cave environment evolution pattern feature includes the first estimated cave environment evolution pattern and the first estimated local karst cave space, and the third cave environment evolution pattern feature includes the third estimated cave environment evolution pattern and the third estimated local karst cave space. Assume that the first estimated cave environment evolution pattern is "environmental changes near the underground river have a certain impact on the overall ecology of the cave", and the first estimated local karst cave space is the area near the underground river; the third estimated cave environment evolution pattern is "environmental changes near the underground river have a significant impact on the distribution of some biological communities in the cave", and the third estimated local karst cave space is the area near the underground river. Calculate the third sub-feature loss between the first estimated cave environment evolution pattern and the third estimated cave environment evolution pattern, for example, quantify the difference in the description of the impact on the overall ecology of the cave and the distribution of some biological communities. At the same time, calculate the fourth sub-feature loss between the first estimated local karst cave space and the third estimated local karst cave space (here, it may be because both are areas near the underground river, the loss is relatively small), so as to generate the second feature loss.

[0044] Parameter learning is performed on the basic cave environment evolution pattern prediction network based on the first characteristic loss and the second characteristic loss. For example, if the first characteristic loss indicates that there is a large difference between the prediction and annotation in terms of biome stability, the network may adjust the parameter weights related to the biome. If the second characteristic loss shows that there is a deviation in the prediction of ecological impacts in the area near the underground river, the network will adjust the parameters related to the environmental factors in the area accordingly. This process is continuously iterated until the network convergence requirements are met, that is, the first characteristic loss and the second characteristic loss are reduced to an acceptable range. At this time, the cave environment risk prediction network and the cave environment evolution pattern prediction network with completed parameter learning are generated.

[0045] After the cave environment risk prediction network and cave environment evolution pattern prediction network with completed parameter learning are generated, it is assumed that new target karst cave environment monitoring data is obtained. This data comes from a new monitoring area of ​​karst cave A or more detailed data obtained by using new monitoring technology in the original monitoring area. This target karst cave environment monitoring data is loaded into the cave environment evolution pattern prediction network and cave environment risk prediction network with completed parameter learning to obtain the target cave environment evolution pattern characteristics corresponding to the target karst cave environment monitoring data and the cave environment risk knowledge points corresponding to the target cave environment evolution pattern characteristics. For example, the target karst cave environment monitoring data shows that the hydrogen sulfide concentration in a certain area suddenly increases, and the number of microbial communities changes abnormally. After loading into the network, the target cave environment evolution pattern characteristics obtained may be "the sudden increase in hydrogen sulfide concentration in this area has caused a sharp change in the structure of the microbial community, which may cause an imbalance in the chemical environment in the cave", and the corresponding cave environment risk knowledge point is "the collapse of the microbial community may cause the material circulation in the cave to be blocked, and may produce the risk of accumulation of harmful gases".

[0046] Based on the above steps, the embodiment of the present application obtains the first template and the second template karst cave environment monitoring data containing rich environmental parameter values, and uses these data to perform multiple rounds of parameter learning on the network, thereby achieving high-precision prediction of the karst cave environment evolution pattern and risk knowledge points. This method can accurately estimate and predict the evolution pattern characteristics of the cave environment and its corresponding risk knowledge points by constructing a cave environment evolution pattern prediction network and a cave environment risk prediction network. In the process of network parameter learning, through continuous iterative optimization, the prediction network gradually approaches the actual situation until the network convergence requirements are met, thereby generating a cave environment risk prediction network and a cave environment evolution pattern prediction network with high-precision prediction capabilities, which not only improves the intelligent level of karst cave environment monitoring, but also achieves accurate prediction of karst cave environment evolution patterns and timely discovery and warning of potential environmental risk points, providing data support for the protection of cave resources and disaster warning.

[0047] In a possible implementation, step S140 includes:

[0048] Step S141, using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task to predict the first template karst cave environment monitoring data, and generate the second cave environment evolution pattern characteristics corresponding to the first template karst cave environment monitoring data.

[0049] Taking the karst cave A mentioned above as an example, in karst cave A, for example, the first template karst cave environmental monitoring data comes from monitoring stations in different areas of the cave, such as the entrance, middle, and deep. Taking the entrance site data as an example, the basic cave environment evolution pattern prediction network will process the monitoring data at the entrance according to its internal algorithm. The environmental state change encoding subnetwork in the network will first extract features from multiple environmental parameter values ​​such as air temperature parameter values, air humidity parameter values, and carbon dioxide concentration parameter values ​​contained therein for each monitoring record at the entrance. Assuming that the air temperature at the entrance is low, the humidity is high, and the carbon dioxide concentration is at a medium level at a certain moment, the environmental state change encoding subnetwork will encode these parameter values ​​according to their mutual relationship and comparison with historical data, and generate an initial feature vector sequence reflecting the change of the cave environment state at the entrance at this moment. Over time, the initial feature vector sequences of many such moments are obtained. Then, based on these initial feature vector sequences, a clustering algorithm is used to cluster similar environmental states into the same category, for example, clustering states with low temperature, high humidity, and stable carbon dioxide concentration into one category, and each category represents an environmental state node. Then, the node-to-node transfer relationship within the environmental state node and the node-to-node transfer relationship between the environmental state nodes are constructed according to the time sequence and the similarity between the states, thereby generating a state transition graph. This state transition graph describes in detail the dynamic change process of the environmental state at the cave entrance over time, and the periodic change process such as the temperature drop and humidity increase from day to night is reflected in the graph. Finally, by traversing the state transition graph, the environmental state at each time point is assigned a unique code according to the time sequence and state transition relationship to generate a state sequence code, and these state sequence codes are arranged in chronological order to form an environmental state change trajectory. After that, the fully connected mapping subnetwork will process this environmental state change trajectory, and finally generate the second cave environment evolution pattern characteristics corresponding to the first template karst cave environment monitoring data by combining its internal predefined parameters and models. This feature may be described as the periodic changes of environmental parameters such as temperature, humidity, and carbon dioxide concentration at the cave entrance and their interactions, resulting in periodic adjustments in the biological community structure at the entrance, and the cave wall has a slow chemical change process under the long-term action of humidity and carbon dioxide.

[0050] Step S142, using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task to predict the second template karst cave environment monitoring data, and generate the third cave environment evolution pattern characteristics corresponding to the second template karst cave environment monitoring data.

[0051] In a possible implementation, the basic cave environment evolution pattern prediction network includes an environment state change encoding subnetwork and a fully connected mapping subnetwork.

[0052] Step S142 includes:

[0053] Step S1421, using the environmental state change encoding subnetwork to encode the environmental state change of the second template karst cave environmental monitoring data, and generate environmental state change trajectory characteristics corresponding to the second template karst cave environmental monitoring data.

[0054] Step S1422, using the cave environment risk prediction network generated in the a+1th round of network parameter learning task, extract local state change features from the environmental state change trajectory features corresponding to the second template karst cave environment monitoring data, and generate a second local state change feature corresponding to the annotated cave environment evolution pattern feature.

[0055] Step S1423: Integrate the environmental state change trajectory features and the local state change features to generate enhanced trajectory features.

[0056] Step S1424, loading the enhanced trajectory features into the fully connected mapping subnetwork to generate the third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data.

[0057] Taking the second template karst cave environment monitoring data in karst cave A as an example, it is assumed that these data mainly come from the area near the underground river in the middle of the cave. First, the environmental state change encoding subnetwork in the basic cave environment evolution pattern prediction network encodes the environmental state change of the second template karst cave environment monitoring data. For the monitoring data in the middle area near the underground river, there are many environmental parameter values ​​in each monitoring record. The environmental state change encoding subnetwork will analyze and process these parameter values. For example, for the cave water flow velocity parameter value, it will consider the flow velocity change amplitude in different seasons, the association between flow velocity change and other environmental parameters. For the air humidity parameter value, it will analyze its response changes when the underground river water flow fluctuates. In this way, an initial feature vector sequence reflecting the change of the environmental state is generated for each monitoring moment, and then according to the previously mentioned clustering algorithm, state transition diagram construction, state sequence encoding generation and other steps, the environmental state change trajectory characteristics corresponding to the second template karst cave environment monitoring data are finally generated. This environmental state change trajectory feature can describe in detail the dynamic changes of the environmental state in the central area close to the underground river over time. For example, when the flow rate of the underground river increases greatly in the rainy season, the humidity of the surrounding air rises rapidly, and when the flow rate slows down in the dry season, the humidity decreases and the carbon dioxide concentration will increase slightly. This trajectory feature is reflected in this.

[0058] Next, the cave environment risk prediction network generated in the a+1th round of network parameter learning task is used to extract local state change features from the environmental state change trajectory features corresponding to the second template karst cave environment monitoring data. For example, for the part of the previously generated environmental state change trajectory features related to the biological community, the cave environment risk prediction network will deeply analyze the relationship between the biological community quantity parameter value, the microbial community quantity parameter value, and other environmental parameters. If the pH value of the underground river changes within a certain period of time, the cave environment risk prediction network will analyze the impact of this change on the number of microbial communities, such as whether the number of microbial communities has a decreasing trend when the pH value changes from weak alkalinity to weak acidity, and what is the rate of decrease, so as to generate the second local state change feature corresponding to the cave environment evolution pattern feature. This second local state change feature highlights the local state changes related to cave environmental risks in the environmental state change trajectory features, such as the local state change of the impact of the change in the pH value of the underground river water on the number of microbial communities in this example.

[0059] Then, the trajectory characteristics of environmental state changes and the local state change characteristics are integrated to generate enhanced trajectory characteristics. In the middle of karst cave A, near the underground river area, the previously obtained trajectory characteristics of environmental state changes, including underground river flow velocity, surrounding air humidity, carbon dioxide concentration, etc., are integrated with the local state change characteristics reflecting the influence of pH on the number of microbial communities. During the integration process, the two will be integrated according to the logical relationship and mutual influence weight between them. For example, if the underground river flow velocity has an indirect effect on the number of microbial communities, then during the integration, the trajectory characteristics related to the flow velocity and the local state change characteristics related to the number of microbial communities will be reasonably combined according to this influence relationship to generate an enhanced trajectory characteristic containing more comprehensive information. This enhanced trajectory feature not only includes the overall changes in the environmental state, but also highlights the local changes related to the cave environmental risks.

[0060] Finally, the enhanced trajectory features are loaded into the fully connected mapping subnetwork to generate the third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data. The fully connected mapping subnetwork processes the enhanced trajectory features according to its internal complex mapping relationship and predefined parameters. In the case of karst cave A, the fully connected mapping subnetwork considers various information in the enhanced trajectory features, such as the relationship between underground river flow rate, changes in the number of microbial communities, air humidity and other factors, and finally generates the third cave environment evolution pattern features. This feature may be described as the change in the flow rate of the underground river in the middle of karst cave A near the underground river area, which indirectly changes the number of microbial communities by affecting factors such as surrounding air humidity and water pH, thereby affecting the biological community structure of the entire area, and this impact may further lead to local adjustments in the circulation of chemical substances inside the cave. At the same time, the scouring effect of the underground river on the cave wall will also have a certain impact on the stability of the cave structure when the flow rate changes.

[0061] In a possible implementation, step S1421 includes:

[0062] Step S1421-1, extracting features from each monitoring record in the second template karst cave environment monitoring data, and generating an initial feature vector sequence reflecting changes in the cave environment state.

[0063] In this embodiment, it is assumed that the second template karst cave environment monitoring data is derived from a monitoring station in a specific area in the middle of cave A, which includes monitoring points at different locations such as underground rivers and surrounding cave walls and ground. Each monitoring record contains multiple environmental parameter values, such as air temperature parameter values, air humidity parameter values, carbon dioxide concentration parameter values, cave water flow velocity parameter values, microbial community number parameter values, etc. For each monitoring record, the environmental state change encoding subnetwork will perform feature extraction on these environmental parameter values. Taking the monitoring record at a certain moment as an example, the air temperature is 15 degrees Celsius, the air humidity is 80%, the carbon dioxide concentration is 0.05%, the cave water flow velocity is 0.5 meters per second, and the number of microbial communities is a certain value. The environmental state change encoding subnetwork will analyze the relationship between these parameter values ​​and the historical data of the area and the preset standard values, such as the fluctuation amplitude of air temperature relative to the historical period, the proportional relationship between air humidity and carbon dioxide concentration, and the possible association between cave water flow velocity and the number of microbial communities. Through these analyses, an initial feature vector sequence containing multiple dimensional feature information is generated, which can preliminarily reflect the characteristics of the cave environment state at that moment.

[0064] Step S1421-2, based on the initial feature vector sequence, use a clustering algorithm to cluster the matching environmental states into the same category, each category represents an environmental state node, and according to the time sequence and the similarity between the states, construct the intra-node transfer relationship within the environmental state node and the inter-node transfer relationship between the environmental state nodes to generate a state transition diagram, which describes the dynamic change process of the cave environment state over time.

[0065] For example, the environmental states of low air temperature, high air humidity, medium carbon dioxide concentration and slow cave water flow are clustered into one category, and this category becomes an environmental state node. In chronological order, when the air temperature gradually rises, the air humidity begins to drop, the carbon dioxide concentration fluctuates slightly and the cave water flow speed increases, these environmental states can be clustered into another category and become another environmental state node. When constructing the intra-node transfer relationship, for the states at different times in the same environmental state node, for example, in the first environmental state node mentioned, as time goes by, the air temperature is low but rises slightly, the air humidity is basically stable, and the carbon dioxide concentration fluctuates slightly. This change relationship of the states at different times in the same category is the intra-node transfer relationship. The conversion between different environmental state nodes, such as the conversion from a node with low temperature, high humidity and low flow rate to a node with high temperature, low humidity and high flow rate, is an inter-node transfer relationship. The state transition diagram constructed in this way describes in detail the dynamic change process of the cave environmental state over time, including the conversion order between different environmental states, the duration of each state, and the degree of association between states.

[0066] Step S1421-3, traverse the state transition diagram, starting from the initial state, assign a unique code to the environmental state at each time point according to the time sequence and state transition relationship, and generate a state sequence code, wherein the unique code is used to uniquely identify the environmental state at that time point.

[0067] The initial state may be the environmental state at a specific moment, such as the environmental state of the middle area of ​​Cave A in the morning. This state is defined as the initial state and is assigned a unique code, such as 001. As time goes by, the environmental state at the next time point is assigned the next unique code, such as 002, based on the previously constructed state transition relationship. This unique code can uniquely identify the environmental state at that time point. In this way, the environmental state at each time point has a specific identifier, which is convenient for subsequent analysis and processing.

[0068] Step S1421-4, arranging the state sequence codes in chronological order to form an environment state change trajectory.

[0069] These sequentially arranged state sequence codes can intuitively show the path of cave environment state changes over time, starting from the initial state code, and arranging the state codes of subsequent time points in sequence to form a complete trajectory of environmental state changes. This trajectory can reflect how the cave environment changes from one state to another over a period of time, such as the change from the low temperature and high humidity state in the morning to the high temperature and low humidity state at noon.

[0070] Step S1421-5, using smoothing filtering to remove noise and jitter in the environmental state change trajectory, using time series analysis to extract periodic characteristics and trend characteristics in the environmental state change trajectory, and using pattern recognition to identify target change patterns or abnormal points in the environmental state change trajectory, to obtain enhanced environmental state change trajectory characteristics.

[0071] In actual monitoring data, there may be some noise and jitter in the trajectory of environmental state changes due to the accuracy limitations of monitoring instruments or sudden interference factors in the environment. For example, monitoring instruments may occasionally have small measurement errors, resulting in unreasonable fluctuations in the air temperature value at a certain point in time. Smoothing filtering can remove these unreasonable fluctuations and make the trajectory of environmental state changes smoother and more stable. Through time series analysis, periodic characteristics in the trajectory of cave environmental state changes can be found. For example, the air temperature and humidity may show periodic changes in the morning, high at noon, and low at night in a day, or the water flow velocity in the cave has obvious periodic changes in different seasons of the year. At the same time, trend characteristics can also be extracted. For example, as the years increase, the carbon dioxide concentration in the cave tends to rise slowly. In addition, pattern recognition technology can be used to identify target change patterns or abnormal points. For example, when large-scale rock collapse or underground river diversion occurs in the cave, it will appear as an abnormal point in the trajectory of environmental state changes, or identify a special target change pattern in which the air humidity and the number of microbial communities drop sharply at the same time at several consecutive time points. Through these operations, we obtain enhanced environmental state change trajectory characteristics, which can more accurately reflect the actual changes in the cave environment and provide a more reliable data basis for subsequent analysis and prediction.

[0072] In a possible implementation, the annotated cave environment evolution pattern characteristics of the first template karst cave environment monitoring data include a first candidate cave environment evolution pattern and a first local karst cave space, the first candidate cave environment evolution pattern characterizing the cave environment evolution pattern corresponding to the annotated cave environment evolution pattern characteristics, and the first local karst cave space characterizing the karst cave environment partition of the annotated cave environment evolution pattern characteristics in the first template karst cave environment monitoring data.

[0073] Step S130 includes:

[0074] Step S131, using the cave environment risk prediction network generated in the a-th round of network parameter learning task to extract local state change characteristics of the first local karst cave space, and generate a first local state change characteristic corresponding to the first local karst cave space.

[0075] Step S132, spatially encode and represent the first local karst cave space to generate first environmental block spatial data corresponding to the first local karst cave space.

[0076] Step S133, load the first local state change characteristics and the first environmental block spatial data into the cave environment risk prediction network generated in the a-th round of network parameter learning task, and obtain the first estimated cave environment evolution pattern and estimated cave environment risk knowledge points corresponding to the first template karst cave environment monitoring data.

[0077] Step S134, performing parameter learning on the cave environment risk prediction network generated in the a-th round of network parameter learning task according to the third feature loss between the first estimated cave environment evolution pattern and the first candidate cave environment evolution pattern, as well as the labeled cave environment risk knowledge points and the estimated cave environment risk knowledge points, to generate the cave environment risk prediction network in the a+1-th round of network parameter learning task.

[0078] In a possible implementation, step S131 includes: obtaining a set threshold value, if the first scale corresponding to the first local state change feature is not greater than the set threshold value, adding a zero-value vector of the second scale, and the sum of the first scale and the second scale is not less than the set threshold value.

[0079] In this embodiment, in the context of karst cave A, the annotated cave environment evolution pattern feature of the first template karst cave environment monitoring data includes a first candidate cave environment evolution pattern and a first local karst cave space. The first candidate cave environment evolution pattern represents the cave environment evolution pattern corresponding to the annotated cave environment evolution pattern feature, and the first local karst cave space indicates the karst cave environment partition of the annotated cave environment evolution pattern feature in the first template karst cave environment monitoring data.

[0080] Assume that in karst cave A, the first local karst cave space is the area near the entrance of cave A. The environmental data of this area is included in the first template karst cave environmental monitoring data. The cave environmental risk prediction network begins to analyze this area. For example, the environmental parameter values ​​of the area near the entrance include air temperature, air humidity, carbon dioxide concentration, wind speed, light intensity, etc. The network will analyze these parameter values ​​to extract local state change characteristics. For the air temperature parameter value, it will analyze the temperature change amplitude and frequency in different time periods. For example, in a period of time, from morning to noon, the air temperature rises from 10 degrees Celsius to 15 degrees Celsius. The amplitude of this temperature rise, the rate of rise, and other information will be used as part of the local state change characteristics. For air humidity, if the humidity rises rapidly from 60% to 90% after rainfall, the sudden change in humidity, the time point of change, and other factors will also be taken into account in the local state change characteristics. At the same time, for parameters such as carbon dioxide concentration, wind speed, and light intensity, the changes in their mutual relationship with other parameters in this local area will also be analyzed. For example, the diffusion of carbon dioxide concentration when wind speed increases, the relationship between changes in light intensity and the respiration of microbial communities (assuming they exist) and thus the carbon dioxide concentration. In this process, it is also necessary to obtain a set threshold value. This set threshold value is a standard value obtained based on a large amount of previous data analysis or theoretical calculations, which is used to measure the scale of local state change characteristics. If after extracting the local state change characteristics of the area near the entrance, it is found that the first scale corresponding to the first local state change feature is not greater than the set threshold value, it is necessary to add a zero value vector of the second scale so that the sum of the first scale and the second scale is not less than the set threshold value. For example, assuming that the set threshold value is a certain value, and the scale of the first local state change feature obtained after analysis is less than this value, then a zero value vector is added according to certain rules to meet the scale requirements. These zero value vectors are added to ensure the integrity and consistency of the data structure in subsequent calculations and analyses.

[0081] Next, for the first local karst cave space near the entrance of cave A, the spatial coding representation will take multiple factors into consideration. For example, the geographical location information of the area, the direction of the entrance, the altitude, etc. Assuming that the entrance faces east and the altitude is 500 meters, these geographical information will be converted into a coding form. At the same time, the spatial range of the area will also be encoded. For example, the area of ​​the area near the entrance is 100 square meters, and this area information will also become part of the coding. In addition, factors such as the connectivity between the area and other areas in the cave will also be considered. If the area near the entrance is connected to a main passage inside the cave, this connectivity relationship will also be encoded into the first environmental block spatial data. By integrating the coding of these geographical, spatial range and connectivity factors, the first environmental block spatial data corresponding to the first local karst cave space is generated.

[0082] Then, when the first local state change characteristics of the area near the entrance (including the change characteristics of parameters such as air temperature, humidity, and carbon dioxide concentration) and the first environmental block spatial data (including encoding information such as geographic location, spatial range, and connectivity) are loaded into this network, the network will calculate according to its predefined algorithm. For example, the network may perform multi-layer analysis and processing on these input data based on machine learning algorithms, such as neural network algorithms. In the first layer of the network, the input local state change characteristics and spatial block data may be re-extracted and integrated, and different types of data may be converted into a unified and easier to analyze feature representation. Then, in subsequent layers, operations are performed based on these characteristics and the existing model parameters in the network. After multiple layers of calculation, the first estimated cave environment evolution pattern and estimated cave environment risk knowledge points corresponding to the first template karst cave environment monitoring data are obtained. It is assumed that the first estimated cave environment evolution pattern obtained is: due to the changes in temperature and humidity in the area near the entrance and the influence of connectivity with the inside of the cave, the structure of the biological community may be adjusted, and some organisms adapted to low temperature and low humidity environments may decrease, while organisms adapted to high temperature and high humidity environments may increase. It is estimated that the knowledge points about cave environmental risks may be: due to the adjustment of the biological community structure in the area near the entrance, the ecological balance may be affected to a certain extent. For example, some organisms have a protective effect on the cave wall (such as secreting special substances to prevent rock weathering), and a reduction in their number may increase the risk of cave wall weathering; at the same time, changes in biological respiration may affect the carbon dioxide concentration, and thus affect the stability of the chemical environment in the cave.

[0083] Finally, for the first estimated cave environment evolution model and the first candidate cave environment evolution model, for example, the first candidate cave environment evolution model is: changes in temperature and humidity in the area near the entrance will cause significant changes in the structure of the biological community, and will cause obvious fluctuations in the chemical environment in the cave. Compare this first candidate cave environment evolution model with the first estimated cave environment evolution model obtained previously, and calculate the third characteristic loss. In terms of comparing the adjustment of the biological community structure, the first candidate model emphasizes significant changes, while the first estimated model is that adjustments may occur. This difference in description needs to be quantified as part of the characteristic loss. For the impact on the chemical environment in the cave, one is obvious fluctuations, and the other may affect stability. This difference should also be quantified into the third characteristic loss. At the same time, for the annotated cave environment risk knowledge points and the estimated cave environment risk knowledge points, the annotated risk knowledge points may clearly point out the specific probability value that a certain biological reduction will lead to an increase in the risk of cave wall weathering, while the estimated risk knowledge points only mention that there is an increased risk. This difference in the quantification of the risk level and the accuracy of the association of risk factors must be quantified. Through these quantified differences, namely the third characteristic loss and the difference between the annotated cave environment risk knowledge points and the estimated cave environment risk knowledge points, the parameters of the cave environment risk prediction network generated in the a-th round of network parameter learning task are learned. The network will adjust its internal parameters according to these differences, such as adjusting the weight values, bias values ​​and other parameters in the neural network, so as to optimize the prediction ability of the network, and finally generate the cave environment risk prediction network in the a+1-th round of network parameter learning task, so that this new network can be more accurate in predicting cave environment risk knowledge points for karst cave environment monitoring data.

[0084] In a possible implementation, before step S133, the method further includes:

[0085] Acquire blending feature data, where the blending feature data is feature data generated after blending the first local state change feature and the first environment block spatial data.

[0086] In this embodiment, in the karst cave A, the first local state change feature includes the changes of various environmental parameters of a specific area (such as the area near the entrance of cave A) over time, such as the fluctuation amplitude of air temperature, the increase and decrease trend of air humidity, the change rate of carbon dioxide concentration, etc.; the first environmental block spatial data reflects the spatial attributes of this specific area, such as the geographical location (latitude and longitude, altitude, etc.) of the area near the entrance, the size of the spatial range, and the connectivity with other areas. The blending feature data is the feature data generated after the first local state change feature and the first environmental block spatial data are blended. This blending is not a simple combination, but is based on specific algorithms and rules. For example, for the area near the entrance, the change in air temperature may be related to the altitude of the area, and the blending process will embody this relationship in a quantitative way. Specifically, it may be possible to multiply the influence coefficient of altitude on air temperature by the actual change value of air temperature according to the empirical formula obtained by physical principles or historical data statistics, and then perform a comprehensive operation with other environmental parameter change values ​​and spatial attribute values ​​that have been similarly processed, thereby generating blending feature data. This blending feature data contains the intrinsic connection information between environmental state changes and spatial attributes.

[0087] Step S133 includes: loading the blending feature data, the first local state change feature and the first environmental block spatial data into the cave environment risk prediction network generated in the a-th round of network parameter learning task, and obtaining the first estimated cave environment evolution pattern and the estimated cave environment risk knowledge point.

[0088] Taking the neural network structure as an example, when the blending feature data, the first local state change feature and the first environmental block spatial data are input, the input layer of the network first receives these data. In the first hidden layer of the network, neurons will perform weighted summation operations on these input data. Each neuron has its own weight, which is determined during the early training or initialization process of the network. For example, a certain feature value associated with air temperature and altitude in the blending feature data may be given a higher weight by a neuron because this feature value may be of great significance in the prediction of cave environmental risks. After the weighted summation of the first hidden layer, a nonlinear transformation is performed through an activation function (such as a ReLU function) to obtain a new feature representation. These new feature representations will continue to perform similar operations in subsequent hidden layers to continuously extract and integrate features. In the last layer of the network, the output layer, the first estimated cave environment evolution pattern and estimated cave environment risk knowledge points will be generated based on the calculation results of the previous layers. The first estimated cave environment evolution pattern may be described as follows: in the area near the entrance of cave A, due to the changing trend of air temperature (derived from the temperature information in the first local state change feature), the spatial attributes of the area (derived from the first environmental block spatial data) and the blending relationship between the two (reflected by the blending feature data), the structure of the biological community will be adjusted in the direction of adapting to temperature changes and spatial environment. The number of some microbial communities adapted to low temperature environments may decrease, while the number of microbial communities adapted to higher temperatures and specific spatial layouts (such as microorganisms in areas near vents can more easily obtain oxygen, etc.) may increase. Knowledge points for estimating cave environmental risks may include that such adjustments in the structure of biological communities may affect the material cycle in the cave, for example, the decomposition process of organic matter in which microorganisms participate may be affected, thereby affecting the stability of the chemical environment in the cave; at the same time, the protective effect of the biological community on the cave wall (such as substances secreted by certain microorganisms can inhibit rock weathering) may change due to changes in community structure, increasing the risk of weathering of the cave wall.

[0089] In a possible implementation, before step S150, the method further includes:

[0090] Step A110, performing first feature selection on the first cave environment evolution pattern characteristics and the third cave environment evolution pattern characteristics to generate first selected feature data.

[0091] In the example of karst cave A, both the first cave environmental evolution pattern feature and the third cave environmental evolution pattern feature contain complex information about the environmental evolution of different areas of the cave. The first feature selection is based on specific selection criteria and algorithms. For example, the first cave environmental evolution pattern feature describes the environmental evolution of a certain area in the cave (such as the area near the underground river in the middle of cave A), including the impact of changes in the velocity of the underground river on the surrounding biological community, the impact of changes in cave wall temperature on the distribution of microbial communities, and other aspects; the third cave environmental evolution pattern feature describes the environmental evolution of another area in the cave (such as a specific area deep in cave A), involving the impact of changes in oxygen concentration on biological respiration, the impact of changes in rock temperature on the stability of cave structure, and other aspects. The first feature selection may be screened according to the importance of each feature to the overall environmental evolution of the cave. For example, features directly related to changes in biodiversity (such as changes in the structure of biological communities, changes in key environmental parameters for biological survival, etc.) will be given priority, while some features that have little impact on the overall environmental evolution or are highly correlated with other features (which can be obtained through correlation analysis) may be excluded. In this way, the first selected feature data is generated, and this data set contains feature information that is critical to the prediction of cave environmental evolution patterns.

[0092] Step A120, loading the blended feature data, the first local state change feature and the first environmental block spatial data into the cave environment risk prediction network generated in the a+1th round of network parameter learning task to obtain the target estimated cave environment evolution pattern.

[0093] This process is similar to the cave environment risk prediction network generated by loading these data into the a-th round network parameter learning task, but because it is in the a+1-th round network parameter learning task, the network parameters have been adjusted and optimized to a certain extent. When the blending feature data, the first local state change feature and the first environmental block spatial data are input, the network calculates according to its internal algorithm structure. For example, the weights in the network may have been updated according to the previous learning task, and the connection strength between neurons has changed, which makes the network process the input data differently. After multiple layers of calculations in the network, the target estimated cave environment evolution pattern is obtained. This target estimated cave environment evolution pattern may be described as an estimate of the environmental evolution pattern of a specific area in the cave (such as the area near the entrance mentioned above or other related areas) after considering the environment-space relationship in the blending feature data, the dynamic changes of the regional environment in the first local state change feature and the spatial attributes in the first environmental block spatial data. For example, in the area near the entrance, the target estimated cave environment evolution pattern may indicate that due to the interaction between some spatial factors that have not been considered before (such as the special influence of the entrance shape on airflow) and changes in environmental conditions (such as seasonal changes in air temperature and humidity), the adjustment amplitude and direction of the biological community structure are different from previous estimates, and the impact of this adjustment on the material circulation and chemical environment stability in the cave has also changed.

[0094] Step A130, matching the first selected feature data with the target estimated cave environment evolution pattern to generate second selected feature data matching the target estimated cave environment evolution pattern.

[0095] In this embodiment, this matching process is to further screen and optimize the feature information related to the target estimated cave environment evolution pattern. For example, some features in the first selected feature data are directly related to the description in the target estimated cave environment evolution pattern, such as the impact of the adjustment of the biological community structure on the material cycle mentioned in the target estimated cave environment evolution pattern, and the first selected feature data contains features related to the biological community structure, then these related features will be screened out as the second selected feature data. This second selected feature data is more focused on the key features that match the target estimated cave environment evolution pattern, and can be more effectively used in the subsequent parameter learning process.

[0096] Step S150 includes: performing parameter learning on the basic cave environment evolution pattern prediction network based on the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the second selected feature data and the third cave environment evolution pattern feature, to generate a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task.

[0097] In this embodiment, the annotated cave environment evolution pattern feature includes accurate annotation information of the cave environment evolution pattern, and the second cave environment evolution pattern feature is an estimate of the cave environment evolution pattern obtained through network prediction. The first feature loss between them is calculated. For example, the annotated cave environment evolution pattern feature points out that the change in the biological community structure in a certain area will lead to the precise change range of the concentration of a specific substance (such as carbon dioxide) within a certain period of time, and the predicted change range of the concentration of this substance in the second cave environment evolution pattern feature is different from it. This difference will be quantified as part of the first feature loss. At the same time, for the second feature loss between the second selected feature data and the third cave environment evolution pattern feature, the key features in the second selected feature data are compared with the relevant description in the third cave environment evolution pattern feature. For example, the feature of the protective effect of the biological community on the cave wall in the second selected feature data is related to the description of the stability of the cave wall in the third cave environment evolution pattern feature. If there is an inconsistency between the two (such as the second selected feature data believes that the biological community has a strong protective effect, while the third cave environment evolution pattern feature shows that the stability of the cave wall is reduced), this inconsistency will be quantified as part of the second feature loss. Based on these first characteristic losses and second characteristic losses, the parameters of the basic cave environment evolution pattern prediction network are learned. The network will adjust its internal parameters according to these loss values, such as adjusting the connection weights and bias values ​​between neurons. If the first characteristic loss indicates that there is a large deviation in the prediction of changes in substance concentrations caused by changes in biological community structure, the network may increase the neuron connection weights related to biological communities and substance concentrations in order to more accurately reflect this relationship in subsequent predictions. By continuously adjusting the parameters until the network converges, the cave environment evolution pattern prediction network in the a+1 round of network parameter learning tasks is generated, so that it can more accurately reflect the actual situation when predicting the cave environment evolution pattern of karst cave environment monitoring data.

[0098] In a possible implementation, the annotated cave environment evolution pattern characteristics of the first template karst cave environment monitoring data include a first candidate cave environment evolution pattern and a first local karst cave space, the first candidate cave environment evolution pattern characterizes the cave environment evolution pattern corresponding to the annotated cave environment evolution pattern characteristics, the first local karst cave space characterizes the karst cave environment partition of the annotated cave environment evolution pattern characteristics in the first template karst cave environment monitoring data, the first cave environment evolution pattern characteristics include a first estimated cave environment evolution pattern and a first estimated local karst cave space, the second cave environment evolution pattern characteristics include a second estimated cave environment evolution pattern and a second estimated local karst cave space, and the third cave environment evolution pattern characteristics include a third estimated cave environment evolution pattern and a third estimated local karst cave space.

[0099] Step S150 may further include:

[0100] The first feature loss is generated based on the first sub-feature loss between the first candidate cave environment evolution pattern and the second estimated cave environment evolution pattern, and the second sub-feature loss between the first local karst cave space and the second estimated local karst cave space.

[0101] The second characteristic loss is generated based on the third sub-characteristic loss between the first estimated cave environment evolution pattern and the third estimated cave environment evolution pattern, and the fourth sub-characteristic loss between the first estimated local karst cave space and the third estimated local karst cave space.

[0102] Based on the first feature loss and the second feature loss, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate the cave environment evolution pattern prediction network in the a+1th round of network parameter learning task.

[0103] In karst cave A, it is assumed that the first candidate cave environment evolution model indicates that in the area near the entrance of cave A (i.e., the first local karst cave space), due to seasonal climate change and airflow exchange inside and outside the cave, the air temperature and humidity will fluctuate periodically. This fluctuation will cause corresponding periodic changes in the microbial community structure. For example, the number of some microorganisms adapted to low temperature and high humidity environments will increase in winter, while the number of microorganisms adapted to high temperature and low humidity environments will increase in summer. At the same time, this change in the microbial community structure will affect the chemical cycle in the cave, such as the production and consumption rate of carbon dioxide. The second estimated cave environment evolution model predicts that in the area near the entrance, although the change in the microbial community structure is related to temperature and humidity, the cycle and amplitude of the change are different from the first candidate cave environment evolution model. For example, the predicted increase in the number of microorganisms adapted to low temperature and high humidity environments in winter is smaller than the actual annotation. Calculating this difference involves quantitative comparison of multiple aspects such as the amplitude and cycle of microbial population changes, thereby obtaining the first sub-feature loss.

[0104] For the second sub-feature loss between the first local karst cave space and the second estimated local karst cave space, it is assumed that the first local karst cave space accurately defines the scope of the area near the entrance, including its boundaries with other areas, specific geographic coordinate ranges, etc. However, the second estimated local karst cave space may have deviations from the actual scope in the prediction, such as inaccurate boundary definition or inclusion of some areas that should not be included. By comparing the differences between the two in terms of spatial scope definition, geographic location accuracy, etc., these differences are quantified to generate the second sub-feature loss. Combining the first sub-feature loss and the second sub-feature loss, the first feature loss is obtained.

[0105] The generation of the second characteristic loss is based on the third sub-characteristic loss between the first estimated cave environment evolution pattern and the third estimated cave environment evolution pattern, and the fourth sub-characteristic loss between the first estimated local karst cave space and the third estimated local karst cave space. For example, the first estimated cave environment evolution pattern describes the environmental evolution of the area near the underground river in the middle of cave A (i.e., the first estimated local karst cave space), and believes that changes in the water flow rate of the underground river will lead to changes in the surrounding soil moisture, thereby affecting the growth state of a small number of special plants (if any) growing in the area, and changes in the growth state of such plants will affect the surrounding microbial communities, changing the distribution and species composition of the microbial communities. In the prediction of the same area by the third estimated cave environment evolution pattern, although it is also believed that the water flow rate of the underground river affects soil moisture and plant growth, the degree of influence on the plant growth state and the specific way of influence on the microbial community are different from those of the first estimated cave environment evolution pattern. By quantifying these differences, such as comparing the predicted differences in plant growth indicators (such as height, number of leaves, etc.) and the predicted differences in changes in the species composition of the microbial community, the third sub-characteristic loss is obtained.

[0106] For the fourth sub-feature loss between the first estimated local karst cave space and the third estimated local karst cave space, it is assumed that the first estimated local karst cave space accurately describes the detailed spatial information of the central area close to the underground river, including the distance range from the underground river, the height of the terrain, etc. However, the third estimated local karst cave space has deviations in the description of these spatial information, such as inaccurate definition of the distance range from the underground river. By comparing and quantifying the differences in these spatial information, the fourth sub-feature loss is obtained. Combining the third sub-feature loss and the fourth sub-feature loss, the second feature loss is generated.

[0107] The parameters of the basic cave environment evolution pattern prediction network are learned based on the first characteristic loss and the second characteristic loss to generate the cave environment evolution pattern prediction network in the a+1 round network parameter learning task. The basic cave environment evolution pattern prediction network contains many parameters, such as weights and biases in the neural network. If the first characteristic loss indicates that there is a large deviation in the prediction of the change in the microbial community structure in the environmental evolution pattern prediction of the area near the entrance (such as a large first sub-characteristic loss), the network will adjust the weights of the input parameters related to the microbial community, such as the weights related to the number of microbial communities, air temperature, humidity, etc. If the second characteristic loss shows that there is a deviation in the prediction of the impact of the underground river flow velocity on plants and microbial communities in the environmental evolution pattern prediction of the central area near the underground river (such as a large third sub-characteristic loss), the network will adjust the weights of the input parameters related to the underground river flow velocity, soil moisture, plant growth indicators, etc. By continuously adjusting the network parameters according to the first characteristic loss and the second characteristic loss until the network convergence requirements are met, the cave environment risk prediction network and the cave environment evolution pattern prediction network with completed parameter learning are generated.

[0108] In a possible implementation manner, after step S150, the method further includes:

[0109] Step S160, obtaining target karst cave environment monitoring data.

[0110] Step S170, loading the target karst cave environment monitoring data into the cave environment evolution pattern prediction network and the cave environment risk prediction network that have completed parameter learning, obtaining the target cave environment evolution pattern characteristics corresponding to the target karst cave environment monitoring data and the cave environment risk knowledge points corresponding to the target cave environment evolution pattern characteristics.

[0111] After completing parameter learning, the target karst cave environment monitoring data is obtained. This target karst cave environment monitoring data can come from a new monitoring area of ​​karst cave A or data obtained by using new monitoring technology or a more sophisticated monitoring frequency for an existing monitoring area. For example, a new monitoring station may be set up deep in cave A, or the monitoring frequency of the area near the entrance may be increased from once an hour to once every half an hour.

[0112] The target karst cave environment monitoring data is loaded into the cave environment evolution pattern prediction network and cave environment risk prediction network that have completed parameter learning, and the target cave environment evolution pattern characteristics corresponding to the target karst cave environment monitoring data and the cave environment risk knowledge points corresponding to the target cave environment evolution pattern characteristics are obtained. For each monitoring record in the target karst cave environment monitoring data, multiple environmental parameter values ​​are included, such as air temperature, humidity, carbon dioxide concentration, etc. When loaded into the cave environment evolution pattern prediction network, the network processes these data according to the parameters it has learned. For example, if the target karst cave environment monitoring data is data from a new monitoring station deep in cave A, the network will predict the target cave environment evolution pattern characteristics based on the special environmental conditions in the depths (such as lower temperature, higher carbon dioxide concentration, etc.) and the previously learned laws of the cave environment evolution pattern. This feature may be described as deep in cave A, due to the lower temperature and higher carbon dioxide concentration, the microbial community structure presents a special distribution state, and this state may be affected by changes in rock temperature over time.

[0113] At the same time, the target karst cave environment monitoring data is loaded into the cave environment risk prediction network to obtain cave environment risk knowledge points corresponding to the target cave environment evolution pattern characteristics. For example, based on the predicted special distribution state of the microbial community structure and possible change trends, the cave environment risk prediction network may conclude that deep in cave A, such changes in the microbial community structure may cause certain rare microorganisms to face risks in survival, thereby affecting the stability of material circulation in the cave; or due to the interaction between the microbial community and the rock (such as the erosion or protection of the rock by microorganisms), the risk of instability of the cave wall structure may increase, and other cave environment risk knowledge points.

[0114] Figure 2 The hardware structure of the karst cave environment data analysis system 100 based on artificial intelligence for implementing the karst cave environment data analysis method based on artificial intelligence provided by an embodiment of the present invention is shown as follows: Figure 2 As shown, the artificial intelligence-based karst cave environment data analysis system 100 may include a processor 110 , a machine-readable storage medium 120 , a bus 130 , and a communication unit 140 .

[0115] 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 that the artificial intelligence-based karst cave environment data analysis system 100 uses to execute or use to complete the exemplary method described in the present invention.

[0116] 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 karst cave environment data analysis method based on artificial intelligence 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.

[0117] The specific implementation process of the processor 110 can refer to the various method embodiments executed by the above-mentioned artificial intelligence-based karst cave environment data analysis system 100. The implementation principles and technical effects are similar, and this embodiment will not be repeated here.

[0118] In addition, an embodiment of the present invention also provides a readable storage medium, in which computer executable instructions are preset. When the processor executes the computer executable instructions, the above-mentioned karst cave environment data analysis method based on artificial intelligence is implemented.

[0119] 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, various features are sometimes combined into one embodiment, drawing or description thereof.

Claims

1. A karst cave environment data analysis method based on artificial intelligence, characterized in that: The method comprises: Acquire first template karst cave environment monitoring data and second template karst cave environment monitoring data, wherein the first template karst cave environment monitoring data carries marked cave environment evolution pattern characteristics and marked cave environment risk knowledge points corresponding to the marked cave environment evolution pattern characteristics, and the second template karst cave environment monitoring data carries marked cave environment risk knowledge points corresponding to the marked cave environment evolution pattern characteristics, and the first template karst cave environment monitoring data and the second template karst cave environment monitoring data both contain monitoring records arranged in chronological order, and each monitoring record contains multiple environmental parameter values, and the environmental parameter values ​​include one or more combinations of air temperature parameter values, air humidity parameter values, carbon dioxide concentration parameter values, air pressure parameter values, cave internal structure change parameter values, oxygen concentration parameter values, wind speed parameter values, light intensity parameter values, methane concentration parameter values, hydrogen sulfide concentration parameter values, cave water flow velocity parameter values, cave water pH parameter values, soil moisture parameter values, rock temperature parameter values, radon gas concentration parameter values, microbial community quantity parameter values, cave wall roughness parameter values, cave top dripping velocity parameter values, negative ion concentration parameter values, and cave dust concentration parameter values; In the a+1th round of network parameter learning task, the cave environment evolution pattern prediction network generated in the ath round of network parameter learning task is used to predict the second template karst cave environment monitoring data to generate a first cave environment evolution pattern feature, wherein the cave environment evolution pattern prediction network is used to estimate the cave environment evolution pattern feature in the karst cave environment monitoring data, and a is a positive integer; Using the first template karst cave environment monitoring data, parameter learning is performed on the cave environment risk prediction network generated in the a-th round of network parameter learning tasks to generate a cave environment risk prediction network in the a+1-th round of network parameter learning tasks, wherein the cave environment risk prediction network is used to predict cave environment risk knowledge points of cave environment evolution pattern characteristics based on the karst cave environment monitoring data; Using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task, the first template karst cave environment monitoring data and the second template karst cave environment monitoring data are predicted to generate second cave environment evolution pattern features corresponding to the first template karst cave environment monitoring data and third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data; According to the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, as well as the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, until the network convergence requirements are met to generate a cave environment risk prediction network and a cave environment evolution pattern prediction network that have completed parameter learning.

2. The karst cave environment data analysis method based on artificial intelligence according to claim 1 is characterized in that: The method uses the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task to predict the first template karst cave environment monitoring data and the second template karst cave environment monitoring data, and generates the second cave environment evolution pattern characteristics corresponding to the first template karst cave environment monitoring data and the third cave environment evolution pattern characteristics corresponding to the second template karst cave environment monitoring data, including: Using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task to predict the first template karst cave environment monitoring data, generating the second cave environment evolution pattern features corresponding to the first template karst cave environment monitoring data; The second template karst cave environment monitoring data is predicted using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning tasks and the cave environment risk prediction network generated in the a+1th round of network parameter learning tasks to generate the third cave environment evolution pattern characteristics corresponding to the second template karst cave environment monitoring data.

3. The karst cave environment data analysis method based on artificial intelligence according to claim 2 is characterized in that: The basic cave environment evolution pattern prediction network includes an environment state change encoding subnetwork and a fully connected mapping subnetwork; The method of using the basic cave environment evolution pattern prediction network in the a+1th round of network parameter learning task and the cave environment risk prediction network generated in the a+1th round of network parameter learning task to predict the second template karst cave environment monitoring data to generate the third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data includes: Using the environmental state change encoding subnetwork to encode the environmental state change of the second template karst cave environmental monitoring data, and generate environmental state change trajectory features corresponding to the second template karst cave environmental monitoring data; Using the cave environment risk prediction network generated in the a+1th round of network parameter learning task, extracting local state change features from the environmental state change trajectory features corresponding to the second template karst cave environment monitoring data, and generating second local state change features corresponding to the annotated cave environment evolution pattern features; Integrating the environmental state change trajectory features and the local state change features to generate enhanced trajectory features; Loading the enhanced trajectory features into the fully connected mapping subnetwork to generate the third cave environment evolution pattern features corresponding to the second template karst cave environment monitoring data; The step of encoding the environmental state change of the second template karst cave environment monitoring data using the environmental state change encoding subnetwork to generate environmental state change trajectory features corresponding to the second template karst cave environment monitoring data includes: Extracting features from each monitoring record in the second template karst cave environment monitoring data to generate an initial feature vector sequence reflecting changes in the cave environment state; Based on the initial feature vector sequence, a clustering algorithm is used to cluster the matching environmental states into the same category, each category represents an environmental state node, and according to the time sequence and the similarity between the states, the intra-node transfer relationship within the environmental state node and the inter-node transfer relationship between the environmental state nodes are constructed to generate a state transition diagram, which describes the dynamic change process of the cave environmental state over time; Traversing the state transition diagram, starting from the initial state, assigning a unique code to the environment state at each time point according to the time sequence and the state transition relationship, and generating a state sequence code, wherein the unique code is used to uniquely identify the environment state at the time point; Arranging the state sequence codes in chronological order to form an environmental state change trajectory; Smoothing filtering is used to remove noise and jitter in the environmental state change trajectory, and time series analysis is used to extract periodic characteristics and trend characteristics in the environmental state change trajectory. Pattern recognition is used to identify target change patterns or abnormal points in the environmental state change trajectory to obtain enhanced environmental state change trajectory characteristics.

4. The karst cave environment data analysis method based on artificial intelligence according to any one of claims 1 to 3, characterized in that: The annotated cave environment evolution pattern characteristics of the first template karst cave environment monitoring data include a first candidate cave environment evolution pattern and a first local karst cave space, wherein the first candidate cave environment evolution pattern represents the cave environment evolution pattern corresponding to the annotated cave environment evolution pattern characteristics, and the first local karst cave space represents the karst cave environment partition of the annotated cave environment evolution pattern characteristics in the first template karst cave environment monitoring data; The method of using the first template karst cave environment monitoring data to perform parameter learning on the cave environment risk prediction network generated in the a-th round of network parameter learning task to generate the cave environment risk prediction network in the a+1-th round of network parameter learning task includes: Using the cave environment risk prediction network generated in the a-th round of network parameter learning task, extracting local state change features of the first local karst cave space to generate first local state change features corresponding to the first local karst cave space; Performing spatial coding representation on the first local karst cave space to generate first environmental block spatial data corresponding to the first local karst cave space; Loading the first local state change feature and the first environmental block spatial data into the cave environment risk prediction network generated in the a-th round of network parameter learning task, obtaining a first estimated cave environment evolution pattern and an estimated cave environment risk knowledge point corresponding to the first template karst cave environment monitoring data; According to the third feature loss between the first estimated cave environment evolution pattern and the first candidate cave environment evolution pattern, as well as the labeled cave environment risk knowledge points and the estimated cave environment risk knowledge points, parameter learning is performed on the cave environment risk prediction network generated in the a-th round of network parameter learning task to generate the cave environment risk prediction network in the a+1-th round of network parameter learning task.

5. The karst cave environment data analysis method based on artificial intelligence according to claim 4 is characterized in that: The method of extracting local state change features of the first local karst cave space using the cave environment risk prediction network generated in the a-th round of network parameter learning task to generate first local state change features corresponding to the first local karst cave space includes: Get the set threshold value; If the first scale corresponding to the first local state change feature is not greater than the set threshold value, a zero-value vector of a second scale is added, and the sum of the first scale and the second scale is not less than the set threshold value.

6. The karst cave environment data analysis method based on artificial intelligence according to claim 4 is characterized in that: Before the first local state change feature and the first environmental block spatial data are loaded into the cave environment risk prediction network generated in the a-th round network parameter learning task, and a first estimated cave environment evolution pattern and an estimated cave environment risk knowledge point corresponding to the first template karst cave environment monitoring data are obtained, the method further includes: Acquire blending feature data, where the blending feature data is feature data generated by blending the first local state change feature and the first environmental block spatial data; The step of loading the first local state change feature and the first environmental block spatial data into the cave environment risk prediction network generated in the a-th round of network parameter learning task to obtain a first estimated cave environment evolution pattern and estimated cave environment risk knowledge points corresponding to the first template karst cave environment monitoring data includes: The blending feature data, the first local state change feature and the first environmental block spatial data are loaded into the cave environment risk prediction network generated in the a-th round of network parameter learning task to obtain the first estimated cave environment evolution pattern and the estimated cave environment risk knowledge points.

7. The karst cave environment data analysis method based on artificial intelligence according to claim 6 is characterized in that: Before performing parameter learning on the basic cave environment evolution pattern prediction network based on the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, and generating the cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, the method further includes: Performing first feature selection on the first cave environment evolution pattern feature and the third cave environment evolution pattern feature to generate first selected feature data; Loading the blending feature data, the first local state change feature and the first environmental block spatial data into the cave environment risk prediction network generated in the a+1th round of network parameter learning task to obtain a target estimated cave environment evolution pattern; Matching the first selected feature data with the target estimated cave environment evolution pattern to generate second selected feature data matching the target estimated cave environment evolution pattern; The method of performing parameter learning on the basic cave environment evolution pattern prediction network based on the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, to generate the cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, includes: According to the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, as well as the second feature loss between the second selected feature data and the third cave environment evolution pattern feature, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task.

8. The karst cave environment data analysis method based on artificial intelligence according to any one of claims 1 to 3, characterized in that: The annotated cave environment evolution pattern characteristics of the first template karst cave environment monitoring data include a first candidate cave environment evolution pattern and a first local karst cave space, the first candidate cave environment evolution pattern represents the cave environment evolution pattern corresponding to the annotated cave environment evolution pattern characteristics, the first local karst cave space represents the karst cave environment partition of the annotated cave environment evolution pattern characteristics in the first template karst cave environment monitoring data, the first cave environment evolution pattern characteristics include a first estimated cave environment evolution pattern and a first estimated local karst cave space, the second cave environment evolution pattern characteristics include a second estimated cave environment evolution pattern and a second estimated local karst cave space, and the third cave environment evolution pattern characteristics include a third estimated cave environment evolution pattern and a third estimated local karst cave space; The method of performing parameter learning on the basic cave environment evolution pattern prediction network based on the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, to generate the cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, includes: generating the first feature loss according to the first sub-feature loss between the first candidate cave environment evolution pattern and the second estimated cave environment evolution pattern, and the second sub-feature loss between the first local karst cave space and the second estimated local karst cave space; generating the second characteristic loss according to the third sub-characteristic loss between the first estimated cave environment evolution pattern and the third estimated cave environment evolution pattern, and the fourth sub-characteristic loss between the first estimated local karst cave space and the third estimated local karst cave space; Based on the first feature loss and the second feature loss, parameter learning is performed on the basic cave environment evolution pattern prediction network to generate the cave environment evolution pattern prediction network in the a+1th round of network parameter learning task.

9. The karst cave environment data analysis method based on artificial intelligence according to any one of claims 1 to 3, characterized in that: The method further comprises: performing parameter learning on the basic cave environment evolution pattern prediction network according to the first feature loss between the labeled cave environment evolution pattern feature and the second cave environment evolution pattern feature, and the second feature loss between the first cave environment evolution pattern feature and the third cave environment evolution pattern feature, generating a cave environment evolution pattern prediction network in the a+1th round of network parameter learning task, until the cave environment risk prediction network and the cave environment evolution pattern prediction network with completed parameter learning are generated when the network convergence requirements are met, and further comprising: Obtaining environmental monitoring data of target karst caves; The target karst cave environment monitoring data is loaded into the cave environment evolution pattern prediction network and the cave environment risk prediction network that have completed parameter learning, and the target cave environment evolution pattern characteristics corresponding to the target karst cave environment monitoring data and the cave environment risk knowledge points corresponding to the target cave environment evolution pattern characteristics are obtained.

10. A karst cave environment data analysis system based on artificial intelligence, characterized in that: The artificial intelligence-based karst cave environment data analysis system 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 artificial intelligence-based karst cave environment data analysis method described in any one of claims 1 to 9 above.

Images