Geological survey method based on big data analysis

Abstract

The invention relates to the technical field of geological survey. The geological survey method based on big data analysis comprises the steps of obtaining multi-source heterogeneous survey data, and performing integration processing on the multi-source heterogeneous survey data to obtain integrated survey data; constructing a time sequence progress analysis model, extracting a progress feature vector of each survey link, and processing each progress feature vector to obtain an abnormal node of progress lag; generating a progress adjustment scheme by using a bottom-up dynamic programming algorithm; acquiring a preset environmental influence parameter, and processing the environmental influence parameter by using a Bayesian network model to obtain an environmental influence result; inputting the environmental influence result into the trained resource scheduling optimization model for processing to obtain a resource configuration scheme; and inputting the progress adjustment scheme and the resource configuration scheme into the trained multi-target optimization model for processing to obtain an exploration progress management strategy so as to realize effective dynamic management of the exploration progress and improve the efficiency of the overall exploration task.

Description

Technical Field

[0001] The present invention relates to the technical field of geological exploration, and particularly to a geological exploration method based on big data analysis. Background Art

[0002] Geological exploration is an important support for the development of many fields. In the field of resource exploration, it can locate various minerals and provide raw materials for industry; in the energy field, it helps to explore oil, natural gas, etc. to ensure energy security; in the field of engineering construction, it understands the geological conditions through exploration to ensure the safety and quality of projects. In geological exploration tasks, the key to the progress of exploration lies in how to effectively manage the exploration progress dynamically. First, the complexity of exploration tasks leads to the diversification of exploration steps, and there are interdependent relationships between different steps. For example, the sequence and progress of steps such as geological sampling, drilling, and geophysical exploration directly affect the overall exploration efficiency. If the progress of a certain link lags behind, it may cause delays in subsequent links and even affect the completion time of the entire exploration task.

[0003] However, in geological exploration tasks, related technologies cannot effectively manage the exploration progress dynamically, resulting in difficulty in improving the efficiency of the overall exploration task.

[0004] In view of this, there is an urgent need for a method to solve the above problems. Summary of the Invention

[0005] Based on this, in order to solve the above technical problems, it is necessary to provide a geological exploration method based on big data analysis to achieve effective dynamic management of the exploration progress and improve the efficiency of the overall exploration task.

[0006] The present application provides a geological exploration method based on big data analysis, and the method includes:

[0007] Obtain multi-source heterogeneous exploration data, and perform integration processing on the multi-source heterogeneous exploration data to obtain integrated exploration data;

[0008] Based on the integrated exploration data, construct a time series progress analysis model, use the sliding window method to extract the progress feature vectors of each exploration link, and process each progress feature vector to obtain abnormal nodes with lagging progress;

[0009] Based on the abnormal nodes, generate a progress adjustment plan using a bottom-up dynamic programming algorithm;

[0010] Obtain preset environmental impact parameters, and process the environmental impact parameters using a Bayesian network model to obtain environmental impact results. The environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data;

[0011] Input the environmental impact results into the trained resource scheduling optimization model for processing to obtain a resource allocation plan, which is used to allocate resources including manpower and equipment;

[0012] Input the progress adjustment plan and the resource allocation plan into the trained multi-objective optimization model for processing to obtain a survey progress management strategy.

[0013] Furthermore, use the Bayesian network model to process the environmental impact parameters to obtain environmental impact results. The environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data, including:

[0014] Use the following formula to calculate the probability distribution results of the impact deviation corresponding to each data based on the preset environmental impact factor data, real-time meteorological data, and terrain data:

[0015]

[0016] where P(E i ) represents the probability distribution of the environmental factor impact, n represents the number of samples, x i represents the measured value of the environmental factor, μ represents the mean value of the environmental factor, and σ represents the standard deviation;

[0017] Based on the probability distribution results, obtain the key impact factors;

[0018] Establish the mapping relationship between environmental factors and survey progress;

[0019] Use the following formula to process the key impact factors based on the mapping relationship to obtain the environmental impact results:

[0020]

[0021] where R represents the environmental impact result, l represents the number of evaluation indicators, γ i represents the weight coefficient of the i-th indicator, λ i represents the time decay coefficient, t represents time, and h i represents the terrain correction coefficient.

[0022] Furthermore, establishing the mapping relationship between environmental factors and survey progress includes:

[0023] Obtain the environmental data of the target area. The environmental data includes weather conditions, terrain conditions, and soil attribute data;

[0024] Construct a regression analysis model between environmental data and survey efficiency;

[0025] Based on the regression analysis model, calculate and confirm the environmental factors whose influence coefficients are greater than the preset threshold;

[0026] Use the random forest algorithm to process environmental factors to obtain key influencing factors;

[0027] Generate a progress prediction result based on the key influencing factors.

[0028] Furthermore, based on the abnormal nodes, use a bottom-up dynamic programming algorithm to generate a progress adjustment plan, including:

[0029] Use the following formula to construct a multi-stage decision-making model and define the transfer relationship between each survey link:

[0030]

[0031] where s t represents the survey state at time t, a t represents the decision-making action at time t, R represents the immediate reward function, γ represents the discount factor, and V represents the value function;

[0032] Use the following formula to process each survey link using the dynamic programming algorithm to obtain the corresponding optimal sub-structure value:

[0033]

[0034] where D k represents the decision value of the k-th survey link, ω ij represents the transfer weight from the i-th task to the j-th task, and d ij represents the transfer distance between tasks;

[0035] Determine the execution order of the survey links based on the optimal sub-structure value;

[0036] Use the following formula to process the execution order based on the similarity algorithm to generate the final progress adjustment plan:

[0037]

[0038]

[0039] where P represents the similarity between two survey links, α k represents the weight coefficient of the k-th feature, sim represents the similarity function of a single feature, x represents the feature vector of the survey link, T(n) represents the optimal adjustment time of n tasks, c k represents the completion time of the k-th task, p i represents the probability of the i-th adjustment plan, and m represents the number of optional adjustment plans.

[0040] Further, input the environmental impact results into the trained resource scheduling optimization model for processing to obtain a resource allocation plan, where the resource allocation plan is used to allocate resources including manpower and equipment, including:

[0041] Construct an objective function based on the environmental impact results, and through the particle swarm algorithm and population evolution iteration, obtain the fitness values corresponding to each initial plan;

[0042] Judge the fitness values to obtain a judgment result, where the judgment result includes that the fitness value reaches the global optimum and the fitness value does not reach the global optimum;

[0043] When the judgment result is that the fitness value does not reach the global optimum, then through the particle swarm algorithm and population evolution iteration, obtain the fitness value that reaches the global optimum;

[0044] When the judgment result is that the fitness value reaches the global optimum, then generate a resource allocation plan according to the fitness value.

[0045] Further, input the progress adjustment plan and the resource allocation plan into the trained multi-objective optimization model for processing to obtain a survey progress management strategy, including:

[0046] Input the progress adjustment plan and the resource allocation plan into the multi-objective optimization model, and process them through the particle swarm algorithm and population evolution iteration to obtain a survey progress management strategy.

[0047] Further, obtain multi-source heterogeneous survey data, and perform integration processing on the multi-source heterogeneous survey data to obtain integrated survey data, including:

[0048] Obtain multi-source heterogeneous survey data;

[0049] Analyze the data format of the multi-source heterogeneous survey data, and extract the corresponding timestamps and accuracy values;

[0050] Based on the data format, timestamps, and accuracy values, adjust the multi-source heterogeneous survey data to obtain multi-source data with unified time and space;

[0051] Based on the consistency of the timestamps and accuracy values, process the multi-source data to obtain integrated survey data.

[0052] Further, based on the optimal substructure values, determine the execution order of the survey links, including:

[0053] Obtain the optimal substructure values corresponding to each survey link, and process each optimal substructure value to obtain the backtracking value of the global optimal path;

[0054] Based on the backtracking value, determine the corresponding survey sequence;

[0055] Generate the execution order of the survey link based on the survey sequence.

[0056] Furthermore, construct an objective function based on the environmental impact results, and through the particle swarm algorithm and population evolution iteration, obtain the fitness values corresponding to each initial plan, including:

[0057] Use the following formula to construct an objective function based on the environmental impact results, and through the particle swarm algorithm and population evolution iteration, obtain the fitness values corresponding to each initial plan:

[0058]

[0059]

[0060] Among them, F(x) represents the particle velocity update equation, ω represents the inertia weight, v represents the current particle velocity, c1 and c2 represent the learning factors, r1 and r2 represent random numbers, p best represents the individual optimal position, g best represents the global optimal position, x represents the current position, P(i) represents the population selection probability calculation equation, β represents the selection pressure parameter, f represents the individual fitness value, N represents the population size, P represents the probability of being selected, fitness represents the fitness calculation function, α represents the weight coefficient, y represents the actual environmental impact value, represents the predicted environmental impact value, and n represents the number of environmental impact indicators.

[0061] Furthermore, based on the consistency of the timestamp and accuracy value, process multi-source data to obtain integrated survey data, including:

[0062] Use the following formula to obtain the data quality evaluation index based on multi-source data:

[0063]

[0064] Among them, Q represents the data quality evaluation index, x i represents the i-th measurement value, μ represents the mean of the measurement values, σ i represents the standard deviation of the i-th measurement value, and n1 represents the sample size.

[0065] The technical solution provided by this application has the following beneficial effects: By providing a geological survey method based on big data analysis, including: obtaining multi-source heterogeneous survey data, and integrating and processing the multi-source heterogeneous survey data to obtain integrated survey data; based on the integrated survey data, constructing a time series progress analysis model, using the sliding window method to extract the progress feature vectors of each survey link, processing each progress feature vector to obtain abnormal nodes with lagging progress; based on the abnormal nodes, using a bottom-up dynamic programming algorithm to generate a progress adjustment plan; obtaining preset environmental impact parameters, using a Bayesian network model to process the environmental impact parameters to obtain an environmental impact result, where the environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data; inputting the environmental impact result into a trained resource scheduling optimization model for processing to obtain a resource allocation plan, where the resource allocation plan is used to allocate resources including manpower and equipment; inputting the progress adjustment plan and the resource allocation plan into a trained multi-objective optimization model for processing to obtain a survey progress management strategy, so as to achieve effective dynamic management of the survey progress and improve the efficiency of the overall survey task. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0067] Figure 1 It is a flowchart of a geological survey method based on big data analysis in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] In order to make the objectives, technical solutions and advantages of this application clearer, the following further elaborates on this application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0069] As Figure 1 shown, this application provides a geological survey method based on big data analysis, and the method includes:

[0070] S101: Obtain multi-source heterogeneous survey data, and integrate and process the multi-source heterogeneous survey data to obtain integrated survey data.

[0071] Specifically, obtain multi-source heterogeneous survey data covering a wide range of sources such as geophysical data, geochemical data, and topographic data, with different formats and structures. Then, integrate and process the multi-source heterogeneous survey data through operations such as data cleaning, format conversion, and spatio-temporal alignment to obtain integrated survey data with a unified format, consistent structure, and convenient for subsequent analysis and use. The beneficial effect of this step is that it can eliminate the differences and contradictions between data, improve data quality, and provide a more comprehensive, accurate, and reliable data basis for subsequent geological survey analysis.

[0072] S102: Based on the integrated survey data, construct a time-series progress analysis model, use the sliding window method to extract the progress feature vectors of each survey link, and process each progress feature vector to obtain abnormal nodes with lagging progress.

[0073] Specifically, based on the integrated survey data covering various geological survey information after integration, build a time-series progress analysis model that can analyze the survey progress situation changing over time. Use the sliding window method to obtain the progress feature vectors, and then perform analysis and calculation on the progress feature vectors to find out those abnormal nodes with lagging progress. The beneficial effect of this step is that it can locate the problems in the survey progress, timely discover abnormal situations that may affect the overall progress, so as to take targeted measures to ensure the efficient and orderly progress of geological survey work. Among them, the time-series progress analysis model is a data analysis tool that collects, organizes, and analyzes relevant data arranged in chronological order, mines the trends, cycles, and other features contained in the data to evaluate and predict the progress of things at different time points, and provides a basis for decision-making and management. It is often applied in many fields such as economy, engineering, and meteorology.

[0074] S103: Based on the abnormal nodes, use the bottom-up dynamic programming algorithm to generate a progress adjustment plan.

[0075] Specifically, after identifying the abnormal nodes with lagging progress in geological survey, use the bottom-up dynamic programming algorithm to generate a progress adjustment plan. The bottom-up dynamic programming algorithm is an algorithm that starts from the optimal solutions of sub-problems and gradually derives the optimal solution of the overall problem. It will first decompose the progress adjustment problem into multiple sub-problems, start solving from the most basic and simplest sub-problems, and then use the solutions of these sub-problems to solve more complex sub-problems until the optimal plan for the entire progress adjustment is obtained. The beneficial effect of this step is that it can comprehensively consider the mutual relationships and constraint conditions between each link in the survey process, generate a scientific and reasonable progress adjustment plan with global optimality, avoid local optimization while ignoring the overall effect, and effectively improve the overall progress and efficiency of geological survey.

[0076] S104: Obtain preset environmental impact parameters, and use a Bayesian network model to process the environmental impact parameters to obtain an environmental impact result. The environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data.

[0077] Specifically, during the geological survey process, it is necessary to obtain preset environmental impact parameters, and then use a Bayesian network model to process these environmental impact parameters. The Bayesian network model is a graphical model based on probabilistic reasoning. It can calculate the possibility and degree of various environmental factors affecting geological surveys based on the causal relationships and probability distributions among known environmental impact parameters, thereby obtaining an environmental impact result. The beneficial effect of this step is that by comprehensively considering various environmental factors and analyzing them with the help of a Bayesian network model, it is possible to more accurately evaluate the potential impact of the environment on geological survey work, formulate countermeasures in advance, reduce the survey risks caused by environmental factors, improve the safety and efficiency of geological survey work, and ensure the smooth progress of the entire survey project.

[0078] S105: Input the environmental impact result into a trained resource scheduling optimization model for processing to obtain a resource allocation plan, which is used to allocate resources including manpower and equipment.

[0079] Specifically, input the environmental impact result into a resource scheduling optimization model that can achieve reasonable resource arrangement for analysis and calculation. This model will comprehensively consider environmental impacts and other relevant factors to obtain a resource allocation plan for resources including manpower and equipment. The beneficial effect of this step is that it is possible to scientifically and reasonably allocate resources based on the actual impact of the environment on survey work, avoid waste or shortage of resources, improve resource utilization efficiency, ensure the smooth progress of geological survey work under different environmental conditions, and obtain the best survey results with the optimal resource input. Among them, the resource scheduling optimization model is a comprehensive analysis and decision-making model aiming to rationally allocate and arrange various available resources among different tasks, projects, or processes by using mathematical algorithms, intelligent strategies, etc., to achieve one or more goals such as maximizing resource utilization efficiency, minimizing costs, and shortest task completion time.

[0080] S106: Input the schedule adjustment plan and the resource allocation plan into a trained multi-objective optimization model for processing to obtain a geological survey schedule management strategy.

[0081] Specifically, the beneficial effects of this step are as follows: It can fully integrate the advantages of schedule adjustment and resource allocation, enabling the geological survey work to reach an optimal state in terms of schedule advancement and resource utilization, improving the overall survey efficiency, reducing costs, and ensuring the high-quality completion of geological survey tasks. Among them, the multi-objective optimization model is a data model used to handle multiple objectives that are mutually restrictive, even conflicting, and need to be optimized simultaneously in a system or problem. It seeks to achieve a certain balance or optimal compromise among these objectives through various mathematical methods and algorithms to obtain a set of non-inferior solutions or Pareto optimal solutions that can make each objective reach a relatively optimal level as much as possible.

[0082] By providing a geological survey method based on big data analysis, including: obtaining multi-source heterogeneous survey data, and integrating and processing the multi-source heterogeneous survey data to obtain integrated survey data; based on the integrated survey data, constructing a time series progress analysis model, using the sliding window method to extract the progress feature vectors of each survey link, processing each progress feature vector to obtain abnormal nodes with lagging progress; based on the abnormal nodes, using a bottom-up dynamic programming algorithm to generate a schedule adjustment plan; obtaining preset environmental impact parameters, using a Bayesian network model to process the environmental impact parameters to obtain environmental impact results, where the environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data; inputting the environmental impact results into a trained resource scheduling optimization model for processing to obtain a resource allocation plan, and the resource allocation plan is used to allocate resources including manpower and equipment; inputting the schedule adjustment plan and the resource allocation plan into a trained multi-objective optimization model for processing to obtain a survey schedule management strategy, so as to realize the effective dynamic management of the survey schedule and improve the efficiency of the overall survey task.

[0083] Furthermore, using a Bayesian network model to process the environmental impact parameters to obtain environmental impact results, where the environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data, may include the following steps:

[0084] (1) Using the following formula, calculate the probability distribution results of the impact deviations corresponding to each data based on the preset environmental impact factor data, real-time meteorological data, and terrain data:

[0085]

[0086] Among them, P(E i ) represents the probability distribution of the environmental factor impact, n represents the sampling quantity, x i represents the measured value of the environmental factor, μ represents the mean value of this environmental factor, and σ represents the standard deviation;

[0087] (2) Based on the probability distribution results, obtain the key impact factors;

[0088] (3) Establish a mapping relationship between environmental factors and survey progress;

[0089] (4) Use the following formula to process the key impact factors based on the mapping relationship to obtain the environmental impact result:

[0090]

[0091] where R represents the environmental impact result, l represents the number of evaluation indicators, γ i represents the weight coefficient of the i-th indicator, λ i represents the time decay coefficient, t represents time, h i represents the terrain correction coefficient.

[0092] Specifically, first, according to the preset environmental impact factor data, real-time meteorological data, and terrain data, calculate the impact deviation probability distribution results corresponding to each data through the formula. Then, determine the key impact factors based on the probability distribution results. Subsequently, establish a mapping relationship between environmental factors and survey progress. Finally, use a formula including the number of evaluation indicators, the weight coefficient of each indicator, the time decay coefficient, time, and the terrain correction coefficient to process the key impact factors according to the established mapping relationship, thereby obtaining the environmental impact result.

[0093] The geological survey method based on big data analysis provided by the embodiments of the present application can more accurately evaluate the impact of environmental factors on the geological survey progress through scientific and rigorous calculations and analyses, provide a reliable basis for subsequent resource allocation and progress management, thereby improving the efficiency and quality of geological survey work and reducing risks and losses caused by environmental factors.

[0094] Further, establishing a mapping relationship between environmental factors and survey progress includes:

[0095] (1) Obtain the environmental data of the target area, and the environmental data includes weather conditions, terrain conditions, and soil property data;

[0096] (2) Construct a regression analysis model between environmental data and survey efficiency;

[0097] (3) Based on the regression analysis model, calculate and confirm the environmental factors whose influence coefficients are greater than the preset threshold;

[0098] (4) Use the random forest algorithm to process the environmental factors to obtain the key impact factors;

[0099] (5) Generate a progress prediction result based on the key impact factors.

[0100] Specifically, to establish the mapping relationship between environmental factors and survey progress, environmental data of the target area needs to be collected first. The weather conditions include different weather types such as sunny days, heavy rains, and sandstorms; the terrain conditions include plains, canyons, steep slopes, etc., and the soil attribute data includes soil texture, pH value, etc. Then, using these environmental data as independent variables and survey efficiency as the dependent variable, a regression analysis model is constructed, which can quantitatively describe the relationship between environmental data and survey efficiency. Based on this regression analysis model, environmental factors with influence coefficients greater than the preset threshold are calculated and screened out. Subsequently, the random forest algorithm is used to process these environmental factors. The random forest consists of multiple decision trees, and the prediction accuracy and stability are improved through ensemble learning, and then the key influencing factors that play a key role in the survey progress are determined. Finally, the progress prediction result is generated based on the key influencing factors.

[0101] The beneficial effect of this embodiment is that it can clearly quantify the impact of the environment on the survey progress, help the survey team plan resources and arrange time in advance, adjust the plan in time when encountering a harsh environment, ensure the efficient and orderly progress of the survey work, and improve the overall work quality and efficiency.

[0102] Further, based on the abnormal nodes, a bottom-up dynamic programming algorithm is used to generate a progress adjustment plan, including:

[0103] (1) Using the following formula, a multi-stage decision-making model is constructed, and the transfer relationship between each survey link is defined:

[0104]

[0105] where s t represents the survey state at time t, a t represents the decision-making action at time t, R represents the immediate reward function, γ represents the discount factor, and V represents the value function;

[0106] (2) Using the following formula, the dynamic programming algorithm is used to process each survey link to obtain the corresponding optimal substructure value:

[0107]

[0108] where D k represents the decision value of the kth survey link, ω ij represents the transfer weight from the ith task to the jth task, and d ij represents the transfer distance between tasks;

[0109] (3) Based on the optimal substructure value, the execution order of the survey links is determined;

[0110] (4) Using the following formula, based on the similarity algorithm, the execution order is processed to generate the final progress adjustment plan:

[0111]

[0112] Among them, P represents the similarity between two survey links, and α k represents the weight coefficient of the k-th feature, sim represents the similarity function of a single feature, x represents the feature vector of the survey link, T(n) represents the optimal adjustment time for n tasks, and c k represents the completion time of the k-th task, and p i represents the probability of the i-th adjustment plan, and m represents the number of optional adjustment plans.

[0113] Specifically, first, a multi-stage decision-making model is constructed using formulas including survey status, decision-making actions, immediate benefit functions, discount factors, and value functions to define the transfer relationships of each survey link. Then, the dynamic programming algorithm is used to process each survey link through formulas involving decision values of survey links, task transfer weights, and transfer distances to obtain the optimal substructure values. Next, the execution order of the survey links is determined based on the optimal substructure values. Finally, with the help of the similarity algorithm, using formulas including feature weight coefficients, single-feature similarity functions, survey-link feature vectors, task optimal adjustment times, etc., the execution order is processed to generate the final schedule adjustment plan.

[0114] The beneficial effects of this embodiment are as follows: It can comprehensively and systematically consider the complex relationships and various factors of each link in geological surveys. Starting from the decisions of each survey link at the bottom layer, it gradually derives the overall optimal plan upwards, avoiding the problem of only focusing on the local while ignoring the global. Through the dynamic programming algorithm, it can find the optimal solution among numerous possible decision-making paths, greatly improving the scientificity and rationality of the schedule adjustment plan. At the same time, generating the final plan based on the similarity algorithm can fully consider the similarities and differences between different survey links, making the plan more flexible and adaptable. For example, when facing geological survey projects in different regions, even if the survey links are roughly the same, due to different feature vectors such as geological conditions and environmental factors, a highly targeted schedule adjustment plan can be generated, effectively ensuring the timely and efficient completion of geological survey projects and avoiding cost increases and resource waste caused by unreasonable schedules.

[0115] Furthermore, the environmental impact results are input into the trained resource scheduling optimization model for processing to obtain a resource allocation plan, and the resource allocation plan is used to allocate resources including manpower and equipment, including:

[0116] (1) Construct an objective function based on the environmental impact results, and through the particle swarm algorithm and population evolution iteration, obtain the fitness values corresponding to each initial plan;

[0117] (2) Judge the fitness value to obtain a judgment result, where the judgment result includes that the fitness value reaches the global optimum and the fitness value does not reach the global optimum;

[0118] (3) When the judgment result is that the fitness value does not reach the global optimum, the particle swarm algorithm and population evolution iteration are used to obtain the fitness value that reaches the global optimum;

[0119] (4) When the judgment result is that the fitness value reaches the global optimum, a resource allocation plan is generated according to the fitness value.

[0120] Specifically, first, a target function is constructed based on the environmental impact results, and the fitness values corresponding to each initial plan are calculated by means of the particle swarm algorithm and population evolution iteration. Subsequently, these fitness values are judged, and the judgment results include that the fitness value reaches the global optimum and does not reach the global optimum. If the judgment result is that it does not reach the global optimum, the particle swarm algorithm and population evolution iteration are used again until the fitness value that reaches the global optimum is obtained; if the judgment result is that it reaches the global optimum, a resource allocation plan for allocating resources such as manpower and equipment is generated according to the fitness value. Among them, the fitness value is calculated by the particle swarm algorithm and population evolution iteration after constructing the target function based on the environmental impact results, and is a quantitative index used to measure the degree of fit or superiority of each initial plan with the optimal solution in the resource allocation problem.

[0121] The beneficial effects of this embodiment are as follows: On the one hand, by constructing a target function and comprehensively considering the environmental impact results, it can ensure that the resource allocation plan fully adapts to the actual working environment, avoiding resource waste or task delays caused by environmental factors. On the other hand, the combination of the particle swarm algorithm and population evolution iteration gives full play to the advantages of swarm intelligence and evolutionary thinking, and can efficiently search for the global optimum solution in complex resource allocation problems, greatly improving the scientificity and rationality of the resource allocation plan. For example, in a large-scale geological exploration project, in the face of complex and changeable environmental impacts in different regions, this method can more accurately allocate the most suitable manpower and equipment resources for each region, ensure the smooth progress of the project, and at the same time reduce costs and improve efficiency.

[0122] Furthermore, the progress adjustment plan and the resource allocation plan are input into the trained multi-objective optimization model for processing to obtain a survey progress management strategy, including:

[0123] The progress adjustment plan and the resource allocation plan are input into the multi-objective optimization model and processed by the particle swarm algorithm and population evolution iteration to obtain a survey progress management strategy.

[0124] Specifically, the progress adjustment plan generated based on abnormal nodes and the resource allocation plan obtained from the resource scheduling optimization model according to environmental impacts are input into the trained multi-objective optimization model. With the help of the particle swarm algorithm and population evolution iterative processing, a survey progress management strategy is finally obtained. The beneficial effects of this embodiment are as follows: It can comprehensively consider multiple objectives such as progress and resources, balance the needs of all parties. For example, under complex geological conditions, it can not only reasonably allocate resources but also improve the overall efficiency and benefits of geological surveys.

[0125] Further, multi-source heterogeneous survey data is obtained, and the multi-source heterogeneous survey data is integrated and processed to obtain integrated survey data, including:

[0126] (1) Obtain multi-source heterogeneous survey data;

[0127] (2) Analyze the data formats of the multi-source heterogeneous survey data and extract the corresponding timestamps and accuracy values;

[0128] (3) Based on the data formats, timestamps, and accuracy values, adjust the multi-source heterogeneous survey data to obtain multi-source data with unified time and space;

[0129] (4) Based on the consistency of timestamps and accuracy values, process the multi-source data to obtain integrated survey data.

[0130] Specifically, first, through format parsing and conversion, the obstacles caused by data format differences can be eliminated, enabling data from different sources to be mutually compatible and collaboratively used. Second, using timestamps and accuracy values to perform time and space unification and consistency processing on the data ensures the reliability and coherence of the integrated data in terms of time and accuracy. For example, when analyzing geological changes in different periods, data with consistent accuracy can improve the accuracy of resource positioning. This embodiment greatly improves the data quality, provides a solid and reliable foundation for subsequent geological survey analysis, model construction, and decision-making based on these data, helps improve the efficiency and accuracy of geological survey work, and reduces incorrect judgments and resource waste caused by data problems.

[0131] Further, based on the optimal substructure values, the execution order of the survey links is determined, including:

[0132] (1) Obtain the optimal substructure values corresponding to each survey link and process each optimal substructure value to obtain the backtracking value of the global optimal path;

[0133] (2) Based on the backtracking value, determine the corresponding survey sequence;

[0134] (3) Based on the survey sequence, generate the execution order of the survey links.

[0135] Specifically, by starting from the locally optimal substructure values, gradually deriving the globally optimal path, and then determining the execution order of the survey links, it is possible to comprehensively and systematically consider the complex interrelationships and various constraints among the various links in the geological survey process. Compared with relying on experience or simple planning, it avoids the problem of low overall efficiency caused by only focusing on the local while ignoring the overall effect. For example, in a large-scale comprehensive geological survey project, involving multiple survey methods and a large number of work links, without a scientific method to determine the execution order, it is very likely to occur resource waste, project schedule delays and other situations. By using the method of determining the execution order based on the optimal structure value, it can ensure that the entire survey project is promoted in the most efficient and reasonable way on the premise of meeting various requirements, improve the success rate of the project, reduce costs, and at the same time improve the accuracy and reliability of the geological survey results.

[0136] Furthermore, based on the environmental impact results, a target function is constructed, and through the particle swarm algorithm and population evolution iteration, the fitness values corresponding to each initial plan are obtained, including:

[0137] Using the following formula, a target function is constructed based on the environmental impact results, and through the particle swarm algorithm and population evolution iteration, the fitness values corresponding to each initial plan are obtained:

[0138]

[0139] Among them, F(x) represents the particle velocity update equation, ω represents the inertia weight, v represents the current particle velocity, c1 and c2 represent the learning factors, r1 and r2 represent random numbers, p best represents the individual optimal position, g best represents the global optimal position, x represents the current position, P(i) represents the population selection probability calculation equation, β represents the selection pressure parameter, f represents the individual fitness value, N represents the population size, P represents the probability of being selected, fitness represents the fitness calculation function, α represents the weight coefficient, y represents the actual environmental impact value, represents the predicted environmental impact value, and n represents the number of environmental impact indicators.

[0140] Specifically, by combining the particle swarm algorithm and population evolution iteration, it is possible to make full use of swarm intelligence and evolutionary mechanisms in complex resource allocation problems, and efficiently search for better resource allocation solutions. The target function is constructed based on the environmental impact results, enabling the resource allocation plan to better adapt to the actual environmental conditions, improve resource utilization efficiency, reduce resource waste or task delays caused by environmental factors, thus ensuring the smooth progress of the geological survey work and enhancing the overall work efficiency.

[0141] Furthermore, based on the consistency of the timestamp and accuracy value, multi-source data is processed to obtain integrated survey data, including:

[0142] Use the following formula to obtain the data quality evaluation index based on multi-source data:

[0143]

[0144] where Q represents the data quality evaluation index, and x i represents the i-th measurement value, μ represents the mean of the measurement values, and σ i represents the standard deviation of the i-th measurement value, and n1 represents the sample size.

[0145] Specifically, by calculating the data quality evaluation index, the quality of multi-source data can be objectively and quantitatively evaluated. Before integrating the survey data, understanding the data quality can help us screen out reliable data, eliminate or correct data with poor quality, thereby improving the quality of the integrated survey data. If it is found that the data collected by some devices has too large a dispersion, these data can be further analyzed or recollected to ensure that the finally obtained integrated survey data is accurate and reliable, providing a solid data foundation for subsequent geological analysis and decision-making, and avoiding wrong judgments and decision-making mistakes caused by data quality problems.

[0146] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0147] The above-described embodiments only represent several implementation manners of the embodiments of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. A geological survey method based on big data analysis, characterized in that: The method comprises: Acquiring multi-source heterogeneous survey data, and integrating and processing the multi-source heterogeneous survey data to obtain integrated survey data; Based on the integrated survey data, a time series progress analysis model is constructed, a sliding window method is used to extract the progress feature vectors of each survey link, and each progress feature vector is processed to obtain abnormal nodes with delayed progress; Based on the abnormal nodes, a schedule adjustment plan is generated using a bottom-up dynamic programming algorithm; Obtaining preset environmental impact parameters, and processing the environmental impact parameters using a Bayesian network model to obtain environmental impact results, wherein the environmental impact parameters include environmental impact factor data, real-time meteorological data, and terrain data; Inputting the environmental impact results into a trained resource scheduling optimization model for processing to obtain a resource allocation plan, wherein the resource allocation plan is used to allocate resources including manpower and equipment; The schedule adjustment plan and the resource allocation plan are input into a trained multi-objective optimization model for processing to obtain a survey schedule management strategy.

2. The geological survey method based on big data analysis according to claim 1, characterized in that: The environmental impact parameters are processed using the Bayesian network model to obtain environmental impact results, wherein the environmental impact parameters include environmental impact factor data, real-time meteorological data and terrain data, including: The following formula is used to calculate the probability distribution result of the impact deviation corresponding to each data based on the preset environmental impact factor data, the real-time meteorological data and the terrain data: Among them, P(E i ) represents the probability distribution of environmental factors, n represents the number of samples, x i represents the measured value of the environmental factor, μ represents the mean value of the environmental factor, and σ represents the standard deviation; Based on the probability distribution results, key influencing factors are obtained; Establish a mapping relationship between environmental factors and survey progress; The key influencing factors are processed based on the mapping relationship using the following formula to obtain the environmental impact result: Among them, R represents the environmental impact result, l represents the number of evaluation indicators, γ i represents the weight coefficient of the i-th indicator, λ i represents the time attenuation coefficient, t represents time, h i Represents the terrain correction factor.

3. The geological survey method based on big data analysis according to claim 2, characterized in that: The establishing of a mapping relationship between environmental factors and survey progress includes: Acquiring environmental data of a target area, wherein the environmental data includes weather conditions, terrain conditions, and soil property data; Constructing a regression analysis model of the environmental data and survey efficiency; Based on the regression analysis model, calculate and confirm environmental factors whose impact coefficients are greater than a preset threshold; The environmental factors are processed using a random forest algorithm to obtain key influencing factors; Based on the key influencing factors, a progress forecast result is generated.

4. The geological survey method based on big data analysis according to claim 1, characterized in that: The process of generating a schedule adjustment plan based on the abnormal node using a bottom-up dynamic programming algorithm includes: Use the following formula to build a multi-stage decision model and define the transfer relationship between the various survey links: Among them, s t represents the survey status at time t, a t represents the decision action at time t, R represents the immediate benefit function, γ represents the discount factor, and V represents the value function; Use the following formula to process each survey link using the dynamic programming algorithm to obtain the corresponding optimal substructure value: Among them, D k represents the decision value of the kth survey link, ω ij represents the transfer weight from the i-th task to the j-th task, d ij represents the transfer distance between tasks; Based on the optimal substructure value, determining the execution order of the surveying links; The execution order is processed using the following formula based on a similarity algorithm to generate the final schedule adjustment plan: Among them, P represents the similarity of the two survey links, α k represents the weight coefficient of the kth feature, sim represents the similarity function of a single feature, x represents the feature vector of the survey phase, T(n) represents the optimal adjustment time of n tasks, c k represents the completion time of the kth task, p i represents the probability of the i-th adjustment plan, and m represents the number of optional adjustment plans.

5. The geological survey method based on big data analysis according to claim 1, characterized in that: The environmental impact results are input into a trained resource scheduling optimization model for processing to obtain a resource allocation plan, which is used to allocate resources including manpower and equipment, including: Based on the environmental impact results, an objective function is constructed, and the fitness value corresponding to each initial solution is obtained through particle swarm algorithm and population evolution iteration; The fitness value is judged to obtain a judgment result, wherein the judgment result includes whether the fitness value reaches the global optimum or not; When the judgment result is that the fitness value has not reached the global optimum, the fitness value reaching the global optimum is obtained through the particle swarm algorithm and the population evolution iteration; When the judgment result is that the fitness value reaches the global optimum, the resource configuration plan is generated according to the fitness value.

6. The geological survey method based on big data analysis according to claim 1, characterized in that: The step of inputting the schedule adjustment plan and the resource allocation plan into the trained multi-objective optimization model for processing to obtain a survey schedule management strategy includes: The progress adjustment plan and the resource allocation plan are input into the multi-objective optimization model, and processed through particle swarm algorithm and population evolution iteration to obtain the survey progress management strategy.

7. The geological survey method based on big data analysis according to claim 1, characterized in that: The acquiring of multi-source heterogeneous survey data and integrating the multi-source heterogeneous survey data to obtain integrated survey data includes: Acquiring the multi-source heterogeneous survey data; Parsing the data format of the multi-source heterogeneous survey data and extracting the corresponding timestamp and precision value; Based on the data format, the timestamp and the precision value, the multi-source heterogeneous survey data is adjusted to obtain multi-source data that is unified in time and space; Based on the consistency of the timestamp and the precision value, the multi-source data is processed to obtain the integrated survey data.

8. The geological survey method based on big data analysis according to claim 4, characterized in that: Determining the execution order of the survey links based on the optimal substructure value includes: Obtaining the optimal substructure value corresponding to each survey link, and processing each optimal substructure value to obtain a backtracking value of a global optimal path; Based on the traceback value, determining a corresponding survey sequence; Based on the survey sequence, the execution order of the survey links is generated.

9. The geological survey method based on big data analysis according to claim 5, characterized in that: The objective function is constructed based on the environmental impact result, and the fitness value corresponding to each initial solution is obtained through particle swarm algorithm and population evolution iteration, including: The objective function is constructed based on the environmental impact results using the following formula, and the fitness values ​​corresponding to each initial solution are obtained through particle swarm optimization and population evolution iteration: Among them, F(x) represents the particle velocity update equation, ω represents the inertia weight, v represents the current velocity of the particle, c1 and c2 represent learning factors, r1 and r2 represent random numbers, and p best represents the optimal position of an individual, g best represents the global optimal position, x represents the current position, P(i) represents the population selection probability calculation equation, β represents the selection pressure parameter, f represents the individual fitness value, N represents the population size, P represents the probability of being selected, fitness represents the fitness calculation function, α represents the weight coefficient, and y represents the actual environmental impact value. It represents the predicted environmental impact value, and n represents the number of environmental impact indicators.

10. The geological survey method based on big data analysis according to claim 7, characterized in that: The step of processing the multi-source data based on the consistency of the timestamp and the precision value to obtain the integrated survey data includes: The following formula is used to obtain the data quality assessment index based on the multi-source data: Among them, Q represents the data quality assessment index, x i represents the i-th measurement value, μ represents the mean of the measurement value, σ i represents the standard deviation of the ith measurement value, and n1 represents the sample size.

Images