Genetic algorithm-based terrain processing method and device, equipment and medium

Abstract

The invention discloses a terrain processing method and device based on a genetic algorithm, equipment and a medium. The method comprises the following steps: initializing a first population; taking the first population as a target population; performing inversion operation and fitness evaluation of the target area based on the target population to obtain a root-mean-square value of each individual in the target population; when the root-mean-square value meets a preset threshold condition, outputting the corresponding individual as an adaptive individual; otherwise, performing genetic manipulation on the target population to obtain a second population; taking the second population as a target population; returning to the step of performing the inversion operation and fitness evaluation of the target area based on the target population until the adaptive individual is output, or reaching the maximum number of iterations of the population; and taking topographic data correspondingly obtained by the inversion parameters corresponding to the adaptive individuals in the inversion operation as a topographic processing result of the target area. The method can efficiently and accurately carry out terrain processing, and can be widely applied to the technical field of data processing.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a terrain processing method, device, equipment and medium based on genetic algorithm. Background Art

[0002] With the deepening of earth science research and the development of resource exploration technology, the accurate inversion of the crust structure, especially the depth of the Moho discontinuity, has become a core demand in the fields of geology and geophysics. As the interface between the crust and the mantle, the accurate determination of the depth of the Moho discontinuity is of great significance for understanding the internal structure of the earth, the mechanism of plate movement, and the exploration of mineral resources.

[0003] In interface inversion technology, the Parker-Oldenburg method is favored for its efficient computing performance based on fast Fourier transform (FFT) and has become one of the classic methods in this field. This method uses gravity anomaly data observed on the ground and combines it with the assumption of underground medium density distribution to invert the shape and depth of the interface. However, when using the Parker-Oldenburg method, the existing technical solutions generally treat the inversion parameters (such as density difference and depth) as fixed constants. This simplified treatment ignores the complexity and variability of underground geological conditions.

[0004] The relationship between the density difference and depth of underground media is not a simple linear relationship, but is affected by a variety of geological factors such as lithology, structure, magmatic activity, etc. Therefore, fixing the inversion parameters will cause the inversion results to fail to accurately reflect the actual underground conditions, thereby reducing the accuracy and reliability of the inversion results.

[0005] In addition, existing solutions also face many challenges in terms of computing efficiency and automation. As the amount of exploration data continues to increase, the overall computing workload has increased significantly, and the demand for computing resources has also increased accordingly. At the same time, due to the lack of a unified execution framework and efficient algorithm optimization, existing solutions are often inefficient when processing large-scale data sets and are difficult to meet the needs of practical applications. Summary of the invention

[0006] The present invention aims to solve the problem of related technical limitations at least to a certain extent. To this end, the present invention proposes a terrain processing method, device, equipment and medium based on genetic algorithm, which can perform terrain processing efficiently and accurately.

[0007] On the one hand, an embodiment of the present invention provides a terrain processing method based on a genetic algorithm, comprising:

[0008] Initialize the inversion parameters of the target number, obtain the first population based on the initialization of the inversion parameters; use the first population as the target population;

[0009] Among them, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth;

[0010] Based on the inversion parameters corresponding to each individual in the target population, the inversion operation and fitness evaluation of the target area are performed to obtain the root mean square value corresponding to each individual in the target population;

[0011] The root mean square value represents the adaptability of the inversion parameters corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the terrain data of the target area;

[0012] When there is a RMS value in the target population that meets the preset threshold condition, the individual corresponding to the RMS value is output as the adapted individual; otherwise,

[0013] Perform genetic operations on the target population based on a preset genetic algorithm to obtain a second population; use the second population as the target population, and increase the number of population iterations by 1;

[0014] Return to execute the steps of performing inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, until the output is a suitable individual, or the number of population iterations reaches the preset maximum number of population iterations;

[0015] The terrain data corresponding to the inversion parameters of the adapted individuals in the inversion operation are used as the terrain processing results of the target area.

[0016] Optionally, initializing the target number of inversion parameters comprises the following steps:

[0017] In response to a first input instruction of the target object, a center point is obtained, a first value is assigned to the center point by a first random number, and a second number of inversion parameters are randomly generated near the center point; and a second value is assigned to the center point by multiplying the first random number by a difference between the first interval and the interval range of the first interval, and a third number of inversion parameters are randomly generated near the center point;

[0018] Wherein, the first random number represents a random number of a normal distribution with a mean of 0 and a standard deviation of 1; the first random number is updated before each generation of the inversion parameter; the first interval represents the value range of the inversion parameter; the sum of the second number and the third number is equal to the target number;

[0019] Alternatively, based on a second random number, a target number of inversion parameters are randomly generated within a first interval; the second random number represents a random value within the second interval;

[0020] Alternatively, a target number of inversion parameters is obtained in response to a second input instruction of the target object.

[0021] Optionally, performing an inversion operation and fitness evaluation on the target area based on the inversion parameters corresponding to each individual in the target population to obtain a root mean square value corresponding to each individual in the target population includes the following steps:

[0022] Based on the inversion parameters corresponding to the individuals in the target population, the topographic data of the target area corresponding to each individual is obtained by processing with the Parker-Oldenburg inversion method;

[0023] Obtain the seismic depth matrix of the target area;

[0024] Traverse the first data values ​​of all location points of the terrain data corresponding to the individual, and collect the second data values ​​of the corresponding location points from the seismic depth matrix; obtain the third data value of the corresponding location point based on the square of the difference between the first data value and the second data value; perform square root processing based on the average value of the third data values ​​of all location points to obtain the root mean square value corresponding to the corresponding individual.

[0025] Optionally, performing a genetic operation on the target population based on a preset genetic algorithm to obtain a second population comprises at least one of the following steps:

[0026] Based on the excellence of each individual in the target population, a target number of individuals are selected from the target population by combining a first probability rule with a first probability threshold to obtain a second population; wherein the excellence is determined based on the inverse of the root mean square value;

[0027] A second population is obtained by performing a crossover operation on a random pair of individuals in the target population using the second probability rule;

[0028] The second population is obtained by performing mutation operation on random individuals in the target population through the second probability rule.

[0029] Optionally, based on the excellence of each individual in the target population, a target number of individuals are selected from the target population by combining a first probability rule with a first probability threshold to obtain a second population, comprising the following steps:

[0030] The reciprocal of the RMS value is taken as the excellence level of each individual in the target population;

[0031] Normalize all the excellence levels to obtain the normalized excellence value corresponding to each individual;

[0032] Initialize the identification number to 0 and initialize the excellent array to empty;

[0033] The first individual of the target population is taken as the data individual, and the excellent value corresponding to the data individual is taken as the output value;

[0034] The output value is used as the array element of the position corresponding to the identification number in the excellent array; the identification number is increased by 1; the next individual after the data individual in the target population is used as the data individual;

[0035] Accumulate the excellent values ​​corresponding to the data individuals to the output value, return to the step of executing the output value as the array element of the position corresponding to the identification number in the excellent array, until the identification number is equal to the total number of all individuals in the target population, and output the excellent array;

[0036] generating a random value within a third interval as a first probability threshold;

[0037] Sequentially traverse the excellent array and obtain the first array element that is greater than the first probability threshold as the target element;

[0038] According to the element number corresponding to the target element, the corresponding individuals are selected from the target population and sorted into the second population;

[0039] Return to the step of generating a random value in the third interval as the first probability threshold until a target number of individuals are selected from the target population to obtain a second population.

[0040] Optionally, performing a crossover operation on a random pair of individuals in the target population by a second probability rule to obtain a second population includes the following steps:

[0041] Based on all RMS values ​​corresponding to the target population, the minimum RMS value is obtained as the target RMS value;

[0042] The ratio of the target RMS value to the preset iterative optimal threshold is rounded up, and the result of the rounding up operation multiplied by the target number is used as the number of crossovers;

[0043] Randomly obtain two different individuals from the target population as crossover individuals; the crossover individuals include the first individual and the second individual;

[0044] generating a random value within a fourth interval as a second probability threshold;

[0045] When the second probability threshold is less than the preset crossover operator probability, weighted crossover is performed on the inversion parameters corresponding to the first individual and the second individual in the crossover individual based on the preset crossover operator ratio; otherwise, weighted crossover is not performed on the crossover individual;

[0046] Add 1 to the first cycle number, and return to the step of randomly obtaining two different individuals from the target population as crossover individuals until the first cycle number is equal to the crossover number, and the second population is obtained;

[0047] The first cycle number is initialized to 0.

[0048] Optionally, performing a mutation operation on random individuals in the target population by using a second probability rule to obtain a second population includes the following steps:

[0049] Randomly obtain an individual from the target population as a mutant individual;

[0050] generating a random value within a fifth interval as a third probability threshold;

[0051] When the third probability threshold is less than the preset mutation operator probability, the inversion parameters corresponding to the mutated individuals are assigned based on the population iteration number, the population maximum iteration number and the value range of the inversion parameters; otherwise, the inversion parameters of the mutated individuals are not assigned;

[0052] The expression for assignment processing is:

[0053] A1=A1+(max-mix)*normal(0,1)*(1-random() (1－curIndex / maxGenCnt) );

[0054] In the formula, A1 represents the inversion parameter of the variant individual; max represents the maximum value of the inversion parameter interval; min represents the minimum value of the inversion parameter interval; normal(0,1) represents a random value with a normal distribution with a mean of 0 and a standard deviation of 1; random() represents a random value generated in the sixth interval; curIndex represents the number of population iterations; maxGenCnt represents the maximum number of population iterations;

[0055] Add 1 to the second cycle number, and return to the step of randomly obtaining an individual from the target population as a mutant individual until the second cycle number equals the target number, and the second population is obtained;

[0056] The second loop number is initialized to 0.

[0057] Optionally, the method further comprises the following steps:

[0058] When the number of population iterations is equal to the maximum number of population iterations, the adapted individuals are output according to the target population obtained in the last cycle.

[0059] On the other hand, an embodiment of the present invention provides a terrain processing device based on a genetic algorithm, comprising:

[0060] The first module is used to initialize the inversion parameters of the target quantity, obtain the first population based on the inversion parameter initialization; and use the first population as the target population;

[0061] Among them, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth;

[0062] The second module is used to perform inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, and obtain the root mean square value corresponding to each individual in the target population;

[0063] The root mean square value represents the adaptability of the inversion parameters corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the terrain data of the target area;

[0064] The third module is used to output the individual corresponding to the root mean square value as the adaptation individual when the root mean square value in the target population meets the preset threshold condition; otherwise,

[0065] The fourth module is used to perform genetic operations on the target population based on a preset genetic algorithm to obtain a second population; the second population is used as the target population, and the number of population iterations is increased by 1;

[0066] The fifth module is used to return to the step of performing the inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, until the adapted individual is output, or the number of population iterations reaches a preset maximum number of population iterations;

[0067] The sixth module is used to use the terrain data obtained by the inversion operation corresponding to the inversion parameters corresponding to the adaptation individual as the terrain processing result of the target area.

[0068] Optionally, the device further comprises:

[0069] The seventh module is used to output the adapted individuals according to the target population obtained in the last cycle when the number of population iterations is equal to the maximum number of population iterations.

[0070] On the other hand, an embodiment of the present invention provides an electronic device, including: a processor and a memory; the memory is used to store programs; the processor executes the program to implement the above-mentioned terrain processing method based on genetic algorithm.

[0071] On the other hand, an embodiment of the present invention provides a computer storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the terrain processing method based on the genetic algorithm.

[0072] The embodiment of the present invention initializes the inversion parameters of the target number, and obtains the first population based on the inversion parameter initialization; the first population is used as the target population; wherein the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seabed depth; based on the inversion parameters corresponding to each individual in the target population, an inversion operation and fitness evaluation of the target area are performed to obtain the root mean square value corresponding to each individual in the target population; wherein the root mean square value represents the adaptability of the inversion parameters corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the terrain data of the target area; when there is a root mean square in the target population If the root value meets the preset threshold condition, the individual corresponding to the corresponding RMS value is output as the adapted individual; otherwise, the target population is genetically operated based on the preset genetic algorithm to obtain the second population; the second population is used as the target population, and the number of population iterations is increased by 1; the steps of performing the inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population are returned to until the adapted individual is output, or the number of population iterations reaches the preset maximum number of population iterations; the terrain data corresponding to the inversion parameters corresponding to the adapted individual in the inversion operation is used as the terrain processing result of the target area. The embodiment of the present invention realizes the dynamic adjustment and optimization of the inversion parameters by introducing the genetic algorithm, thereby more accurately reflecting the complexity and variability of the underground geological conditions. In addition, by optimizing the process architecture and improving the degree of automation, the present invention also significantly improves the computing efficiency and practicality, providing strong support for terrain structure research and resource exploration. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0074] Figure 1 A schematic diagram of an implementation environment for performing terrain processing based on a genetic algorithm provided by an embodiment of the present invention;

[0075] Figure 2 A schematic diagram of a flow chart of a terrain processing method based on a genetic algorithm provided by an embodiment of the present invention;

[0076] Figure 3 A schematic diagram of the expansion process of the inversion operation and fitness evaluation provided by the embodiment of the present invention;

[0077] Figure 4 A schematic diagram of the development process of genetic operations provided by an embodiment of the present invention;

[0078] Figure 5 A schematic diagram of the expansion flow of the selection operation provided by an embodiment of the present invention;

[0079] Figure 6 A schematic diagram of the expansion process of the crossover operation provided by an embodiment of the present invention;

[0080] Figure 7 A schematic diagram of the expansion process of the mutation operation provided by an embodiment of the present invention;

[0081] Figure 8 A schematic diagram of the overall process of a terrain processing method based on a genetic algorithm provided in an embodiment of the present invention;

[0082] Fig. 9 A schematic diagram of the structure of a terrain processing device based on a genetic algorithm provided by an embodiment of the present invention;

[0083] Fig.10 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0084] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0085] It should be noted that, although the functional modules are divided in the system schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first / S100", "second / S200", etc. in the specification, claims and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0086] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0087] It is understandable that the terrain processing method based on genetic algorithm provided in the embodiment of the present invention can be applied to any computer device with data processing and computing capabilities, and this computer device can be various terminals or servers. When the computer device in the embodiment is a server, the server is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Optionally, the terminal is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this.

[0088] To facilitate the understanding of the technical solution of the present invention, the specific terms of the technical features that may appear in the embodiments of the present invention are first explained:

[0089] Seabed topography: the solid earth surface under water. Due to the cover of seawater, the ups and downs of the seabed topography are difficult to observe directly. The early plumb-bob sounding method was time-consuming and of low precision. Since the 1920s, ships have used echo sounders during navigation, which can quickly measure the depth of the seabed and reveal the truth of the seabed topography in combination with precise positioning. In the present invention, the seabed topography refers to the depth of seawater at various latitudes and longitudes of the ocean.

[0090] Heuristic search is also called informed search. It uses the heuristic information of the problem to guide the search, so as to reduce the search scope and reduce the complexity of the problem. This search process using heuristic information is called heuristic search. For example, genetic algorithm is a classic heuristic search algorithm.

[0091] Genetic Algorithm (GA) is a computational model that simulates the biological evolution process, using mechanisms such as natural selection and gene recombination to search and optimize complex problems. It is a heuristic search algorithm, and its basic idea is to create a group of artificial individuals for optimization and solution by simulating the mechanisms of natural inheritance and evolution. In a genetic algorithm, each individual represents a potential solution, and each individual has its own fitness evaluation criteria. By selecting, crossing over, and mutating these individuals, the adaptability of the current population is gradually improved and the potential solution space is explored in order to find the optimal solution. Genetic algorithms have the characteristics of a wide range of applications, strong global search, and easy parallel computing. They are suitable for optimizing and solving various types of problems, including combinatorial optimization, continuous optimization, and multi-objective optimization. Compared with other algorithms, genetic algorithms can efficiently search for the global optimal solution and are widely used in various fields, including but not limited to engineering optimization design, artificial intelligence tasks, financial analysis and prediction, etc.

[0092] Decimal coding: It is a counting method that counts by one when the number reaches ten. The first digit weight is 10^0, the second digit is 10^1, and the Nth digit is 10^(N-1). The value of the number is equal to the sum of (value × digit weight) of each digit. The numbers used in daily life are all coded in decimal.

[0093] The Parker-Oldenburg inversion method is a frequency domain density interface iterative inversion method proposed by Oldenburg based on the Parker formula. Since this method has the advantage of fast calculation, its application has been rapidly developed. However, there is a downward extension factor in the formula, which causes the formula to have an oscillation phenomenon of high-frequency signals during the inversion iteration process, and as the inversion depth increases, this phenomenon becomes more obvious, which seriously affects the convergence of the inversion result. To solve this problem, a low-pass filter is usually applied in the inversion process to eliminate the high-frequency oscillation phenomenon. The Parker-Oldenburg inversion method is one of the important methods for calculating the seabed topography, and is also the starting point of the core of the present invention. At present, the input parameters of the method (filter-related parameters, density difference, depth), etc. are designed according to constants based on experience, but in fact, there is still room for tuning and mining these parameters.

[0094] RMS: The abbreviation of Root Mean Square. The result of taking the square root of the sum of the squares of N items divided by N is the root mean square result.

[0095] like Figure 1 FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the invention. Figure 1The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected to a network wirelessly or wired to complete data transmission and exchange.

[0096] Server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0097] In addition, the server 101 can also be a node server in the blockchain network. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm.

[0098] The terminal 102 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal 102 and the server 101 may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present invention.

[0099] Based on the example Figure 1 In the implementation environment shown, an embodiment of the present invention provides a terrain processing method based on a genetic algorithm. The following is explained using the example of the terrain processing method based on a genetic algorithm being applied to a server 101. It can be understood that the terrain processing method based on a genetic algorithm can also be applied to a terminal 102.

[0100] Reference Figure 2 , Figure 2 The flowchart of the terrain processing method based on genetic algorithm applied to the server provided in the embodiment of the present invention, the execution subject of the terrain processing method based on genetic algorithm can be any of the aforementioned computer devices (including servers or terminals). Figure 2 , the method comprises the following steps:

[0101] S100, initializing the inversion parameters of the target number, obtaining a first population based on the initialization of the inversion parameters; and using the first population as the target population;

[0102] Among them, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth;

[0103] It should be noted that the present invention may also include a data preparation step to initialize various variables, parameters, and thresholds that may be applied. Specifically, the parameters that the present invention may involve are explained as follows:

[0104] The parameters mainly include two parts: the parameters of the improved genetic algorithm based on real number coding and the parameters of the Parker-Oldenburg inversion method, etc.

[0105] There are a total of 2 variables searched by the improved genetic algorithm based on real number coding, which are contrast (physical meaning: density difference, unit g / cm 3 ), and depth (physical meaning: depth, unit km). Then, the parameters of the improved genetic algorithm are respectively:

[0106] The maximum density difference maxContrast: the maximum value of contrast, which is determined by the average value of the target search area, generally obtained according to experience, and satisfies minContrast < maxContrast. In most cases, it is sufficient to satisfy maxContrast - minContrast >= 0.4, and it can be set according to the actual situation.

[0107] The minimum density difference minContrast: the minimum value of contrast, which is determined by the average value of the target search area, generally obtained according to experience, and satisfies minContrast < maxContrast. In most cases, it is sufficient to satisfy maxContrast - minContrast >= 0.4, and it can be set according to the actual situation.

[0108] The maximum depth maxDepth: the maximum value of depth, which is determined by the average value of the target search area, generally obtained according to experience, and satisfies maxDepth < minDepth. In most cases, it is sufficient to satisfy maxDepth - minDepth >= 10, and it can be set according to the actual situation.

[0109] The minimum depth minDepth: the minimum value of depth, which is determined by the average value of the target search area, generally obtained according to experience, and satisfies maxDepth < minDepth. In most cases, it is sufficient to satisfy maxDepth - minDepth >= 10, and it can be set according to the actual situation.

[0110] The number of individuals in the population sizePop, the individual scale of the genetic algorithm, is taken according to requirements, and the default is 100.

[0111] The maximum number of population iterations maxGenCnt, the number of population reproduction times of the genetic algorithm, that is, the number of iterations of the genetic algorithm, is determined according to requirements and the default value is 200.

[0112] Crossover operator probability crossoverRate, a parameter used by the crossover operator, has a value range of [0,1] and is determined based on demand. The default value is 0.9.

[0113] Mutation operator probability mutationRate, a parameter used by the mutation operator, has a value range of [0,1] and is determined based on demand. The default value is 0.1.

[0114] Crossover operator ratio crossoverAlpha, a parameter used by the crossover operator, with a value range of [0,1]. The value is determined according to requirements and the default value is 0.5.

[0115] Iterate the optimal threshold minIterDelt. If an individual less than or equal to minIterDelt is found, it means that the optimal solution is found and the entire search process ends directly.

[0116] The main parameters of the Parker-Oldenburg inversion method are as follows:

[0117] Density difference contrast, unit: g / cm 3 , which is one of the variables searched by means of the improved genetic algorithm, that is, the variable is obtained by searching the improved genetic algorithm and is one of the two search directions mentioned in the present invention.

[0118] Depth depth: unit is km, this is one of the variables searched by means of the improved genetic algorithm, that is, the variable is obtained by searching the improved genetic algorithm, and is one of the two search directions mentioned in the present invention.

[0119] Gravity outlier matrix GravityMat: It is a matrix composed of (longitude, latitude, gravity outlier) triplets, generally data observed by seismic measurement, satellite measurement or other means, and is a constant in the present invention.

[0120] Seismic depth matrix SemisMat: It is a matrix composed of (longitude, latitude, depth) triplets, generally data observed by seismic measurement, satellite measurement or other means, and is a constant in the present invention.

[0121] The iteration threshold criterio, i.e., the iteration rms threshold of the Parker-Oldenburg inversion method, is generally a constant, and the default value is 0.1, which is a constant in the present invention.

[0122] Frequency threshold WH: low-pass filter threshold, the initial value is obtained based on experience, the default value is 0.1, and it is a constant in the present invention.

[0123] Frequency threshold SH: high-pass filter threshold, the initial value is obtained based on experience, the default value is 0.12, and it is a constant in the present invention.

[0124] Window function threshold truncation: It is a signal sampling weighting function in signal processing. The initial value is obtained based on experience, and the default value is 0.1. It is a constant in the present invention.

[0125] It should be noted that, in some embodiments, initializing the target number of inversion parameters may include the following steps:

[0126] In response to a first input instruction of the target object, a center point is obtained, a first value is assigned to the center point by a first random number, and a second number of inversion parameters are randomly generated near the center point; and a second value is assigned to the center point by multiplying the first random number by a difference between the first interval and the interval range of the first interval, and a third number of inversion parameters are randomly generated near the center point;

[0127] Wherein, the first random number represents a random number of a normal distribution with a mean of 0 and a standard deviation of 1; the first random number is updated before each generation of the inversion parameter; the first interval represents the value range of the inversion parameter; the sum of the second number and the third number is equal to the target number;

[0128] It should be noted that the first interval includes a density difference interval (determined based on a preset maximum density difference value and a preset minimum density difference value) and a seafloor average depth interval (determined based on a preset maximum depth value and a preset minimum depth value);

[0129] For example, in some specific implementations, based on the center point generation, it can be implemented as follows:

[0130] If the center point is given, the contrast and depth are generated as follows:

[0131] The generation formula of the first individual gene is as follows, which is mainly generated near the center point.

[0132] contrast=initContrast+normal(0,1)*0.01;

[0133] depth=initDepth+normal(0,1)*0.1;

[0134] Other ways in which individual genes are generated are:

[0135] contrast=initContrast+(maxContrast-minContrast)*normal(0,1);

[0136] depth=initDepth+(maxDepth-minDepth)*normal(0,1);

[0137] Alternatively, based on a second random number, a target number of inversion parameters are randomly generated within a first interval; the second random number represents a random value within the second interval;

[0138] For example, in some specific implementations, random generation can be implemented as follows:

[0139] If no center point is given, it is randomly generated within the search interval. The generation method of contrast and depth is as follows:

[0140] contrast=minContrast+(maxContrast-minContrast)*random();

[0141] depth=minDepth+(maxDepth-minDepth)*random();

[0142] Alternatively, in response to a second input instruction of the target object, an inversion parameter of the target quantity is obtained.

[0143] Exemplarily, in some specific implementations, it can also be specified manually, in which case the initialization genes of all individuals need to be specified manually.

[0144] Among them, initContrast and initDepth represent the center point (corresponding to the center point of density difference and the center point of average depth of the seabed); random() represents the random value generated in [0,1]; normal(0,1) represents the normal distribution with mean 0 and standard deviation 1, and the obtained random number satisfies the normal distribution, thereby ensuring the fairness of gene generation. The present invention adopts decimal real number coding, and the entire calculation is based on decimal, so no additional encoding and decoding operations are required.

[0145] Specifically, the implementation of population initialization includes: the population is composed of a large number of individuals, each individual has two genes, representing the density difference contrast and the average depth of the seabed depth. The present invention uses decimal real numbers to represent gene encoding. The core of population initialization is to generate sizePop individuals and initialize the corresponding genes for each individual. The generation of the population can be achieved through the above strategy.

[0146] S200, performing inversion operation and fitness evaluation on the target area based on the inversion parameters corresponding to each individual in the target population, and obtaining a root mean square value corresponding to each individual in the target population;

[0147] The root mean square value represents the adaptability of the inversion parameters corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the terrain data of the target area;

[0148] It should be noted that in some embodiments, Figure 3 As shown, step S200 may include the following steps: S201, based on the inversion parameters corresponding to the individuals in the target population, obtain the terrain data of the target area corresponding to each individual through the Parker-Oldenburg inversion method; S202, obtain the seismic depth matrix of the target area; S203, traverse the first data values ​​of all position points of the terrain data corresponding to the individual, and collect the second data values ​​of the corresponding position points from the seismic depth matrix; S204, based on the square of the difference between the first data value and the second data value, obtain the third data value of the corresponding position point; S205, perform square root processing based on the average value of the third data values ​​of all position points to obtain the root mean square value corresponding to the corresponding individual.

[0149] Exemplarily, in some specific implementations, first, the seabed topography is calculated based on the current population: for the current population, the gene contrast and depth of each individual are traversed one by one, and then the corresponding seabed topography data is calculated based on the Parker-Oldenburg inversion method. Because it is a decimal real number encoding, the gene does not require additional encoding and decoding operations before being applied to the Parker-Oldenburg inversion method. Then, the fitness function value of each individual is evaluated: at this time, each individual in the population has calculated the corresponding seabed topography data, and then combined with the seismic depth matrix SemisMat, the corresponding RMS is calculated. Each individual will calculate its own RMS, and finally an array of sizePop is obtained.

[0150] S300, when there is a root mean square value in the target population that meets the preset threshold condition, output the individual corresponding to the corresponding root mean square value as the adapted individual, and jump to step S600; otherwise, execute the subsequent steps;

[0151] When there are multiple root mean square values ​​that meet the preset threshold conditions at the same time, all root mean square values ​​that meet the preset threshold conditions are sorted from small to large, and the individual corresponding to the first root mean square value after sorting is output as the adapted individual;

[0152] Exemplarily, in some specific implementations, by judging whether there is an individual whose value is less than a threshold, all individuals in the population are traversed to judge whether there is an RMS less than the iterative optimal threshold minIterDelt. If so, go to step S600, indicating that the optimal individual (optimal solution) has been found, otherwise execute the next step.

[0153] S400, performing genetic operations on the target population based on a preset genetic algorithm to obtain a second population; taking the second population as the target population, and increasing the number of population iterations by 1;

[0154] It should be noted that in some embodiments, Figure 4 As shown, a genetic operation is performed on a target population based on a preset genetic algorithm to obtain a second population, including at least one of the following steps: S41, based on the excellence of each individual in the target population, a target number of individuals are selected from the target population through a first probability rule combined with a first probability threshold to obtain a second population; wherein the excellence is determined based on the inverse of the root mean square value; S42, a crossover operation is performed on a random pair of individuals in the target population through a second probability rule to obtain a second population; S43, a mutation operation is performed on a random individual in the target population through the second probability rule to obtain a second population.

[0155] The target population executed by any of the above genetic operations may be the result of the second population obtained by other genetic operations.

[0156] Among them, in some embodiments, such as Figure 5 As shown, based on the excellence of each individual in the target population, selecting a target number of individuals from the target population by combining a first probability rule with a first probability threshold to obtain a second population may include the following steps:

[0157] S411, taking the reciprocal of the root mean square value as the excellence level corresponding to each individual in the target population;

[0158] S412, normalizing all excellence levels to obtain a normalized excellence value corresponding to each individual;

[0159] S413, initializing the identification number to 0, and initializing the outstanding array to empty;

[0160] S414, taking the first individual of the target population as a data individual, and taking the excellent value corresponding to the data individual as an output value;

[0161] S415, using the output value as the array element of the position corresponding to the identification number in the excellent array; increasing the identification number by 1; using the next individual of the data individual in the target population as the data individual;

[0162] S416, adding the excellent value corresponding to the data individual to the output value, returning to the step of using the output value as the array element of the position corresponding to the identification number in the excellent array, until the identification number is equal to the total number of all individuals in the target population, and outputting the excellent array;

[0163] S417, generating a random value in the third interval as a first probability threshold;

[0164] S418, sequentially traverse the excellent array, and obtain the first array element that is greater than the first probability threshold as the target element;

[0165] S419, selecting corresponding individuals from the target population according to the element number corresponding to the target element and arranging them into the second population;

[0166] S420, returning to the step of generating a random value in the third interval as the first probability threshold, until a target number of individuals are selected from the target population, and a second population is obtained.

[0167] For example, in some specific implementations, the selection operator may be applied to individuals based on probability, which may be implemented as follows:

[0168] The selection operator is mainly based on probability to select more outstanding individuals. The smaller the RMS, the better the individual, and the larger the RMS, the worse the individual. The present invention uses 1 / RMS to represent the excellence of an individual. The larger the 1 / RMS, the better the individual. Formula (1) is used to solve the sum of the priorities of all individuals; formula (2) is to calculate the normalized individual priority; referring to formula (3), the sum of all individual priorities is 1; formula (4) is used to calculate the array of cumulative priority sums. How to select outstanding individuals? Calculate a random number R0 in the range of [0,1], traverse the accRMS array to find the first individual greater than R0, and the individual will be selected.

[0169] Through this selection operator, the larger the averRMS, the greater the probability of being selected, while individuals with smaller averRMS still have a chance to be selected. The reason for this design is that the current excellent individuals mean that the gene expression is excellent and worthy of being selected, but considering that the reason for the low averRMS of individuals may be that the gene expression is relatively slow, they still need to be selected with a certain probability.

[0170]

[0171]

[0172] Among them, in some embodiments, such as Figure 6 As shown, performing a crossover operation on a random pair of individuals in the target population by a second probability rule to obtain a second population may include the following steps:

[0173] S421. Based on all RMS values ​​corresponding to the target population, obtain the minimum RMS value as the target RMS value;

[0174] S422, performing an upward rounding operation on the ratio of the target RMS value to the preset iterative optimal threshold, and taking the result of the upward rounding operation multiplied by the target number as the number of crossovers;

[0175] S423, randomly obtaining two different individuals from the target population as crossover individuals; the crossover individuals include the first individual and the second individual;

[0176] S424, generating a random value in the fourth interval as a second probability threshold;

[0177] S425, when the second probability threshold is less than the preset crossover operator probability, weighted crossover is performed on the inversion parameters corresponding to the first individual and the second individual in the crossover individuals based on the preset crossover operator ratio; otherwise, weighted crossover is not performed on the crossover individuals;

[0178] S426, add 1 to the first loop number, return to execute the step of randomly obtaining two different individuals from the target population as crossover individuals, until the first loop number is equal to the crossover number, and obtain the second population; wherein the first loop number is initialized to 0.

[0179] For example, in some specific implementations, the crossover operator may be applied to individuals based on probability, which may be implemented as follows:

[0180] The crossover operator selects two individuals and then crosses their genes. Referring to formula (5), minRMS is the minimum value of the RMS of the current population, round() means rounding up, and crossCnt means the number of crossovers. The advantage of this design is that when minRMS is larger, it means the distance from the global optimal solution is farther, and the crossover operator operates more; when minRMS is smaller, it means the distance from the global optimal solution is closer, and the crossover operator can guarantee at least sizePop times.

[0181]

[0182] For a single crossover operation, based on the excellent individuals obtained by the selection operator, two individuals A1 and A2 are randomly selected; then the random number R2 in [0,1] is calculated. If R2 is less than crossoverRate, it is determined to cross over these two individuals A1 and A2; then one is randomly selected from Case1, Case2 and Case3 for the crossover operator operation; then a single crossover operator operation is completed; this process is repeated, and crossCnt crossover operations are performed, and finally the crossover operation of the entire population is completed.

[0183] Case 1: Density difference crossover, only the density difference crossover operation is performed. The crossover formula is as follows:

[0184] t1 = A1[contrast];

[0185] t2 = A2[contrast];

[0186] A1[contrast]=t1*crossoverAlpha+t2*(1-crossoverAlpha);

[0187] A2[contrast]=t2*crossoverAlpha+t1*(1-crossoverAlpha);

[0188] Case 2: Depth crossover, only depth crossover operation is performed. The crossover formula is as follows:

[0189] t1 = A1[depth];

[0190] t2 = A2[depth];

[0191] A1[depth]=t1*crossoverAlpha+t2*(1-crossoverAlpha);

[0192] A2[depth]=t2*crossoverAlpha+t1*(1-crossoverAlpha);

[0193] Case 3: Density difference crossover and depth crossover, all genes are crossovered, the operation formula is as above.

[0194] To summarize, the total number of crossovers is calculated by minRMS, and then individuals are randomly selected to confirm whether to participate in the crossover based on probability; individuals participating in the crossover choose which genes to participate in the crossover based on probability.

[0195] Among them, in some embodiments, such as Figure 7 As shown, performing a mutation operation on random individuals in the target population by using a second probability rule to obtain a second population may include the following steps:

[0196] S431, randomly obtaining an individual from the target population as a mutant individual;

[0197] S432, generating a random value in the fifth interval as a third probability threshold;

[0198] S433, when the third probability threshold is less than the preset mutation operator probability, the inversion parameters corresponding to the mutated individuals are assigned values ​​based on the population iteration number, the population maximum iteration number and the value range of the inversion parameters; otherwise, the inversion parameters of the mutated individuals are not assigned values;

[0199] The expression for assignment processing is:

[0200] A1=A1+(max-mix)*normal(0,1)*(1-random() (1－curIndex / maxGenCnt) );

[0201] In the formula, A1 represents the inversion parameter of the variant individual; max represents the maximum value of the inversion parameter interval; min represents the minimum value of the inversion parameter interval; normal(0,1) represents a random value with a normal distribution with a mean of 0 and a standard deviation of 1; random() represents a random value generated in the sixth interval; curIndex represents the number of population iterations; maxGenCnt represents the maximum number of population iterations;

[0202] S434, adding 1 to the second cycle number, returning to the step of randomly obtaining an individual from the target population as a mutant individual, until the second cycle number is equal to the target number, and the second population is obtained;

[0203] The second loop number is initialized to 0.

[0204] For example, in some specific implementations, a mutation operator is applied to an individual based on probability. The mutation operator selects a single individual and then selects a gene for mutation. The maximum number of mutations is fixed to sizePop, and then a random number R3 in [0,1] is calculated. If R3 is less than mutationRate, an individual is selected to be mutated; then one is randomly selected from Case1 and Case2 for the mutation operator operation.

[0205] Case 1: Density difference mutation, only the density difference mutation operation is performed. The mutation formula is as follows:

[0206]

[0207] Case 2: Deep mutation, only deep mutation operation is performed. The mutation formula is as follows:

[0208]

[0209] Among them, random() represents a random value generated in [0,1]; normal(0,1) represents a normal distribution with a mean of 0 and a standard deviation of 1; curIndex represents the current generation of the population.

[0210] In general, the maximum number of mutations is sizePop, and then whether mutation is needed is confirmed based on the mutationRate probability; if mutation is clear, individuals are randomly selected, and then which gene is selected for mutation.

[0211] S500, returning to the step of performing inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, until a suitable individual is output, or the number of population iterations reaches a preset maximum number of population iterations;

[0212] For example, in some specific implementations, after the operation of step S400, it means that a population reproduction is completed, and a new population is obtained. Through natural selection of survival of the fittest, it is determined whether the global optimal solution is searched based on the new population. Among them, steps S200 to S500 are executed in a loop, and each execution indicates that the population reproduction is iterated once until the maximum number of iterations of the population maxGenCnt is reached.

[0213] In some embodiments, the method may further include the following steps: when the number of population iterations is equal to the maximum number of population iterations, outputting the adapted individuals according to the target population obtained in the last cycle.

[0214] For example, in some specific implementations, the minimum value among the root mean square values ​​corresponding to all individuals in the target population obtained in the last cycle may be obtained, and the individual corresponding to the minimum value may be output as the adapted individual.

[0215] S600: Using the terrain data corresponding to the inversion parameters corresponding to the adaptation individuals in the inversion operation as the terrain processing result of the target area.

[0216] For example, in some specific implementations, the optimal solution of the population is searched, and there are only two solutions obtained here: the optimal vector that satisfies the iterative optimal threshold minIterDelt, or the optimal vector obtained by searching the entire space (i.e., the individual confirmed based on the minimum root mean square value of the last cycle). At this point, the entire search process ends, and not only the optimal solution is obtained, but also the corresponding seabed topography data.

[0217] In order to explain the principle of the technical solution of the present invention in detail, the overall process of the present invention is described below in combination with some specific embodiments. It is easy to understand that the following is an explanation of the technical principle of the present invention and cannot be regarded as a limitation of the present invention.

[0218] First of all, it should be noted that, although the Parker-Oldenburg inversion method currently occupies an important position in the study of crustal structure, its common technical limitations cannot be ignored. The primary problem is that the traditional technical solution simply treats the key parameters in the inversion process, namely density difference and depth, as fixed constants. This linearization assumption ignores the complex nonlinear relationship of the underground medium, which directly leads to the limited accuracy of the inverted seabed topography and the inability to accurately reflect the real geological structure.

[0219] Secondly, in order to solve the optimal parameters of Parker-Oldenburg inversion, the existing methods mostly use genetic algorithms for exploration. However, practice shows that the crossover operator and mutation operator design in these algorithms cannot fully adapt to the characteristics of the inversion problem, resulting in low efficiency of the solution process, large fluctuations in the average solution accuracy, and difficulty in stably outputting high-quality inversion results.

[0220] Furthermore, from the perspective of computational efficiency and automation, existing technical solutions face huge challenges. Due to the huge amount of overall computation and the lack of an efficient and unified execution framework to integrate and optimize the computational process, the automated implementation of terrain inversion becomes extremely difficult. This not only increases the cost of manual intervention, but also limits the accuracy and practicality of the inversion results, making the final calculation results rougher than the real terrain, making it difficult to meet the needs of high-precision exploration and research.

[0221] In view of this, the present invention proposes a method for solving seabed topography based on an improved genetic algorithm and Parker-Oldenburg inversion using real number coding, which finds the optimal parameters through the improved genetic algorithm and uses the Parker-Oldenburg inversion method to obtain a more accurate seabed topography. Figure 8 , the present invention may include the following process steps:

[0222] Step 1: Parameter initialization. The parameters mainly include two parts: the improved genetic algorithm parameters based on real number coding and the Parker-Oldenburg inversion method parameters (for specific parameter details, please refer to the above parameter description and will not be repeated here).

[0223] Step 2: Initialize the population. The population is composed of a large number of individuals, each of which has two genes, representing the density difference contrast and the average depth of the seabed depth. The present invention uses decimal real numbers to represent gene encoding. The core of population initialization is to generate sizePop individuals and initialize the corresponding genes for each individual. There are three strategies for population generation:

[0224] Case 1: Generated based on the center point. If the center point is given, the generation of contrast and depth is as follows:

[0225] The generation formula of the first individual gene is as follows, which is mainly generated near the center point.

[0226] contrast=initContrast+normal(0,1)*0.01;

[0227] depth=initDepth+normal(0,1)*0.1;

[0228] Other ways in which individual genes are generated are:

[0229] contrast=initContrast+(maxContrast-minContrast)*normal(0,1);

[0230] depth=initDepth+(maxDepth-minDepth)*normal(0,1);

[0231] Case 2: Random generation. If no center point is given, it is randomly generated within the search interval. The generation methods of contrast and depth are as follows:

[0232] contrast=minContrast+(maxContrast-minContrast)*random();

[0233] depth=minDepth+(maxDepth-minDepth)*random();

[0234] Case 3: Manually specified. In this case, the initialization genes of all individuals need to be manually specified.

[0235] Among them, initContrast and initDepth represent the center point; random() represents the random value generated in [0,1]; normal(0,1) represents the normal distribution with a mean of 0 and a standard deviation of 1. The obtained random number satisfies the normal distribution, thereby ensuring the fairness of gene generation. The present invention uses decimal real number encoding, and the entire calculation is based on decimal, so no additional encoding and decoding operations are required. Go to step 3.

[0236] Step 3: Calculate the seafloor topography based on the current population. For the current population, traverse the gene contrast and depth of each individual one by one, and then calculate the corresponding seafloor topography data based on the Parker-Oldenburg inversion method. Because it is a decimal real number encoding, the gene does not require additional encoding and decoding operations before being applied to the Parker-Oldenburg inversion method. Go to step 4.

[0237] Step 4: Evaluate the fitness function value of each individual. At this point, each individual in the population has calculated the corresponding seafloor topography data, and then combined with the seismic depth matrix SemisMat, the corresponding RMS is calculated. Each individual will calculate its own RMS, and finally get an array of sizePop. Go to step 4.

[0238] Step 5: Determine whether there are individuals with RMS less than the threshold. Traverse all individuals in the population and determine whether there are individuals with RMS less than the iterative optimal threshold minIterDelt. If so, go to step 11, indicating that the optimal individual (optimal solution) has been found. Otherwise, go to step 6.

[0239] Step 6: Apply the selection operator to the individual based on probability. The selection operator mainly selects better individuals based on probability. The smaller the RMS, the better the individual, and the larger the RMS, the worse the individual. The present invention uses 1 / RMS to indicate the excellence of an individual. The larger the 1 / RMS, the better the individual. Formula (1) is used to solve the sum of the priorities of all individuals; formula (2) is used to calculate the normalized individual priority; referring to formula (3), the sum of all individual priorities is 1; formula (4) is used to calculate the array of cumulative priority sums. How to select excellent individuals? Calculate a random number R0 in the range of [0,1], traverse the accRMS array to find the first individual greater than R0, and the individual will be selected.

[0240] Through this selection operator, the larger the averRMS, the greater the probability of being selected, while individuals with smaller averRMS still have a chance to be selected. The reason for this design is that the current excellent individuals mean that the gene expression is excellent and worthy of being selected, but considering that the reason for the low averRMS of individuals may be that the gene expression is relatively slow, they still need to be selected with a certain probability.

[0241]

[0242]

[0243] For example, if the array of the original RMS is [3,7,1,9], there are four individual RMSs, namely 3, 7, 1, and 9.

[0244] According to formula (1), sum = 1 / 3 + 1 / 7 + 1 / 1 + 1 / 9 = 1.5873;

[0245] According to formula (2), we get the averRMS array as [1 / 3 / 1.5873, 1 / 7 / 1.5873, 1 / 1 / 1.5873, 1 / 9 / 1.5873] and calculate averRMS = [0.21, 0.09, 0.63, 0.07]. According to formula (3), we can verify that the sum of the averRMS array is 1.

[0246] The accRMS array is calculated by formula (4):

[0247] accRMS[0] = 0.21;

[0248] accRMS[1] = 0.21 + 0.09 = 0.3;

[0249] accRMS[2] = 0.21 + 0.09 + 0.63 = 0.93;

[0250] accRMS[3] = 0.21 + 0.09 + 0.63 + 0.07 = 1;

[0251] Then calculate the random value R1. If R1 < accRMS[0], then accRMS[0] is selected; if R1 >= accRMS[0] and R1 < accRMS[1], then accRMS[1] is selected; if R1 >= accRMS[1] and R1 < accRMS[2], then accRMS[2] is selected; if R1 >= accRMS[2] and R1 < accRMS[3], then accRMS[3] is selected; if R1 >= accRMS[3], then accRMS[3] is selected, and this is the critical case.

[0252] In this way, through sizePop runs of the selection operator, finally sizePop excellent individuals are obtained. Note that the same individual may be selected multiple times. Go to step 7.

[0253] Step 7: Apply the crossover operator to the individuals based on probability. The crossover operator is to select two individuals and then perform gene crossover. Referring to formula (5), minRMS is the minimum value of the RMS of the current population, round() means rounding up, and crossCnt represents the number of crossovers. The advantage of this design is that when minRMS is larger, it means the distance from the global optimal solution is farther, and at this time the crossover operator operates more; when minRMS is smaller, it means the distance from the global optimal solution is closer, and the crossover operator can ensure at least sizePop times.

[0254]

[0255] For a single crossover operation, based on the excellent individuals obtained by the selection operator, randomly select two individuals A1 and A2; then calculate the random number R2 in [0, 1]. If R2 is less than crossoverRate, then determine to perform crossover on these two individuals A1 and A2; then randomly select one from Case1, Case2, and Case3 to perform the crossover operator operation; then the single crossover operator operation is completed; repeat this way, perform crossCnt crossover operations, and finally complete the crossover operation of the entire population.

[0256] Case1: Density difference crossover, only perform the crossover operation of the density difference. The crossover formula is as follows:

[0257] t1 = A1[contrast];

[0258] t2 = A2[contrast];

[0259] A1[contrast]=t1*crossoverAlpha+t2*(1-crossoverAlpha);

[0260] A2[contrast]=t2*crossoverAlpha+t1*(1-crossoverAlpha);

[0261] Case 2: Depth crossover, only depth crossover operation is performed. The crossover formula is as follows:

[0262] t1 = A1[depth];

[0263] t2 = A2[depth];

[0264] A1[depth]=t1*crossoverAlpha+t2*(1-crossoverAlpha);

[0265] A2[depth]=t2*crossoverAlpha+t1*(1-crossoverAlpha);

[0266] Case 3: Density difference crossover and depth crossover, all genes are crossovered, the operation formula is as above.

[0267] To summarize, the total number of crossovers is calculated by minRMS, and then individuals are randomly selected to confirm whether to participate in the crossover based on probability; individuals participating in the crossover choose which genes to participate in the crossover based on probability.

[0268] Here is a specific example. If minRMS is 1.2, minIterDelt is 1, sizePop is 4, and crossoverAlpha is 0.8, then corssCnt = round(1.2 / 1)*4 = 8, that is, at most 8 pairs of individuals participate in the crossover operation. Here, we take the crossover of a pair of individuals as an example. Assume that the density difference of individual A1 is A1[contrast] = 0.5, and the depth of individual A1 is A1 [depth] = 20; the density difference of individual A2 is A2[contrast] = 0.3, and the depth of individual A2 is A2 [depth] = 10. If the random number R2 is less than crossoverRate, A1 and A2 are definitely crossed.

[0269] If we only do the density intersection:

[0270] A1[contrast]=0.5*0.8+(1-0.8)*0.3=0.46,

[0271] A2[contrast]=0.3*0.8+(1-0.8)*0.5=0.34;

[0272] If only the depth intersection is done, then:

[0273] A1[depth]=20*0.8+(1-0.8)*10=18,

[0274] A2[depth]=10*0.8+(1-0.8)*20=12;

[0275] If you do the intersection of density difference and depth at the same time:

[0276] A1[contrast]=0.5*0.8+(1-0.8)*0.3=0.46,

[0277] A2[contrast]=0.3*0.8+(1-0.8)*0.5=0.34;

[0278] A1[depth]=20*0.8+(1-0.8)*10=18,

[0279] A2[depth]=10*0.8+(1-0.8)*20=12;

[0280] Go to step 8.

[0281] Step 8: Apply the mutation operator to individuals based on probability. The mutation operator selects a single individual and then selects genes for mutation. The maximum number of mutations is fixed to sizePop, and then a random number R3 in [0,1] is calculated. If R3 is less than mutationRate, an individual is selected for mutation; then one is randomly selected from Case1 and Case2 for mutation operator operation.

[0282] Case 1: Density difference mutation, only the density difference mutation operation is performed. The mutation formula is as follows:

[0283]

[0284] Case 2: Deep mutation, only deep mutation operation is performed. The mutation formula is as follows:

[0285]

[0286] Among them, random() represents a random value generated in [0,1]; normal(0,1) represents a normal distribution with a mean of 0 and a standard deviation of 1; curIndex represents the current generation of the population.

[0287] In general, the maximum number of mutations is sizePop, and then whether mutation is needed is confirmed based on the mutationRate probability; if mutation is clear, individuals are randomly selected, and then which gene is selected for mutation.

[0288] Step 9: Get the next generation population. After the operations from step 6 to step 8, it means that one population reproduction is completed, and a new population is obtained. Through natural selection of survival of the fittest, determine whether the global optimal solution is found based on the new population. Go to step 10.

[0289] Step 10: Determine whether the maximum number of population iterations is met. Steps 3 to 9 are executed in a loop. Each execution indicates that the population reproduces once until the maximum number of population iterations maxGenCnt is reached. If it is not exceeded, go to step 3, otherwise go to step 11.

[0290] Step 11: Find the optimal solution. This step yields the optimal solution for the entire search process. There are only two solutions: the optimal vector that satisfies the iterative optimal threshold minIterDelt, or the optimal vector obtained by searching the entire space. At this point, the entire search process ends, and not only the optimal solution is obtained, but also the corresponding seafloor topography data.

[0291] In summary, the present invention proposes a method for solving seabed topography based on an improved genetic algorithm based on real number coding and Parker-Oldenburg inversion, which finds the optimal parameters by using an improved genetic algorithm based on real number coding and uses the Parker-Oldenburg inversion method to obtain a more accurate seabed topography. Compared with the prior art, the present invention has at least the following beneficial effects:

[0292] Improved genetic algorithm with real number coding: In view of the shortcomings of traditional methods in optimizing Parker-Oldenburg inversion parameters (density difference and depth), this scheme innovatively proposes an improved genetic algorithm based on real number coding. The real number coding strategy shows excellent advantages in continuous parameter optimization and continuous gradient problems, and can explore the parameter space more finely, thereby accurately locating the optimal density difference and depth combination, and significantly improving the accuracy of terrain inversion.

[0293] Optimize the design of crossover and mutation operators: To further improve the performance of the algorithm, we carefully designed crossover and mutation operators suitable for real number coding. These optimized operators not only enhance the algorithm's search capability in a wide parameter space, but also effectively reduce the computational complexity of the genetic algorithm, making the search process more efficient and robust. This improvement ensures that the improved genetic algorithm can achieve both improved accuracy and efficiency, providing strong technical support for terrain inversion.

[0294] Fusion strategy to improve overall performance: This solution cleverly combines the improved genetic algorithm based on real number coding with the Parker-Oldenburg inversion method to form an efficient and accurate terrain inversion framework. This fusion strategy not only retains the advantages of the Parker-Oldenburg method in processing gravity anomaly data, but also greatly improves the accuracy and computational efficiency of the inversion results through the parameter optimization capability of the genetic algorithm. Ultimately, this solution provides a strong guarantee for achieving high-precision and high-efficiency seabed topography exploration and research.

[0295] On the other hand, Fig. 9 As shown, the embodiment of the present invention provides a terrain processing device 900 based on a genetic algorithm, which may include:

[0296] The first module 901 is used to initialize the inversion parameters of the target number, obtain the first population based on the inversion parameter initialization, and use the first population as the target population;

[0297] Among them, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth;

[0298] The second module 902 is used to perform inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, and obtain the root mean square value corresponding to each individual in the target population;

[0299] The root mean square value represents the adaptability of the inversion parameters corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the terrain data of the target area;

[0300] The third module 903 is used to output the individual corresponding to the root mean square value as the adapted individual when the root mean square value of the target population meets the preset threshold condition; otherwise,

[0301] The fourth module 904 is used to perform genetic operation on the target population based on a preset genetic algorithm to obtain a second population; the second population is used as the target population, and the number of population iterations is increased by 1;

[0302] The fifth module 905 is used to return to the step of performing the inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, until the adapted individual is output, or the number of population iterations reaches a preset maximum number of population iterations;

[0303] The sixth module 906 is used to use the terrain data obtained by the inversion operation corresponding to the inversion parameters corresponding to the adaptation individuals as the terrain processing result of the target area.

[0304] In some embodiments, the apparatus may further include:

[0305] The seventh module is used to output the adapted individuals according to the target population obtained in the last cycle when the number of population iterations is equal to the maximum number of population iterations.

[0306] The contents of the method embodiments of the present invention are all applicable to the device embodiments. The functions specifically implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0307] On the other hand, an embodiment of the present invention further provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above-mentioned hydrate stability region bottom boundary prediction method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0308] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0309] like Fig.10 As shown, Fig.10 The hardware structure of an electronic device 1000 of another embodiment is illustrated. The electronic device 1000 includes:

[0310] The processor 1001 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0311] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the network node population optimization method of the embodiment of the present invention;

[0312] Input / output interface 1003, used to implement information input and output;

[0313] The communication interface 1004 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0314] A bus 1005 , which transmits information between various components of the device (e.g., the processor 1001 , the memory 1002 , the input / output interface 1003 , and the communication interface 1004 );

[0315] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .

[0316] The electronic device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0317] The contents of the method embodiments of the present invention are all applicable to the electronic device embodiments. The functions specifically implemented by the electronic device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0318] Another aspect of an embodiment of the present invention further provides a computer-readable storage medium, the storage medium stores a program, and the program is executed by a processor to implement the above method.

[0319] The contents of the method embodiments of the present invention are all applicable to the computer-readable storage medium embodiments. The functions specifically implemented by the computer-readable storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0320] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0321] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0322] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A terrain processing method based on genetic algorithm, characterized in that: The following steps are involved: Initializing the inversion parameters of the target number, obtaining a first population based on the initialization of the inversion parameters; and using the first population as the target population; Wherein, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth; Based on the inversion parameters corresponding to each of the individuals in the target population, an inversion operation and fitness evaluation are performed on the target area to obtain a root mean square value corresponding to each of the individuals in the target population; The root mean square value represents the adaptability of the inversion parameter corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the topographic data of the target area; When the root mean square value in the target population meets the preset threshold condition, the individual corresponding to the root mean square value is output as the adapted individual; otherwise, Performing genetic operations on the target population based on a preset genetic algorithm to obtain a second population; using the second population as the target population, and increasing the number of population iterations by 1; Return to the step of performing the inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population until the adapted individual is output, or the number of population iterations reaches a preset maximum number of population iterations; The terrain data corresponding to the inversion parameters corresponding to the adapted individuals in the inversion operation is obtained as the terrain processing result of the target area.

2. The terrain processing method based on genetic algorithm according to claim 1, characterized in that: The initialization of the target number of inversion parameters comprises the following steps: In response to a first input instruction of the target object, a center point is obtained, a first value is assigned to the center point by a first random number, and a second number of inversion parameters are randomly generated near the center point; and a second value is assigned to the center point by multiplying the first random number by a difference between the first interval range and the first interval, and a third number of inversion parameters are randomly generated near the center point; Wherein, the first random number represents a normally distributed random number with a mean of 0 and a standard deviation of 1; the first random number is updated before each generation of the inversion parameter; the first interval represents a value range of the inversion parameter; the sum of the second number and the third number is equal to the target number; Alternatively, based on a second random number, the target number of inversion parameters is randomly generated within the first interval; the second random number represents a random value within the second interval; Alternatively, the inversion parameters of the target quantity are obtained in response to a second input instruction of the target object.

3. The terrain processing method based on genetic algorithm according to claim 1, characterized in that: The inversion operation and fitness evaluation of the target area are performed based on the inversion parameters corresponding to each individual in the target population to obtain the root mean square value corresponding to each individual in the target population, including the following steps: Based on the inversion parameters corresponding to the individuals in the target population, the terrain data of the target area corresponding to each individual is obtained by processing using the Parker-Oldenburg inversion method; Acquire a seismic depth matrix of the target area; Traversing the first data values ​​of all location points of the terrain data corresponding to the individual, collecting the second data values ​​of the corresponding location points from the seismic depth matrix; obtaining the third data value of the corresponding location point based on the square of the difference between the first data value and the second data value; performing square root processing based on the average value of the third data values ​​of all the location points to obtain the root mean square value corresponding to the corresponding individual.

4. The terrain processing method based on genetic algorithm according to claim 1, characterized in that: The step of performing a genetic operation on the target population based on a preset genetic algorithm to obtain a second population comprises at least one of the following steps: Based on the excellence of each of the individuals in the target population, selecting the target number of individuals from the target population by combining a first probability rule with a first probability threshold, to obtain the second population; wherein the excellence is determined based on the inverse of the root mean square value; Performing a crossover operation on a random pair of individuals in the target population using a second probability rule to obtain the second population; The second population is obtained by performing a mutation operation on the random individuals in the target population using a second probability rule.

5. The terrain processing method based on genetic algorithm according to claim 4 is characterized in that: The step of selecting the target number of individuals from the target population based on the excellence of each individual in the target population by combining a first probability rule with a first probability threshold to obtain the second population comprises the following steps: Taking the reciprocal of the root mean square value as the excellence level corresponding to each individual in the target population; Normalizing all the excellence levels to obtain a normalized excellence value corresponding to each individual; Initialize the identification number to 0 and initialize the excellent array to empty; Taking the first individual of the target population as a data individual, and taking the excellent value corresponding to the data individual as an output value; Using the output value as the array element at the position corresponding to the identification number in the excellent array; increasing the identification number by 1; using the next individual of the data individual in the target population as the data individual; Accumulate the excellent value corresponding to the data individual to the output value, return to the step of using the output value as the array element of the position corresponding to the identification number in the excellent array, until the identification number is equal to the total number of all the individuals of the target population, and output the excellent array; generating a random value within a third interval as a first probability threshold; Sequentially traverse the excellent array to obtain the first array element that is greater than the first probability threshold as the target element; According to the element number corresponding to the target element, the corresponding individuals are selected from the target population and arranged into the second population; Return to the step of generating a random value within the third interval as the first probability threshold, until the target number of individuals are selected from the target population, and the second population is obtained.

6. The terrain processing method based on genetic algorithm according to claim 4, characterized in that: The step of performing a crossover operation on a random pair of individuals in the target population by using a second probability rule to obtain the second population comprises the following steps: Based on all the root mean square values ​​corresponding to the target population, obtaining the minimum root mean square value as a target root mean square value; Performing an upward rounding operation on the ratio of the target RMS value to the preset iterative optimal threshold, and taking the result of multiplying the result of the upward rounding operation by the target number as the number of crossovers; Randomly obtain two different individuals from the target population as crossover individuals; the crossover individuals include a first individual and a second individual; generating a random value within a fourth interval as a second probability threshold; When the second probability threshold is less than a preset crossover operator probability, weighted crossover is performed on the inversion parameters corresponding to the first individual and the second individual in the crossover individuals based on a preset crossover operator ratio; otherwise, the weighted crossover is not performed on the crossover individuals; Add 1 to the first cycle number, return to the step of randomly obtaining two different individuals from the target population as crossover individuals, until the first cycle number is equal to the crossover number, and the second population is obtained; Wherein, the first cycle number is initialized to 0.

7. The terrain processing method based on genetic algorithm according to claim 4, characterized in that: The step of performing a mutation operation on the random individuals in the target population by using a second probability rule to obtain the second population comprises the following steps: Randomly obtain one of the individuals from the target population as a mutant individual; generating a random value within a fifth interval as a third probability threshold; When the third probability threshold is less than the preset mutation operator probability, the inversion parameter corresponding to the mutant individual is assigned a value based on the population iteration number, the population maximum iteration number and the value interval of the inversion parameter; otherwise, the inversion parameter of the mutant individual is not assigned a value; The expression of the assignment process is: A1=A1+(max-mix)*normal(0,1)*(1-random() (1－curIndex / maxGenCnt) ); In the formula, A1 represents the inversion parameter of the variant individual; max represents the maximum value of the inversion parameter interval; min represents the minimum value of the inversion parameter interval; normal(0,1) represents a random value with a normal distribution with a mean of 0 and a standard deviation of 1; random() represents a random value generated in the sixth interval; curIndex represents the number of population iterations; maxGenCnt represents the maximum number of population iterations; Add 1 to the second cycle number, and return to the step of randomly obtaining one of the individuals from the target population as a mutant individual until the second cycle number is equal to the target number, and the second population is obtained; Wherein, the second cycle number is initialized to 0.

8. A terrain processing device based on genetic algorithm, characterized in that: include: A first module is used to initialize the inversion parameters of a target number, and obtain a first population based on the inversion parameter initialization; taking the first population as the target population; Wherein, the inversion parameters correspond to individuals in the target population; the inversion parameters include density difference and average seafloor depth; The second module is used to perform inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, and obtain the root mean square value corresponding to each individual in the target population; The root mean square value represents the adaptability of the inversion parameter corresponding to the individual to participate in the inversion operation of the target area; the result of the inversion operation is the topographic data of the target area; The third module is used to output the individual corresponding to the corresponding root mean square value as the adapted individual when the root mean square value in the target population meets the preset threshold condition; otherwise, The fourth module is used to perform genetic operation on the target population based on a preset genetic algorithm to obtain a second population; use the second population as the target population, and increase the number of population iterations by 1; The fifth module is used to return to the step of performing the inversion operation and fitness evaluation of the target area based on the inversion parameters corresponding to each individual in the target population, until the adapted individual is output, or the number of population iterations reaches a preset maximum number of population iterations; The sixth module is used to use the terrain data obtained in the inversion operation corresponding to the inversion parameters corresponding to the adapted individual as the terrain processing result of the target area.

9. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 7.

10. A computer storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the method according to any one of claims 1 to 7 when executed by the processor.

Images