Hydrometeorological change point detection method and system

Abstract

The invention relates to a hydro meteorology change point detection method and system, and belongs to the technical field of hydro meteorology scientificity, and the method comprises the steps: obtaining the observation data of a hydro meteorology time sequence; the number of change points of the observation data is initially 1; judging whether the number of change points is greater than a given maximum value or not; when the number of the change points is smaller than or equal to the maximum value, calculating the change rate of the fitness function under the current number of the change points according to a change point set obtained by carrying out change point detection on the observation data by utilizing a memetic algorithm and a corresponding fitness function value, and adding 1 to the number of the change points until the number of the change points is larger than the maximum value; and when the number of change points is greater than a given maximum value, determining an optimal number of change points and an optimal change point set according to the change rate of the fitness function corresponding to different numbers of change points. According to the invention, the dynamic change of the time sequence data can be adaptively processed, and the performance of hydro-meteorological change point detection is improved.

Description

Technical Field

[0001] This application relates to the field of hydrological and meteorological science and technology, and in particular to a method and system for detecting hydrological and meteorological change points. Background Technology

[0002] Environmental changes caused by natural factors or human activities are the main reason for fluctuations in hydrological and meteorological time series data. For example, natural disasters such as tsunamis, typhoons, and floods can cause abrupt changes in hydrological and meteorological data. At the same time, these changes in climate conditions also have a serious impact on agriculture, energy, transportation, and other fields. Therefore, the detection of turning points in hydrological and meteorological time series data has become one of the important topics in hydrological and meteorological scientific research. It not only helps in understanding the trends and characteristics of hydrological and meteorological data changes, but also provides a foundation for the study of climate and hydrological cycles.

[0003] Hydrometeorological change point detection refers to the process of monitoring, analyzing, and judging changes in hydrometeorological elements (such as wind speed, precipitation, and flow) using intelligent algorithms, aiming to reveal how hydrometeorological data changes over time. Currently, the detection of hydrometeorological changes is mainly based on time series change point detection technology. Existing change point detection methods can be classified from different perspectives as follows: according to the number of change points, they can be divided into single-change point and multi-change point detection methods; according to the dimension of the time series, they can be divided into univariate and multivariate time series change point detection. However, determining the number and location of change points remains a key issue in change point detection. Summary of the Invention

[0004] Therefore, it is necessary to provide a method and system for detecting hydrological and meteorological change points, which can better adapt to the dynamic changes in time series data and improve the performance of hydrological and meteorological change point detection, based on the above technical problems.

[0005] In a first aspect, embodiments of this application provide a method for detecting hydro-meteorological change points, the method comprising: S1. Obtain observational data of hydrological and meteorological time series; S2. Initialize the number of variable points in the observation data to 1; S3. Determine if the number of variable points is greater than the given maximum value; S4. When the number of change points is less than or equal to the given maximum value, use the meme algorithm to detect change points in the observed data to obtain the set of change points and the corresponding fitness function value. S5. Based on the set of variable points and the corresponding fitness function values, calculate the rate of change of the fitness function for the current number of variable points, and increment the number of variable points by 1. S6. Repeat steps S4 and S5 until the number of variable points is greater than the given maximum value. S7. When the number of variables is greater than the given maximum value, determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables.

[0006] In one embodiment, the method for detecting change points in the observed data is as follows: S41. Initialize the population size in the meme algorithm. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability , obtain the initialization parameters; S42. Randomly generate the initial population based on the initialization parameters. ; S43, Set the current population iteration count. The current population is ; S44. Determine the current population iteration count. Is it less than the number of population iterations? ; S45. If the current population iteration count is... Less than the number of population iterations Based on crossover probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning. S46. Merge the current population and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population. The current population iteration count is incremented by 1; the new population becomes the current population for the next cycle. S47. Repeat steps S45 to S46 until the current population iteration count is reached. Greater than or equal to the number of population iterations ; S48. If the current population iteration count is... Greater than or equal to the number of population iterations The individual with the highest fitness function value in the current population is taken as the set of change points for the observation data.

[0007] In one embodiment, the initial population Include Individuals, each individual is composed of It consists of several variable points.

[0008] In one embodiment, the formula for calculating the rate of change of the fitness function is: ; In the formula, Indicates the number of variable points as The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points. The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points as The fitness function value corresponding to the set of variable points obtained by the time-mode algorithm.

[0009] In one embodiment, the fitness function value is calculated as follows: ; In the formula, It is by The first of the observed data determined by the first change point The length of each segment It is the first The covariance of each segment, It is a regularization term. It is a regularization parameter. .

[0010] In one embodiment, the covariance is calculated as follows: ; in, ; In the formula, and They are the first The variable points at both ends of a segment, It is the first in the hydrological and meteorological time series observation data Data points, It is the first The mean of each segment.

[0011] In one embodiment, the method for determining the optimal number of change points and the set of change points is as follows: Based on the rate of change of the fitness function corresponding to different numbers of variables, the number of variables corresponding to the largest rate of change is selected as the optimal number of variables, and the set of variables corresponding to the optimal number of variables is the optimal set of variables.

[0012] In one embodiment, the acquired hydro-meteorological time series observation data are characterized by meteorological elements.

[0013] Secondly, embodiments of this application provide a hydrological and meteorological change point detection system, including: The data acquisition module is used to acquire hydrological and meteorological time series observation data; An initialization module, connected to the data acquisition module, is used to initialize the number of variable points in the observation data; The settings module is used to set a given maximum value; The judgment module, connected to the initialization module and the setting module, is used to determine whether the number of variable points is greater than a given maximum value; The detection module, connected to the judgment module, is used to detect changes in the observed data using the meme algorithm when the number of change points is less than or equal to a given maximum value, thereby obtaining a set of change points and the corresponding fitness function value. The calculation module, connected to the detection module, is used to calculate the rate of change of the fitness function under the current number of variable points based on the set of variable points and the corresponding fitness function values. The determination module, connected to the calculation module and the judgment module, is used to determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables when the number of variables is greater than a given maximum value.

[0014] In one embodiment, the detection module includes: The initialization submodule is used to initialize the population size. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability This yields the initialization parameters and the current population iteration count. ; The generation submodule, connected to the initialization submodule, is used to randomly generate an initial population based on initialization parameters. and based on cross probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning, as well as by merging the current population. and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population; The determination submodule, connected to the initialization submodule and the generation submodule, is used to determine the current population iteration number. Is it less than the number of population iterations? ; The output submodule, connected to the generation submodule, is used to perform iterations in the current population. Greater than or equal to the number of population iterations When the individual with the highest fitness function value in the current population is selected as the set of change points for the observation data, the data is analyzed.

[0015] The above-mentioned hydro-meteorological change point detection method uses a meme algorithm to find the change point that maximizes the fitness function, thus transforming the hydro-meteorological change point detection problem into a change point detection problem for multivariate time series data. It has the advantages of high detection accuracy and ease of implementation. The change rate of the fitness function is used to adaptively determine the number of change points in the observation data.

[0016] The above-mentioned method for detecting hydrological and meteorological change points can provide a foundation for research in multiple fields such as ecological environment, agriculture and forestry, and urban planning and management. At the same time, this application is not limited to the detection of hydrological and meteorological data, but can also be used as a method for detecting changes in time series data in other fields.

[0017] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart illustrating the hydrological and meteorological change point detection method provided in the embodiments of this application.

[0019] Figure 2 This is a schematic diagram of the process of using a meme algorithm to detect change points in observation data, provided in one embodiment of this application.

[0020] Figure 3 This is a schematic diagram of the overall process of the hydrological and meteorological change point detection method provided in one embodiment of this application.

[0021] Figure 4 This is a structural block diagram of the hydrological and meteorological change point detection system provided in the embodiments of this application.

[0022] Figure 5 This is a block diagram of the detection module structure provided in the embodiments of this application.

[0023] Figure 6 This is a schematic diagram of the result of change point detection on a multivariate hydrological and meteorological dataset provided in one embodiment of this application.

[0024] In the diagram, 1 is the hydrological and meteorological change point detection system, 10 is the data acquisition module, 20 is the initialization module, 30 is the setting module, 40 is the judgment module, 50 is the detection module, 51 is the initialization submodule, 52 is the generation submodule, 53 is the judgment submodule, 54 is the output submodule, 60 is the calculation module, and 70 is the determination module. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0026] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0027] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0028] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0029] In a first aspect, embodiments of this application provide a method for detecting hydro-meteorological change points, as shown in Figure 1, comprising the following steps: S1. Obtain observational data of hydrological and meteorological time series.

[0030] Specifically, the acquired hydro-meteorological time series observation data are characterized by meteorological elements.

[0031] For example, acquiring hydrological and meteorological time-series observation data from meteorological stations. Each data point Depend on Characterized by meteorological elements, , The length of the observed data.

[0032] S2. Initialize the number of variable points in the observation data to 1.

[0033] For example, observation data The initial number of variable points is .

[0034] S3. Determine if the number of variable points is greater than the given maximum value. .

[0035] It should be noted that the given maximum value This can be set according to actual needs. For example, a given maximum value... It can be set to 10.

[0036] S4. When the number of change points is less than or equal to the given maximum value, use the meme algorithm to detect change points in the observed data to obtain the set of change points and the corresponding fitness function value.

[0037] In this embodiment, the meme algorithm is used to find the change point that maximizes the fitness function, so as to transform the hydrological and meteorological change point detection problem into the change point detection problem of multivariate time series data, which has the advantages of high detection accuracy and easy implementation.

[0038] In one embodiment of this application, as Figure 2 As shown, the specific steps for detecting change points in observed data using the meme algorithm include: S41. Initialize the population size in the meme algorithm. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability , and obtain the initialization parameters.

[0039] Population size The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability This can be configured according to actual needs. For example, initializing the population size for observation data. The population size newly generated in the 100th generation. 50, number of population iterations 100, crossover probability The mutation probability is 0.6. 0.2, local fine-tuning probability It is 0.2.

[0040] S42. Randomly generate the initial population based on the initialization parameters. .

[0041] Specifically, the population is generated randomly. Include Individuals, each individual is composed of It consists of several variable points.

[0042] S43, Set the current population iteration count. The current population is .

[0043] S44. Determine the current population iteration count. Is it less than the number of population iterations? .

[0044] S45. If the current population iteration count is... Less than the number of population iterations Based on crossover probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning.

[0045] Specifically, the crossover method is as follows: randomly select pairs of individuals in the current population according to the crossover probability, and generate offspring using single-point crossover (binary) or arithmetic crossover (real number).

[0046] Specifically, the mutation method is as follows: randomly select the gene position of an individual in the current population according to the mutation probability, and modify the gene by bit flipping (binary) or Gaussian mutation (real number).

[0047] Specifically, the local fine-tuning method is as follows: make small adjustments (e.g., with a boundary of [-5,5]) to individuals in the current population according to the local fine-tuning probability (using random perturbation to avoid premature convergence and strengthen local search).

[0048] S46. Merge the current population and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population. And increment the current population iteration count by 1, i.e., the current population iteration count. .

[0049] In this embodiment of the application, the new population serves as the current population for the next cycle.

[0050] S47. Repeat steps S45 to S46 until the current population iteration count is reached. Greater than or equal to the number of population iterations .

[0051] S48. If the current population iteration count is... Greater than or equal to the number of population iterations , will the current population The individuals with the highest fitness function values ​​are used as the set of change points in the hydro-meteorological time series.

[0052] S5. Based on the set of variable points and the corresponding fitness function values, calculate the rate of change of the fitness function for the current number of variable points, and increment the number of variable points by 1.

[0053] In this embodiment of the application, by calculating the rate of change of the fitness function under different numbers of change points, the number of change points in the observation data can be adaptively determined.

[0054] In this embodiment of the application, the formula for calculating the rate of change of the fitness function is: ; In the formula, Indicates the number of variable points as The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points as The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points as The fitness function value corresponding to the set of variable points obtained by the time-mode algorithm.

[0055] The method for calculating the fitness function value: ; In the formula, It is by The first of the observed data determined by the first change point The length of each segment It is the first The covariance of each segment, It is a regularization term. It is a regularization parameter. .

[0056] The method for calculating covariance is as follows: ; in, ; In the formula, and They are the first The variable points at both ends of a segment, It is the first in the hydrological and meteorological time series observation data Data points, It is the first The mean of each segment.

[0057] S6. Repeat steps S4 and S5 until the number of variable points is greater than the given maximum value.

[0058] S7. When the number of variables is greater than the given maximum value, determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables.

[0059] Specifically, the method for determining the optimal number of variables and the set of variables is as follows: based on the rate of change of the fitness function corresponding to different numbers of variables, select the number of variables corresponding to the largest rate of change as the optimal number of variables, and the set of variables corresponding to the optimal number of variables is the optimal set of variables.

[0060] For example, based on the set of fitness function change rates corresponding to all number of changes during the change point detection process. Choose the number of change points corresponding to the largest rate of change of the fitness function. The optimal number of variables is: The set of change points corresponding to each time is the moment when a change occurs in the hydrological and meteorological time series (i.e., the optimal set of change points).

[0061] In one exemplary embodiment, such as Figure 3 As shown, a method for detecting hydro-meteorological change points includes the following specific steps: S101. Obtain observational data of hydrological and meteorological time series.

[0062] S102. Initialize the number of variable points in the observation data to 1.

[0063] S103. Determine whether the number of variable points is greater than the given maximum value.

[0064] S104. When the number of change points is less than or equal to the given maximum value, initialize the population size in the meme algorithm. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability , and obtain the initialization parameters.

[0065] S105. Randomly generate the initial population based on the initialization parameters. .

[0066] S106, Set the current population iteration count. The current population is .

[0067] S107. Determine the current population iteration count. Is it less than the number of population iterations? .

[0068] S108. If the current population iteration count is... Less than the number of population iterations Based on crossover probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning.

[0069] S109, Merge the current population and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population. And set the current population iteration count. .

[0070] S110. Repeat steps S108 to S109 until the current population iteration count is reached. Greater than or equal to the number of population iterations .

[0071] S111, if the current population iteration count is... Greater than or equal to the number of population iterations , will the current population The individuals with the highest fitness function values ​​are used as the set of change points in the hydro-meteorological time series.

[0072] S112. Based on the set of variable points and the corresponding fitness function values, calculate the rate of change of the fitness function under the current number of variable points, and increment the number of variable points by 1.

[0073] S113. Repeat steps S104 to S112 until the number of variable points is greater than the given maximum value.

[0074] S114. When the number of variables is greater than the given maximum value, determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables.

[0075] The hydrological and meteorological change point detection method described in this application can provide a foundation for research in multiple fields such as ecological environment, agriculture and forestry, and urban planning and management. At the same time, this application is not limited to the detection of hydrological and meteorological data, but can also be used as a method for detecting changes in time-series data involved in other fields.

[0076] Secondly, embodiments of this application provide a hydro-meteorological change point detection system, as shown in Figure 4, including: Data acquisition module 10 is used to acquire hydrological and meteorological time series observation data; Initialization module 20, connected to data acquisition module 10, is used to initialize the number of variable points in the observation data; Setting module 30 is used to set a given maximum value; The judgment module 40, connected to the initialization module 20 and the setting module 30, is used to determine whether the number of variable points is greater than a given maximum value; The detection module 50, connected to the judgment module 40, is used to detect changes in the observed data using the meme algorithm when the number of change points is less than or equal to a given maximum value, thereby obtaining a set of change points and the corresponding fitness function value. The calculation module 60, connected to the detection module 50, is used to calculate the rate of change of the fitness function under the current number of variable points based on the set of variable points and the corresponding fitness function values. The determination module 70, connected to the calculation module 60 and the judgment module 40, is used to determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables when the number of variables is greater than a given maximum value.

[0077] In one embodiment of this application, the detection module 50, as shown... Figure 5 As shown, it includes: Initialization submodule 51 is used to initialize the population size. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability This yields the initialization parameters and the current population iteration count. ; Generation submodule 52, connected to the initialization submodule 51, is used to randomly generate an initial population based on initialization parameters. Based on crossover probability, mutation probability, and local fine-tuning probability, new individuals are generated by crossover, mutation, and local fine-tuning of the current population. Additionally, the current population and all new individuals are merged into a new population, and the individual with the highest fitness function value is selected. Individuals generate a new population; The judgment submodule 53, connected to the initialization submodule 51 and the generation submodule 52, is used to determine whether the current population iteration count is less than the population iteration count. The output submodule 54, connected to the generation submodule 52, is used to select the individual with the highest fitness function value in the current population as the set of change points.

[0078] To verify the effectiveness of the embodiments of this application, simulation verification was performed. The embodiments of this application were verified based on simulated univariate time series data, and the obtained observation data... , , , The first and the second, respectively. The, the Data points, Set the maximum number of change points for the length of the observed data. The parameter population size in the meme algorithm The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability and regularization parameters The values ​​were set to 100, 50, 100, 0.6, 0.2, 0.2, and 0.001, respectively. A time series containing four points of change was generated using a second-order autoregressive model (AR(2)), the specific expression of which is as follows: ; ; in, , , It is a random variable that follows a standard Gaussian distribution (mean is 0, variance is 1).

[0079] The simulation results are shown in Table 1. The time series change points (i.e. change points) detected by the meme algorithm are 199, 401, 601 and 802, respectively. It can be found that the change points detected by this application are very close to the actual change points, the error is within an acceptable range, and the number of change points obtained is completely consistent with the real situation, indicating that this application has good detection performance.

[0080] Table 1 Simulation Experiment Results

[0081] To verify the accuracy of the embodiments in this application, verification was conducted using a multivariate hydrological and meteorological dataset. The experimental results are as follows: Figure 6 As shown.

[0082] Hydrometeorological time series data are among the fundamental data used to study phenomena related to Earth. The UCI Machine Learning Database provides hourly meteorological data from Beijing meteorological stations, including meteorological elements such as dew point, air pressure, and PM2.5. For example... Figure 6 As shown in the embodiment of this application, hourly meteorological data from Beijing meteorological stations provided by the UCI machine learning database were used to detect change points. The experiment selected dew point and air pressure data, with a total length of 18,000 bytes, and standardized the data. The parameter settings in this embodiment are the same as those in the simulation experiment. While it is impossible to accurately determine the true location of change in real data, the quality of the detection results can be judged by observing the trend of data changes. Figure 6The blue and orange lines represent the two-dimensional data of dew point and air pressure in the hourly meteorological data of Beijing meteorological stations, respectively. The black dashed line is the location of the trend change (i.e. the change point) detected in the embodiment of this application. The results detected by this application are basically consistent with the changing trend of the hydrological and meteorological time series. Therefore, this application can accurately realize the change point detection of multivariate hydrological and meteorological data.

[0083] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0084] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for detecting hydrological and meteorological change points, characterized in that, The method includes: S1. Obtain observational data of hydrological and meteorological time series; S2. Initialize the number of variable points in the observation data to 1; S3. Determine if the number of variable points is greater than the given maximum value; S4. When the number of change points is less than or equal to the given maximum value, use the meme algorithm to detect change points in the observed data to obtain the set of change points and the corresponding fitness function value. S5. Based on the set of variable points and the corresponding fitness function values, calculate the rate of change of the fitness function for the current number of variable points, and increment the number of variable points by 1. S6. Repeat steps S4 and S5 until the number of variable points is greater than the given maximum value. S7. When the number of variables is greater than the given maximum value, determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables.

2. The method according to claim 1, characterized in that, The method for detecting change points in the observed data is as follows: S41. Initialize the population size in the meme algorithm as follows: The size of the newly generated population in the offspring is The number of population iterations is The crossover probability is The mutation probability is The probability of local fine-tuning is , obtain the initialization parameters; S42. Randomly generate the initial population based on the initialization parameters. ; S43, Set the current population iteration count. The current population is ; S44. Determine the current population iteration count. Is it less than the number of population iterations? ; S45. If the current population iteration count is... Less than the number of population iterations Based on crossover probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning. S46. Merge the current population and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population. The current population iteration count is incremented by 1; the new population becomes the current population for the next cycle. S47. Repeat steps S45 to S46 until the current population iteration count is reached. Greater than or equal to the number of population iterations ; S48. If the current population iteration count is... Greater than or equal to the number of population iterations The individual with the highest fitness function value in the current population is taken as the set of change points for the observation data.

3. The method according to claim 2, characterized in that, The initial population Include Individuals, each individual is composed of It consists of several variable points.

4. The method according to claim 1, characterized in that, The formula for calculating the rate of change of the fitness function is: ； In the formula, Indicates the number of variable points as The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points as The fitness function value corresponding to the set of change points obtained by the time-mode algorithm. Indicates the number of variable points as The fitness function value corresponding to the set of variable points obtained by the time-mode algorithm.

5. The method according to claim 4, characterized in that, The fitness function value is calculated as follows: ； In the formula, It is by The first of the observed data determined by the first change point The length of each segment It is the first The covariance of each segment, It is a regularization term. It is a regularization parameter. .

6. The method according to claim 5, characterized in that, The method for calculating covariance is as follows: ； in, ; In the formula, and They are the first The variable points at both ends of a segment, It is the first in the hydrological and meteorological time series observation data Data points, It is the first The mean of each segment.

7. The method according to claim 1, characterized in that, The method for determining the optimal number of variables and the set of variables is as follows: Based on the rate of change of the fitness function corresponding to different numbers of variables, the number of variables corresponding to the largest rate of change is selected as the optimal number of variables, and the set of variables corresponding to the optimal number of variables is the optimal set of variables.

8. The method according to claim 1, characterized in that, The acquired hydro-meteorological time series observation data are characterized by meteorological elements.

9. A hydrological and meteorological change point detection system, characterized in that, include: The data acquisition module is used to acquire hydrological and meteorological time series observation data; An initialization module, connected to the data acquisition module, is used to initialize the number of variable points in the observation data; The settings module is used to set a given maximum value; The judgment module, connected to the initialization module and the setting module, is used to determine whether the number of variable points is greater than a given maximum value; The detection module, connected to the judgment module, is used to detect changes in the observed data using the meme algorithm when the number of change points is less than or equal to a given maximum value, thereby obtaining a set of change points and the corresponding fitness function value. The calculation module, connected to the detection module, is used to calculate the rate of change of the fitness function under the current number of variable points based on the set of variable points and the corresponding fitness function values. The determination module, connected to the calculation module and the judgment module, is used to determine the optimal number of variables and the set of variables based on the rate of change of the fitness function corresponding to different numbers of variables when the number of variables is greater than a given maximum value.

10. The system according to claim 9, characterized in that, The detection module includes: The initialization submodule is used to initialize the population size. The size of the new population generated in the offspring Number of population iterations Crossover probability Probability of mutation Local fine-tuning probability This yields the initialization parameters and the current population iteration count. ; The generation submodule, connected to the initialization submodule, is used to randomly generate an initial population based on initialization parameters. and based on cross probability Probability of mutation Local fine-tuning probability For the current population New individuals are generated through crossover, mutation, and local fine-tuning, as well as by merging the current population. and the population composed of all new individuals The one with the highest fitness function value Individuals generate a new population; The determination submodule, connected to the initialization submodule and the generation submodule, is used to determine the current population iteration number. Is it less than the number of population iterations? ; The output submodule, connected to the generation submodule, is used to perform iterations in the current population. Greater than or equal to the number of population iterations When the individual with the highest fitness function value in the current population is selected as the set of change points for the observation data, the data is analyzed.

Images