Hydro-ecological state prediction method, device, equipment and medium

Abstract

The present disclosure provides a hydrological ecological state prediction method, device, equipment and medium, and relates to the technical field of data analysis. The method comprises: obtaining historical hydrological ecological data, and preprocessing the historical hydrological ecological data to obtain hydrological ecological optimization data; determining hydrological feature data based on the hydrological ecological optimization data, and determining the hydrological ecological state corresponding to the hydrological feature data; constructing a dynamic Markov chain model about time sequence based on the hydrological feature data and the hydrological ecological state; and taking the current hydrological ecological data and the current hydrological ecological state as the input of the dynamic Markov chain model to predict the hydrological ecological state at a target time. The technical solution in the present disclosure can improve the efficiency and accuracy of hydrological ecological state prediction.

Description

Technical Field

[0001] This disclosure relates to the field of data analysis technology, and more specifically, to a method, apparatus, equipment, and medium for predicting hydrological and ecological conditions. Background Technology

[0002] Hydrological and ecological changes are a complex natural process, and their patterns of change are often influenced by a variety of factors, such as climate, topography, geology, soil, and human activities. Due to the intertwined influence of these factors and their potential unpredictability, high-precision prediction of hydrological and ecological changes is a challenge.

[0003] In related technologies, statistical methods are typically used to analyze historical hydrological and ecological data, extract patterns of hydrological and ecological changes, and then use these patterns to predict future hydrological and ecological conditions. However, this method has poor prediction efficiency. Furthermore, the prediction results are highly limited by the continuity and representativeness of historical data. Due to the nonlinear characteristics of nature and the uncertainty of future environmental conditions, predictions based solely on past trends often fail to accurately grasp the evolutionary trends of hydrological and ecological conditions, resulting in poor prediction accuracy.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this disclosure is to provide a method, device, electronic device, and computer-readable storage medium for predicting hydrological and ecological states, thereby overcoming, to at least a certain extent, the problems of insufficient prediction efficiency and accuracy in related hydrological and ecological state prediction technologies.

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0007] According to a first aspect of the present disclosure, a method for predicting hydrological and ecological states is provided, comprising: acquiring historical hydrological and ecological data, and preprocessing the historical hydrological and ecological data to obtain optimized hydrological and ecological data; determining hydrological characteristic data based on the optimized hydrological and ecological data, and determining the hydrological and ecological state corresponding to the hydrological characteristic data; constructing a dynamic Markov chain model about a time series based on the hydrological characteristic data and the hydrological and ecological state; and using the current hydrological and ecological data as the input of the dynamic Markov chain model to predict the hydrological and ecological state at a target time.

[0008] In some example embodiments of this disclosure, based on the foregoing scheme, the preprocessing of the historical hydrological and ecological data to obtain optimized hydrological and ecological data includes: performing missing value detection and outlier detection on the historical hydrological and ecological data to determine missing values ​​and outliers in the historical hydrological and ecological data; filling the missing values ​​with mean interpolation and removing the outliers to obtain preliminary hydrological and ecological data; and standardizing and normalizing the preliminary hydrological and ecological data to obtain optimized hydrological and ecological data.

[0009] In some example embodiments of this disclosure, based on the foregoing scheme, the step of determining hydrological characteristic data based on the hydrological and ecological optimization data and determining the hydrological ecological state corresponding to the hydrological characteristic data includes: determining hydrological characteristic data based on the hydrological and ecological optimization data, wherein the hydrological characteristic data includes one or more of vegetation characteristic data, runoff characteristic data, soil moisture characteristic data, topographic characteristic data, and climate characteristic data; and determining the hydrological ecological state corresponding to the hydrological characteristic data based on a preset state standard.

[0010] In some example embodiments of this disclosure, based on the foregoing scheme, the step of constructing a dynamic Markov chain model about a time series based on the hydrological feature data and the hydrological ecological state includes: constructing a state transition probability matrix about the hydrological feature data and the time series based on the hydrological ecological state; constructing a dynamic Markov chain model based on the state transition probability matrix; and using the dynamic Markov chain model as the hydrological ecological prediction model.

[0011] In some example embodiments of this disclosure, based on the foregoing scheme, the step of constructing a dynamic Markov chain model based on the state transition probability matrix includes: determining an initial state transition probability matrix and optimizing the initial state transition probability matrix based on maximizing the likelihood function; constructing a dynamic Markov chain model based on the optimized state transition probability matrix and verifying the prediction performance of the dynamic Markov chain model based on a cross-validation method.

[0012] In some example embodiments of this disclosure, based on the foregoing scheme, the method for predicting hydrological and ecological status further includes: when the hydrological characteristic data is a single variable, predicting the hydrological and ecological status at the target time based on a combined model of an autoregressive integral moving average model and a long short-term memory network.

[0013] In some exemplary embodiments of this disclosure, based on the aforementioned scheme, the prediction of the hydrological and ecological state at a target time using the combined model of the autoregressive integral moving average model and the long short-term memory network includes: predicting the linear part of the hydrological and ecological state based on the autoregressive integral moving average model to obtain a linear prediction result and a model residual sequence; using the model residual sequence as input to the long short-term memory network to predict the nonlinear part of the hydrological and ecological state to obtain a nonlinear prediction result; and determining the hydrological and ecological state at the target time based on the sum of the linear prediction result and the nonlinear prediction result.

[0014] According to a second aspect of the present disclosure, a hydrological and ecological state prediction device is provided. The device includes: a data processing module, configured to acquire historical hydrological and ecological data and preprocess the historical hydrological and ecological data to obtain optimized hydrological and ecological data; a state determination module, configured to determine hydrological characteristic data based on the optimized hydrological and ecological data and determine the hydrological and ecological state corresponding to the hydrological characteristic data; a model building module, configured to construct a dynamic Markov chain model about a time series based on the hydrological characteristic data and the hydrological and ecological state; and a state prediction module, configured to use current hydrological and ecological data as input to the dynamic Markov chain model to predict the hydrological and ecological state at a target time.

[0015] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory storing computer-readable instructions, which, when executed by the processor, implement the above-described method for predicting hydrological and ecological states.

[0016] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for predicting hydrological and ecological states as described above.

[0017] The technical solutions provided in this disclosure may have the following beneficial effects:

[0018] The hydrological and ecological state prediction method in this exemplary embodiment has several advantages. First, by preprocessing historical hydrological and ecological data, data quality can be ensured, resulting in more representative optimized hydrological and ecological data. Second, by using a dynamic Markov chain model to predict the hydrological and ecological state, compared with the traditional static Markov model, the dynamic model can dynamically adjust the state transition probability according to the changes in the time series, better capturing the non-stationary characteristics of hydrological and ecological changes over time. Third, by combining the optimized hydrological feature data with the corresponding hydrological and ecological state, the constructed dynamic Markov chain model can comprehensively consider the impact of multiple factors on the hydrological and ecological system state. Fourth, based on the efficiency of the dynamic Markov chain model, even when processing large amounts of data and complex state transitions, it can maintain a high computing speed, which helps to quickly predict the hydrological and ecological state. Thus, it improves the efficiency and accuracy of hydrological and ecological state prediction to a certain extent.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0021] Figure 1 A schematic diagram of an exemplary system architecture for predicting hydrological and ecological states, to which embodiments of the present disclosure can be applied, is shown.

[0022] Figure 2 The illustration schematically shows a flowchart of a method for predicting hydrological and ecological conditions according to some embodiments of the present disclosure.

[0023] Figure 3 The illustration shows a flowchart of a process for predicting the hydro-ecological state at a target time based on a combined model according to some embodiments of the present disclosure.

[0024] Figure 4 The diagram illustrates a block diagram of a hydrological and ecological state prediction device according to some embodiments of the present disclosure.

[0025] Figure 5 The schematic diagram illustrates the structural schematic of a computer system of an electronic device according to some embodiments of the present disclosure.

[0026] Figure 6A schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure is shown.

[0027] In the accompanying drawings, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this specification as detailed in the appended claims.

[0029] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “the,” and “the” as used in this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0030] It should be understood that although the terms first, second, third, etc., may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this specification, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0031] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0032] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., may be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring aspects of this disclosure.

[0033] Furthermore, the accompanying drawings are for illustrative purposes only and are not necessarily drawn to scale. The block diagrams shown in the drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0034] In the context of current ecological environmental protection and sustainable development, hydrological and ecological status has become an important indicator for measuring the quality of the natural environment. However, the mechanisms of hydrological and ecological change are complex, involving the interaction of multiple factors, including but not limited to climate change, topography, geological structure, soil type, and the impact of human activities. The intertwined influences and potential unpredictability of these factors pose certain challenges to the accurate prediction of hydrological and ecological conditions.

[0035] In related technologies, statistical methods are typically used to analyze historical hydrological and ecological data, extract patterns of hydrological and ecological changes, and then use these patterns to predict future hydrological and ecological conditions. However, this approach has poor prediction efficiency. Furthermore, the predictive effectiveness is highly limited by the continuity and representativeness of historical data. Due to the nonlinear characteristics of nature and the uncertainty of future environmental conditions, predictions based solely on past trends often fail to accurately grasp the evolutionary trends of hydrological and ecological conditions, resulting in poor prediction accuracy. Some methods predict hydrological and ecological conditions by constructing physical models. However, the establishment of physical models requires detailed prior data, including high-resolution geographical, climatic, and soil parameters. The collection and integration of this data is arduous and costly. Moreover, the construction and solution process of physical models is highly complex, not only time-consuming and labor-intensive but also demanding on computational resources. The models also exhibit weak generalization ability and poor prediction efficiency.

[0036] To address all or part of the technical problems in the aforementioned related technologies, a method for predicting hydrological and ecological states is first proposed in the exemplary embodiments of this disclosure. Figure 1 A schematic diagram of an exemplary system architecture for predicting hydrological and ecological states, to which embodiments of the present disclosure can be applied, is shown.

[0037] like Figure 1As shown, system architecture 100 may include one or more of terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables. Terminal devices 101, 102, and 103 may be various devices used for collecting historical hydrological and ecological data, including but not limited to sensors, meteorological monitoring equipment, hydrological monitoring equipment, remote sensing satellites, drones, and other electronic devices with hydrological and ecological data acquisition capabilities. It should be understood that... Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0038] The hydrological and ecological state prediction method provided in this embodiment can be executed based on the aforementioned terminal devices and server. For example, historical hydrological and ecological data can be collected through preset terminal devices 101, 102, and 103. These terminal devices 101, 102, and 103 can then send this historical hydrological and ecological data to server 105 via network 104. Upon receiving the historical hydrological and ecological data, server 105 can first preprocess the data to obtain optimized hydrological and ecological data. Then, based on the optimized data, it can determine hydrological characteristic data and the corresponding hydrological and ecological state. Next, based on the hydrological characteristic data and the hydrological and ecological state, it can construct a dynamic Markov chain model for the time series. Finally, it uses the current hydrological and ecological data as input to the dynamic Markov chain model to predict the hydrological and ecological state at the target time, obtaining the prediction result. However, those skilled in the art will readily understand that the above application scenarios are merely illustrative and are not limited to this exemplary embodiment.

[0039] Figure 2 The illustration schematically shows a flowchart of a method for predicting hydrological and ecological states according to some embodiments of the present disclosure. Reference Figure 2 As shown, the method for predicting the hydrological and ecological state may include the following steps:

[0040] Step S210: Obtain historical hydrological and ecological data, and preprocess the historical hydrological and ecological data to obtain optimized hydrological and ecological data;

[0041] Step S220: Determine hydrological characteristic data based on hydrological and ecological optimization data, and determine the hydrological and ecological status corresponding to the hydrological characteristic data;

[0042] Step S230: Construct a dynamic Markov chain model about the time series based on hydrological characteristic data and hydrological ecological status;

[0043] Step S240: Use the current hydrological and ecological data as input to the dynamic Markov chain model to predict the hydrological and ecological state at the target time.

[0044] According to the hydrological and ecological state prediction method in the above example embodiments, on the one hand, by preprocessing historical hydrological and ecological data, data quality can be ensured, resulting in more representative optimized hydrological and ecological data; on the other hand, by using a dynamic Markov chain model to predict the hydrological and ecological state, compared with the traditional static Markov model, the dynamic model can dynamically adjust the state transition probability according to the changes in the time series, better capturing the non-stationary characteristics of hydrological and ecological changes over time; furthermore, by combining the optimized hydrological feature data with the corresponding hydrological and ecological state, the constructed dynamic Markov chain model can comprehensively consider the impact of multiple factors on the hydrological and ecological system state; and thirdly, based on the efficiency of the dynamic Markov chain model, even when processing large amounts of data and complex state transitions, it can maintain a high computing speed, which helps to quickly predict the hydrological and ecological state; thus, to a certain extent, it improves the efficiency and accuracy of hydrological and ecological state prediction.

[0045] The following will further explain the method for predicting the hydrological and ecological state in the example embodiments.

[0046] In step S210, historical hydrological and ecological data are acquired and preprocessed to obtain optimized hydrological and ecological data.

[0047] Historical hydrological and ecological data can represent a series of quantitative information records about changes in hydrology and the ecological environment accumulated over a period of time through various monitoring methods and technologies. Preprocessing refers to the data optimization steps performed before analyzing historical hydrological and ecological data.

[0048] In some embodiments, historical hydrological and ecological data may include meteorological data, hydrological data, surface data, and other suitable data such as human behavior data. For example, meteorological data may include temperature, precipitation, evaporation, relative humidity, wind speed and direction, etc.; hydrological data may include water flow velocity, water level, etc.; surface data may include land cover type (such as forest, grassland, cultivated land, water bodies, urbanized areas, etc.), vegetation cover, soil moisture, height, slope, aspect, near-infrared reflectance, and red light reflectance, etc.; and human behavior data may include water consumption, pollutant emissions, etc.

[0049] In some embodiments, meteorological and hydrological data can be acquired through sensing devices such as monitoring equipment at weather and hydrological stations, soil moisture meters, remote sensing satellites, and drones. Surface data can also be acquired via remote sensing satellites, and historical hydrological and ecological data can be obtained through other means such as publicly available datasets, meteorological websites, or hydrological websites. Furthermore, when using monitoring equipment or other devices for data collection, the collection frequency and location can be set, and the collection period can be any suitable time period, such as one hour, one day, or one month.

[0050] In some embodiments, preprocessing historical hydrological and ecological data to obtain optimized hydrological and ecological data may include the following steps: performing missing value detection and outlier detection on the historical hydrological and ecological data to identify missing and outlier values ​​in the historical hydrological and ecological data; filling in missing values ​​using the mean imputation method and removing outlier values ​​to obtain preliminary hydrological and ecological data; and standardizing and normalizing the preliminary hydrological and ecological data to obtain optimized hydrological and ecological data.

[0051] Missing value detection refers to the process of identifying missing or unrecorded data in historical hydrological and ecological data. Outlier detection refers to identifying data in historical hydrological and ecological data that are significantly different from most other hydrological and ecological data, which may be due to measurement errors, input errors, or other anomalies. Mean imputation refers to replacing all missing values ​​of a variable with the average of all non-missing observations of that variable.

[0052] Specifically, outlier detection can be performed using other suitable methods such as the Z-score (Standard Score), the Interquartile Range Rule (IQR), and density estimation.

[0053] For example, outlier detection using the Z-score method can be performed by first calculating the sample mean μ and sample standard deviation σ for each variable that needs outlier detection, such as temperature or precipitation. Then, the outlier is detected using the following equation (1). z Value calculation:

[0054] (1)

[0055] in, σ represents the actual observed value of the variable, μ represents the sample mean of the variable, and σ represents the sample standard deviation of the variable.

[0056] Next, set a judgment threshold. For example, the judgment threshold can be 3, indicating that when... z When the absolute value of the value is greater than 3, the value can be...z The observed values ​​corresponding to the values ​​are marked as outliers; finally, these outliers are removed.

[0057] Alternatively, mean imputation of missing values ​​can be performed using the following equation (2):

[0058] (2)

[0059] in, This represents the sum of all non-missing values ​​in a specific variable. n This indicates the number of non-missing values.

[0060] After imputing missing values ​​and removing outliers from historical hydrological and ecological data, preliminary hydrological and ecological screening data can be obtained. Then, the hydrological and ecological data can be standardized and normalized to ensure consistent scales for different features, adjust data distribution, eliminate unnecessary biases, and improve the training efficiency and prediction accuracy of the model in subsequent processes.

[0061] Specifically, the preliminary hydrological and ecological screening data can be normalized using the following equation (3) to scale the data to the [0,1] interval:

[0062] (3)

[0063] in, This represents the standardized data value. Represents the original data value. This represents the maximum value among the variables. This represents the minimum value among the variables.

[0064] The hydrological and ecological screening data can be normalized by the following formula (4) to convert them into a standard normal distribution, so that the data have a mean of 0 and a standard deviation of 1.

[0065] (4)

[0066] in, This represents the normalized data value. Represents the original data value. u This represents the average value of the variable. σ The standard deviation of a variable.

[0067] In step S220, hydrological characteristic data are determined based on hydrological and ecological optimization data, and the hydrological and ecological status corresponding to the hydrological characteristic data is determined.

[0068] Hydrological characteristic data can represent features that significantly influence changes in the hydrological and ecological state, selected or constructed based on the principles of hydrological ecology, such as vegetation characteristics, runoff characteristics, soil moisture characteristics, and other suitable features. The hydrological and ecological state can represent the comprehensive condition of a target river and its surrounding ecosystem within a specific time period. For example, the hydrological and ecological state can include precipitation status, river flow status, vegetation cover status, soil moisture status, and water quality status. Specifically, precipitation status can include drought, normal, and rainy; river flow status can include low water level, normal water level, and high water level; vegetation cover status can include poor, moderate cover, and lush; soil moisture status can include dry, suitable, and moist; and water quality status can include excellent, good, and poor.

[0069] In some embodiments, hydrological characteristic data is determined based on hydrological and ecological optimization data, and the corresponding hydrological and ecological state is determined. Specifically, the following steps are included: determining hydrological characteristic data based on hydrological and ecological optimization data, wherein the hydrological characteristic data includes one or more of vegetation characteristic data, runoff characteristic data, soil moisture characteristic data, topographic characteristic data, and climate characteristic data; and determining the hydrological and ecological state corresponding to the hydrological characteristic data based on a preset state standard.

[0070] Specifically, vegetation characteristic data can be represented by the Normalized Difference Vegetation Index (NDVI). NDVI is a vegetation index based on near-infrared and red light reflectance, used to quantify vegetation cover density in a specific area, in order to assess the health status and coverage of vegetation. Specifically, the Normalized Difference Vegetation Index can be calculated using the following equation (5):

[0071] (5)

[0072] in, The near-infrared reflectance of multispectral remote sensing images. This refers to the red band reflectance of the multispectral remote sensing image. For example, determining the vegetation cover status corresponding to the vegetation feature data based on a preset vegetation status standard can be achieved when... NDVI When the value is less than 0.1, the vegetation cover is considered poor. NDVI When the value is greater than 0.1 and less than 0.5, the vegetation cover status is medium cover. NDVI When the value is greater than 0.5, the vegetation cover is considered lush.

[0073] Furthermore, runoff characteristic data can be obtained by calculating the amount of water flowing through a certain cross-section of a river within a certain time period. Specifically, runoff can be calculated using the following equation (6):

[0074] (6)

[0075] Wherein, Q represents runoff in cubic meters per second; P represents precipitation in millimeters; I represents initial loss, which is evaporation and vegetation interception before runoff occurs, in millimeters; Ea represents actual evapotranspiration, including evaporation and plant transpiration, in millimeters; Ei represents deep infiltration, i.e., water exceeding the depth of plant roots, in millimeters; R represents reservoir capacity change in millimeters; G represents aquifer discharge or inflow, with positive values ​​indicating inflow and negative values ​​indicating outflow, in millimeters; and A represents catchment area in square meters. For example, for small rivers, when Q is less than 20, the corresponding river flow status is low; when Q is greater than 20 and less than 80, the corresponding river flow status is normal; and when Q is greater than 80, the corresponding river flow status is high. Of course, in other embodiments of this disclosure, different river flow status standards can be set for different rivers.

[0076] Furthermore, soil moisture characteristics can be represented by water content, which can be collected by hydrological monitoring sensors installed on the ground to reflect the soil's ability to absorb and evaporate water. Specifically, soil water content W can be calculated using the following formula (7):

[0077] (7)

[0078] Where Mw can represent the mass of water in the soil, and M can represent the mass of dry soil.

[0079] After obtaining the soil moisture content, the soil moisture status of the target river area can be determined according to the preset soil moisture status standard.

[0080] In addition, topographic feature data can be represented by topographic data such as elevation, slope, and aspect of the target river area, which can be obtained through remote sensing technology. Climate feature data can be represented by climate data such as temperature, rainfall, and evaporation, which can be obtained from meteorological station records or satellite remote sensing data.

[0081] Next, referring to 2, in step S230, a dynamic Markov chain model about the time series is constructed based on hydrological characteristic data and hydrological ecological status.

[0082] The Dynamic Markov Chain (DMC) model represents a model that allows state transition probabilities to change over time, essentially an extension of the basic Markov chain. In a basic Markov chain, the transition probabilities between states are fixed; that is, the probability of moving from one state to another does not change over time. In a DMC, however, the state transition probabilities can be a function of time, reflecting the dynamic characteristics of the system's evolution over time. In hydrological and ecological state prediction, precipitation, runoff, climate, topography, and vegetation cover may change over time. Using a DMC model based on time series data, the patterns and trends of hydrological and ecological state evolution over time can be better predicted.

[0083] In one example embodiment, a dynamic Markov chain model based on hydrological feature data and hydrological ecological status is constructed based on time series data. The specific steps include: constructing a state transition probability matrix based on hydrological feature data and time series data based on hydrological ecological status; constructing a dynamic Markov chain model based on the state transition probability matrix; and using the dynamic Markov chain model as a hydrological ecological prediction model.

[0084] The state transition probability matrix can be used to represent a square matrix of transition probabilities between different states in a dynamic Markov chain. Specifically, the state transition probability matrix S of the dynamic Markov chain model can be defined. Based on collected historical hydrological and ecological data and their corresponding hydrological and ecological states, the state transition probability matrix of the model, i.e., the set of possible states, is determined. For example, when there exists... n When there are several hydrological and ecological states, the state transition probability matrix can be defined as S = {S1, S2, ... S...} n}

[0085] The probability of transition between different hydrological and ecological states is calculated based on historical hydrological and ecological data. n A Markov chain representing a hydrological and ecological state can be obtained. n × n The state transition probability matrix. In time... t From the state i Transition to state j The probability can be used It indicates that it is time. t The function is assumed to be influenced by environmental factors, namely historical hydrological and ecological data. E The impact. Therefore, It can be represented as That is, the state transition probability is E ( t )and t The function, where, The elements in the matrix can represent the probability of transitioning from state i to state j. The following conditions must be met:

[0086]

[0087]

[0088] Where Σ represents summation, This means that the condition is satisfied for all possible j.

[0089] At a specific time t And environmental factors, namely historical hydrological and ecological data E ( t Under these conditions, the model operates at a certain time. t From state i Transition to state j The probability is A state transition matrix is ​​used to represent all state transition probabilities. Each row of the matrix corresponds to a state, and each column corresponds to a time point. The element in the i-th row and j-th column of the matrix is... The dynamic Markov chain model constructed based on the state transition probability matrix can then be expressed as:

[0090] In one example embodiment, constructing a dynamic Markov chain model based on the state transition probability matrix includes the following steps: determining the initial state transition probability matrix and optimizing the initial state transition probability matrix based on maximizing the likelihood function; constructing a dynamic Markov chain model based on the optimized state transition probability matrix and verifying the prediction performance of the dynamic Markov chain model based on the cross-validation method.

[0091] Maximizing the likelihood (MLE) function is a commonly used parameter estimation method. Its goal is to find a set of parameters that maximizes the probability of the observed data occurring. Specifically, in a dynamic Markov chain model, for a given set of observed state sequences, maximizing the likelihood function means adjusting the parameters in the state transition probability matrix to maximize the probability of that state sequence occurring based on these parameters. Cross-validation can be used to estimate the model's performance on unknown data, avoiding overfitting.

[0092] Specifically, for a dynamic Markov chain model, the parameter that needs to be initialized is the state transition probability matrix. Each element in the transition probability matrix can be initialized to an equal probability or set with an initial value based on some known prior knowledge. For example, if the transitions between all states are equally likely, then each element in the state transition probability matrix can be initialized to 1 / n (where n is the number of states). Furthermore, if there is prior knowledge of some state transition probabilities, the initial value can be set based on that prior knowledge, and the sum of each row of the state transition probability matrix is ​​1.

[0093] For dynamic Markov chain models, the data used are historical hydrological and ecological data and their corresponding hydrological and ecological states. Parameter learning can be performed by maximizing the model's likelihood function, which can be... ,in, Indicates model parameters, X Represents observation data, L Let represent the likelihood function. By solving this likelihood function, we can obtain a set of model parameters that maximize the probability of observing actual data under these parameters.

[0094] Specifically, for the data {X1, X2, ... X... T The likelihood function of a dynamic Markov chain model can be expressed as: Here, P represents the parameters in the dynamic Markov chain model, i.e., the elements in the state transition probability matrix. The goal is to find a set of parameters. This makes the likelihood function L reach its maximum.

[0095] Furthermore, the predictive performance of the dynamic Markov chain model can be validated using cross-validation. This can be achieved by using historical hydrological and ecological data and their corresponding hydrological and ecological states as the training set, and then dividing the training set into... k A subset can be set k Use other suitable values ​​such as 5, 10, or 15. Each time you use it... k -1 subset is used for training, and the remaining subset is used for validation; loop. k Each time, a different subset of validations is selected, resulting in... k The average of the model performance evaluation results is taken as the final evaluation result. For each validation, the model's goodness of fit and prediction error, such as mean squared error (MSE), can be calculated to obtain... k The mean square error results are then calculated using the following equation (8), where, This represents the average value of the mean square error.

[0096] (8)

[0097] The standard deviation of the mean square error can be calculated using the following equation (9), where, It can represent the standard deviation of the mean square error.

[0098] (9)

[0099] The mean and standard deviation of the mean squared error (MSE) can be used to represent the predictive performance and stability of a dynamic Markov chain model. A smaller mean MSE indicates better predictive performance, while a smaller standard deviation indicates better model stability.

[0100] In some embodiments, when training a dynamic Markov chain model, the number of iterations and the learning rate can be appropriately increased. Specifically, during model training, the model may learn slowly, failing to achieve the expected prediction accuracy within a pre-set number of iterations. In this case, the number of iterations can be appropriately increased, allowing the model to learn and adjust more times, thereby improving the model's learning performance. The learning rate is the step size of the model's learning process. An excessively large learning rate may cause the model to oscillate during the learning process, affecting the stability of predictions; an excessively small learning rate may cause the model to learn too slowly. In such cases, the learning rate can be appropriately adjusted, selecting a suitable learning rate based on the performance on the validation set.

[0101] Next, refer to Figure 2 In step S240, the current hydrological and ecological data is used as the input to the dynamic Markov chain model to predict the hydrological and ecological state at the target time.

[0102] The current hydrological and ecological data can represent current meteorological data, hydrological data, surface data, human behavior data, and other suitable hydrological and ecological data for the target river area. Using this current hydrological and ecological data as input to a dynamic Markov chain model, and setting a prediction time point, the model predicts the probability distribution of the target river area at that time point based on the current hydrological and ecological data. The hydrological and ecological state at that time point is then predicted based on the state with the highest probability distribution.

[0103] Furthermore, after predicting the hydrological and ecological state at a specific point in the future, the prediction results can be analyzed and interpreted. Based on the analysis results, potential problems in the future hydrological and ecological state can be predicted, such as whether there will be water scarcity or the risk of floods. Corresponding countermeasures can be formulated based on these analysis results, such as preparing for flood prevention in advance and improving irrigation systems, so as to transform the model prediction results into specific applications and provide support for decision-making.

[0104] In one example embodiment, the method for predicting hydrological and ecological status may further include the following steps: when the hydrological characteristic data is a single variable, the hydrological and ecological status at the target time is predicted based on a combined model of an autoregressive integral moving average model and a long short-term memory network.

[0105] Specifically, hydrological characteristic data with a single variable can be any type of topographic characteristic data, such as altitude, slope, and aspect, or any type of climate characteristic data, such as temperature, rainfall, and evaporation. Furthermore, the hydrological ecological state corresponds to the hydrological characteristic data. For example, when the hydrological characteristic data is rainfall, the hydrological ecological state can be rainfall intensity; when the hydrological characteristic data is evaporation, the hydrological ecological state can be soil moisture status.

[0106] Among them, the Autoregressive Integrated Moving Average (ARIMA) model is suitable for linear time series data and can be used to capture the trend of linear changes in hydrological and ecological states and predict its linear component. Long Short-Term Memory (LSTM) networks are suitable for handling nonlinear relationships in time series and long-term memory tasks. By combining the linear prediction results of the ARIMA model with the nonlinear prediction results of the LSTM model, a comprehensive prediction of the hydrological and ecological state at the target time can be obtained. This combines the advantages of both models, considering both the linear trend of the data and capturing potential nonlinear complex relationships, thus improving the accuracy and robustness of the prediction.

[0107] Furthermore, for multivariate hydrological characteristic data, the hydrological and ecological state can be predicted using a dynamic Markov chain model. For single-variable hydrological characteristic data, the hydrological and ecological state at a target time can be predicted using a combination model of autoregressive integral moving average and long short-term memory network. Using different models for different types of hydrological characteristic data can significantly improve the accuracy of prediction and the practicality of the model.

[0108] In one example embodiment, reference Figure 3 As shown, the prediction of the hydrological and ecological state at a target time based on a combined model of autoregressive integral moving average and long short-term memory network includes the following steps:

[0109] Step S310: Based on the autoregressive integral moving average model, predict the linear part of the hydrological and ecological state to obtain the linear prediction results and the model residual sequence.

[0110] Specifically, the difference order *d*, the order of autoregression *p*, and the order of the moving average *q* in the autoregressive integral moving average model are first determined. Based on the determined orders (p, d, q), an ARIMA(p, d, q) model is constructed and trained using historical data to learn the linear trend and autocorrelation of the data. The trained ARIMA model is then used to linearly predict the hydrological and ecological state at a future target time. Simultaneously, the model residual sequence generated during the prediction process is recorded, reflecting the nonlinear components in the data and fluctuations not captured by the model.

[0111] Step S320: The model residual sequence is used as the input of the long short-term memory network to predict the nonlinear part of the hydrological and ecological state, and the nonlinear prediction result is obtained.

[0112] Specifically, an LSTM network architecture is predefined, including an input layer, several LSTM hidden layers, and an output layer. The preprocessed residual sequence is divided into a training set and a validation set. The LSTM network is trained using the training set data, and the network weights are adjusted through backpropagation and optimization algorithms to minimize the prediction error. Simultaneously, the validation set is used for verification to prevent overfitting. Then, the trained LSTM model is used to predict the residual sequence, obtaining the prediction results of the nonlinear component of the hydrological and ecological state, thereby capturing the complex dynamics and nonlinear patterns in the residuals.

[0113] Step S330: Determine the hydrological and ecological state at the target time based on the sum of the linear and nonlinear prediction results. Specifically, the linear and nonlinear prediction results can be weighted and summed, and the combined value of the linear and nonlinear prediction results can be used as the final predicted value of the hydrological and ecological state at the target time.

[0114] The hydrological and ecological state prediction method in the above example embodiments has several advantages. First, by preprocessing historical hydrological and ecological data, data quality can be ensured, resulting in more representative optimized hydrological and ecological data. Second, by using a dynamic Markov chain model to predict the hydrological and ecological state, compared with the traditional static Markov model, the dynamic model can dynamically adjust the state transition probability according to the changes in the time series, better capturing the non-stationary characteristics of hydrological and ecological changes over time. Third, by combining the optimized hydrological feature data with the corresponding hydrological and ecological state, the constructed dynamic Markov chain model can comprehensively consider the impact of multiple factors on the hydrological and ecological system state. Fourth, based on the efficiency of the dynamic Markov chain model, even when processing large amounts of data and complex state transitions, it can maintain a high computing speed, which helps to quickly predict the hydrological and ecological state. Thus, it improves the efficiency and accuracy of hydrological and ecological state prediction to a certain extent.

[0115] It should be noted that although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0116] Next, in this embodiment of the disclosure, a device for predicting hydrological and ecological states is also provided, with reference to... Figure 4 As shown, the hydrological and ecological state prediction device 400 can be composed of a data processing module 401, a state determination module 402, a model building module 403, and a state prediction module 404. Specifically: the data processing module 401 can acquire historical hydrological and ecological data and preprocess it to obtain optimized hydrological and ecological data; the state determination module 402 can determine hydrological characteristic data based on the optimized hydrological and ecological data and determine the corresponding hydrological and ecological state; the model building module 403 can construct a dynamic Markov chain model about the time series based on the hydrological characteristic data and the hydrological and ecological state; and the state prediction module 404 can use the current hydrological and ecological data and the current hydrological and ecological state as inputs to the dynamic Markov chain model to predict the hydrological and ecological state at the target time.

[0117] It should be noted that the specific details of each part of the above-mentioned hydrological and ecological state prediction device have been described in detail in the implementation method section of the hydrological and ecological state prediction method. For any undisclosed details, please refer to the implementation method section of the method section, and therefore will not be repeated here.

[0118] Furthermore, in an exemplary embodiment of this disclosure, an electronic device capable of implementing the above-described method for predicting hydrological and ecological states is also provided.

[0119] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be embodied in the following forms: a completely hardware embodiment, a completely software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0120] The following reference Figure 5 To describe an electronic device 500 according to such an embodiment of the present disclosure. Figure 5 The electronic device 500 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0121] like Figure 5As shown, the electronic device 500 is presented in the form of a general-purpose computing device. The components of the electronic device 500 may include, but are not limited to: at least one processing unit 510, at least one storage unit 520, a bus 530 connecting different system components (including storage unit 520 and processing unit 510), and a display unit 540.

[0122] The storage unit stores program code, which can be executed by the processing unit 510 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 510 can perform actions such as... Figure 2 Step S210 shows the acquisition of historical hydrological and ecological data, and the preprocessing of the historical hydrological and ecological data to obtain optimized hydrological and ecological data; Step S220 shows the determination of hydrological characteristic data based on the optimized hydrological and ecological data, and the determination of the hydrological and ecological state corresponding to the hydrological characteristic data; Step S230 shows the construction of a dynamic Markov chain model about the time series based on the hydrological characteristic data and the hydrological and ecological state; Step S240 shows the prediction of the hydrological and ecological state at the target time by using the current hydrological and ecological data as the input of the dynamic Markov chain model.

[0123] Storage unit 520 may include readable media in the form of volatile storage units, such as random access memory (RAM) 521 and / or cache memory 522, and may further include read-only memory (ROM) 523.

[0124] Storage unit 520 may also include a program / utility 524 having a set (at least one) program module 525, such program module 525 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0125] Bus 530 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0126] Electronic device 500 can also communicate with one or more external devices 570 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 500, and / or with any device that enables electronic device 500 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 550. Furthermore, electronic device 500 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 560. As shown, network adapter 560 communicates with other modules of electronic device 500 via bus 530. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0127] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0128] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the methods described above is stored. In some possible embodiments, various aspects of this disclosure may also be implemented as a program product including program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0129] refer to Figure 6 As shown, a program product 600 for implementing the above-described method for predicting hydrological and ecological states according to embodiments of the present disclosure is described. This product may be a portable compact disk read-only memory (CD-ROM) and includes program code, and may run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, the readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0130] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0131] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0132] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, electromagnetic waves, etc., or any suitable combination thereof.

[0133] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0134] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0135] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0136] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0137] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for predicting hydrological and ecological conditions, characterized in that, include: Historical hydrological and ecological data are acquired and preprocessed to obtain optimized hydrological and ecological data. Based on the aforementioned hydrological and ecological optimization data, hydrological characteristic data are determined, and the corresponding hydrological and ecological status is determined. Based on the hydrological and ecological state, a state transition probability matrix is ​​constructed for the hydrological feature data and time series. An initial state transition probability matrix is ​​determined and optimized based on maximizing the likelihood function; a dynamic Markov chain model is constructed based on the optimized state transition probability matrix, and the prediction performance of the dynamic Markov chain model is verified based on the cross-validation method. The current hydrological and ecological data are used as input to the dynamic Markov chain model to predict the hydrological and ecological state at the target time. When the hydrological characteristic data is a single variable, the linear part of the hydrological ecological state is predicted based on the autoregressive integral moving average model to obtain the linear prediction result and the model residual sequence; the model residual sequence is used as the input of the long short-term memory network to predict the nonlinear part of the hydrological ecological state to obtain the nonlinear prediction result. The hydrological and ecological state at the target time is determined based on the sum of the linear and nonlinear prediction results.

2. The method for predicting hydrological and ecological states according to claim 1, characterized in that, The preprocessing of the historical hydrological and ecological data to obtain optimized hydrological and ecological data includes: Missing value detection and outlier detection are performed on the historical hydrological and ecological data to identify missing and outlier values ​​in the historical hydrological and ecological data. The missing values ​​were filled using the mean interpolation method, and the outliers were removed to obtain the initial hydrological and ecological screening data. The preliminary hydrological and ecological data are standardized and normalized to obtain the optimized hydrological and ecological data.

3. The method for predicting hydrological and ecological states according to claim 1, characterized in that, The step of determining hydrological characteristic data based on the hydrological and ecological optimization data, and determining the hydrological and ecological state corresponding to the hydrological characteristic data, includes: Hydrological characteristic data are determined based on the aforementioned hydrological and ecological optimization data. The hydrological characteristic data includes one or more of the following: vegetation characteristic data, runoff characteristic data, soil moisture characteristic data, topographic characteristic data, and climate characteristic data. Based on preset state standards, the hydrological ecological state corresponding to the hydrological characteristic data is determined.

4. A device for predicting hydrological and ecological states, used to implement the method for predicting hydrological and ecological states according to any one of claims 1 to 3, characterized in that, The device includes: The data processing module is used to acquire historical hydrological and ecological data and preprocess the historical hydrological and ecological data to obtain optimized hydrological and ecological data. The state determination module is used to determine hydrological characteristic data based on the hydrological and ecological optimization data, and to determine the hydrological and ecological state corresponding to the hydrological characteristic data. The model building module is used to construct a dynamic Markov chain model about the time series based on the hydrological feature data and the hydrological ecological state. The state prediction module is used to use the current hydrological and ecological data as input to the dynamic Markov chain model to predict the hydrological and ecological state at the target time.

5. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method for predicting the hydrological and ecological state according to any one of claims 1-3 by executing the executable instructions.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for predicting the hydrological and ecological state as described in any one of claims 1-3.

Images