A monitoring and updating method and device for ecological flow data

Abstract

The present application relates to the field of ecological data monitoring technology, and in particular to a monitoring and updating method and device for ecological flow data, which collects stress indicators related to ecological flow balance; based on the pre-deployed stress update log, standardizes and integrates the stress indicators, and outputs the standardized and integrated stress indicators; deploys a monitoring and updating model structure for ecological flow data based on the stress indicators and the dependent associations between the stress indicators; determines a monitoring and updating model for ecological flow data based on the standardized and integrated stress indicators and the monitoring and updating model structure for ecological flow data; collects the balance stress data of the ecological flow to be updated, and outputs the balance stress update result of the ecological flow to be updated in combination with the monitoring and updating model for ecological flow data. The present invention ensures data consistency, improves the efficiency of data updating and the management and traceability of the updating process, and provides a scientific basis for ecosystem management and protection.

Description

Technical Field

[0001] The present application relates to the technical field of ecological data monitoring, and in particular to a monitoring and updating method and device for ecological flow data. Background Art

[0002] Ecological data is a complex, open, dynamic, non-equilibrium and non-linear data. Some studies on the stability of ecological data only focus on the biological part of ecological data (such as taxonomy, biodiversity, etc.), and some consider both biological and non-biological parts, but only through statistical means (simply linking the two. In a complex and changing environment, biological and non-biological parts are coupled together through non-linear phase parameters, and due to the openness of ecological data, the two themselves often have complex structures. Therefore, previous methods cannot objectively and accurately monitor the stability of ecological data.

[0003] In order to truly describe the accuracy of ecological data across data, space and time, and to scientifically evaluate and predict the accuracy of ecological data, it is necessary to construct an ecological data accuracy monitoring model based on machine learning methods. In addition to considering the importance of each species indicator of ecological data itself, the topological structure and accuracy indicators of the interspecies interaction network will also be considered. The accuracy of ecological data will be monitored and predicted and incorporated into the model, which will greatly reduce the subjectivity of previous monitoring methods, overcome the incomparability of the accuracy of ecological data with different differences, and make the monitoring results of ecological data accuracy more accurate.

[0004] In the existing methods for monitoring the accuracy of ecological data, increasing the amount of data in the model will increase the difficulty of processing, which has certain limitations. Some ecological data indicator data variables have a greater impact on the accuracy of ecological data, and certain selection is required to achieve low-cost and high-accuracy ecological data accuracy monitoring. In addition, for traditional ecological data monitoring, there may only be manual methods for monitoring, which will result in excessive costs, and the data used for manual monitoring has large errors and low accuracy. Summary of the invention

[0005] To solve the above problems, this application provides the following technical solutions:

[0006] According to a first aspect of the present invention, the present invention claims protection for a monitoring and updating method for ecological flow data, comprising:

[0007] Collect stress indicators related to ecological flow balance;

[0008] Based on the pre-deployed coercion update log, the coercion indicators are standardized and integrated, and the standardized and integrated coercion indicators are output; wherein the pre-deployed coercion update log is used to illustrate the correlation level and existence possibility level of the existence degree of each coercion indicator;

[0009] Based on the stress indicators and the dependencies between them, a monitoring and updating model structure for ecological flow data is deployed;

[0010] Determining a monitoring and updating model for ecological flow data based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data;

[0011] The balance stress data of the ecological flow to be updated are collected, and based on the balance stress data and the monitoring and updating model for the ecological flow data, the balance stress update result of the ecological flow to be updated is output.

[0012] Furthermore, the monitoring and updating model for ecological flow data is an LSTM model algorithm;

[0013] The balance stress data of the ecological flow to be updated are stress indicators associated with the balance of the ecological flow to be updated and / or abnormal types of the ecological flow to be updated;

[0014] The monitoring and updating model based on the balance stress data and the ecological flow data outputs the balance stress update result of the ecological flow to be updated, including:

[0015] Based on the pre-deployed stress update log, the balance stress data is standardized and integrated, and the standardized and integrated balance stress data is output;

[0016] Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated; wherein the balance stress scenario includes a deep stress scenario, a general stress scenario and a balance scenario;

[0017] And / or, the abnormal type of the ecological flow to be updated after standardized integration is sent to the LSTM model algorithm to determine the coercion indicator of the abnormal type of the ecological flow to be updated.

[0018] Furthermore, the standardized integrated stress index associated with the ecological flow balance to be updated is sent to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated, including:

[0019] Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm, and outputting the possibility distribution of the equilibrium stress scenario corresponding to each stress indicator based on hypothesis deduction analysis;

[0020] If the probability value of the deep stress scene in the balanced stress scene corresponding to any stress indicator exceeds the preset stress threshold, it is determined that the stress indicator is in the deep stress scene;

[0021] If the possibility values ​​of the deep stress scenes in the balanced stress scenes corresponding to the stress indicators do not exceed the preset stress thresholds, it is determined that the stress indicators are not in the deep stress scenes.

[0022] Furthermore, the abnormal type of the ecological flow to be updated after standardized integration is sent to the LSTM model algorithm to determine whether the ecological flow to be updated has a stress indicator of the abnormal type, including:

[0023] Based on the abnormal type of the ecological traffic to be updated after standardized integration, in the output network element of the LSTM model algorithm, the possibility distribution of the abnormal type is adjusted to a deep stress scenario, and based on inductive analysis, the possibility distribution of the balanced stress scenario corresponding to each stress indicator is determined, and the stress indicator whose possibility value of the deep stress scenario exceeds the preset stress threshold is used as a stress indicator of the existence of this abnormal type.

[0024] Furthermore, the standardization and integration of the coercion indicators based on the pre-deployed coercion update log includes:

[0025] Based on the pre-deployed coercion update log, determine the correlation level and existence possibility level of the existence degree corresponding to each coercion indicator;

[0026] Based on the correlation and existence possibility of the existence degree corresponding to each stress indicator, the equilibrium stress level corresponding to each stress indicator is determined;

[0027] Based on the equilibrium stress level corresponding to each stress indicator, each stress indicator is standardized and integrated.

[0028] Furthermore, based on the stress indicators and the dependencies between the stress indicators, a monitoring and updating model structure for ecological flow data is deployed, including:

[0029] Based on the stress indicator, deploy a stress indicator dataset;

[0030] Each stress indicator is used as a network element variable, and the dependency relationship between the stress indicators is determined, and the network element variables are connected to output multiple queues;

[0031] The satisfaction degree between each queue and the stress indicator dataset is calculated, and the queue with the largest satisfaction degree is used as the monitoring update model structure for ecological flow data.

[0032] Furthermore, the monitoring and updating model for ecological flow data is determined based on the standardized integrated stress index and the monitoring and updating model structure for ecological flow data, including:

[0033] Determine the probability distribution corresponding to each stress indicator based on the monitoring and updating model structure for ecological flow data;

[0034] The monitoring and updating model structure for ecological flow data is taken as a parameter learning object, and the standardized and integrated stress indicators are imported into the monitoring and updating model structure for ecological flow data. Parameter learning is used to correct the monitoring and updating model structure for ecological flow data to determine the monitoring and updating model for ecological flow data.

[0035] According to a second aspect of the present invention, the present invention claims protection for a monitoring and updating device for ecological flow data, comprising:

[0036] Collection unit, used to collect stress indicators related to ecological flow balance;

[0037] An integration unit, configured to standardize and integrate the coercion indicators based on a pre-deployed coercion update log, and output the standardized and integrated coercion indicators; wherein the pre-deployed coercion update log is used to illustrate the correlation level and existence possibility level of the existence degree of each coercion indicator;

[0038] A deployment unit, used for deploying a monitoring and updating model structure for ecological flow data based on the stress indicators and the dependencies between the stress indicators;

[0039] The deployment unit is further used to determine a monitoring and updating model for ecological flow data based on the standardized integrated stress index and the monitoring and updating model structure for ecological flow data;

[0040] An updating unit, used for collecting the balance stress data of the ecological flow to be updated, and outputting the balance stress update result of the ecological flow to be updated based on the balance stress data and the monitoring and updating model for ecological flow data;

[0041] The monitoring and updating device for ecological flow data is used to execute the monitoring and updating method for ecological flow data.

[0042] The present application relates to the field of ecological data monitoring technology, and in particular to a monitoring and updating method and device for ecological flow data, which collects stress indicators related to ecological flow balance; based on the pre-deployed stress update log, standardizes and integrates the stress indicators, and outputs the standardized and integrated stress indicators; deploys a monitoring and updating model structure for ecological flow data based on the stress indicators and the dependent associations between the stress indicators; determines a monitoring and updating model for ecological flow data based on the standardized and integrated stress indicators and the monitoring and updating model structure for ecological flow data; collects the balance stress data of the ecological flow to be updated, and outputs the balance stress update result of the ecological flow to be updated in combination with the monitoring and updating model for ecological flow data. The present invention ensures data consistency, improves the efficiency of data update and the management and traceability of the update process, and provides a scientific basis for ecosystem management and protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A workflow diagram of a monitoring and updating method for ecological flow data as claimed in an embodiment of the present application;

[0044] Figure 2 This is a structural module diagram of a monitoring and updating device for ecological flow data as claimed in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0047] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0048] Figure 1 FIG. 1 is a flowchart of an implementation method for monitoring and updating ecological flow data provided by an embodiment of the present invention. Figure 1 As shown, the method includes:

[0049] Step 110: Collect stress indicators related to ecological flow balance.

[0050] In this embodiment, the research results on the reasons for the reduction of ecological water storage containers can be analyzed and reviewed, and combined with expert experience, stress indicators directly or indirectly related to the ecological flow balance can be determined. Specifically, the stress indicators related to the ecological flow balance can include the effects of incorrect water use, poor dam repair, flow and water changes, heavy rainfall, abnormal aquatic plants and animals, etc., which are not limited here.

[0051] Step 120: Based on the pre-deployed coercion update log, the coercion indicators are standardized and integrated, and the standardized and integrated coercion indicators are output; wherein the pre-deployed coercion update log is used to illustrate the correlation level and existence possibility level of the existence degree of each coercion indicator.

[0052] In this embodiment, when monitoring and updating ecological flow data, it is necessary to make a comprehensive judgment based on the possibility of the existence of coercion and the order of the degree of existence.

[0053] Exemplarily, the possibility of coercion existence is determined according to data value or state value standards, and the degree of existence is determined according to correlation, and a result strategy for the possibility of coercion element (indicator) existence and a result strategy for the degree of coercion indicator existence can be output.

[0054] For any coercion indicator, if the coercion indicator is greater than 0.7 according to the data value identification standard, or the status value is identified and revised as often existing, then the possibility level of the coercion indicator is level 5; if the coercion indicator is between 0.07 and 0.7 according to the data value identification standard, or the status value is identified and revised as possibly existing, then the possibility level of the coercion indicator is level 4; if the coercion indicator is between 0.007 and 0.07 according to the data value identification standard, or the status value is identified and revised as occasionally existing, then the possibility level of the coercion indicator is level 3; if the coercion indicator is between 0.0007 and 0.007 according to the data value identification standard, or the status value is identified and revised as rarely existing, then the possibility level of the coercion indicator is level 2; if the coercion indicator is not greater than 0.0007 according to the data value identification standard, or the status value is identified and revised as extremely unlikely to exist, then the possibility level of the coercion indicator is level 1. It should be understood that the classification criteria are for illustration purposes only and do not constitute any specific limitation to this scheme. The classification criteria can be determined by analyzing and summarizing the responses of relevant staff members through a questionnaire survey.

[0055] For any coercion indicator, if the order of the degree of existence of the coercion indicator is catastrophic, then the relevance level of the degree of existence of the coercion indicator is level 5; if the order of the degree of existence of the coercion indicator is very relevant, then the relevance level of the degree of existence of the coercion indicator is level 4; if the order of the degree of existence of the coercion indicator is relevant, then the relevance level of the degree of existence of the coercion indicator is level 3; if the order of the degree of existence of the coercion indicator is general, then the relevance level of the degree of existence of the coercion indicator is level 2; if the order of the degree of existence of the coercion indicator is irrelevant, then the relevance level of the degree of existence of the coercion indicator is level 1. It should be understood that the classification standard is only for illustrative purposes and does not constitute a specific limitation on this scheme. The classification standard can be determined by analyzing and summarizing the responses of relevant staff members in the form of a questionnaire survey.

[0056] Based on the judgment criteria, deploy the coercion update log (i.e., the coercion result strategy table);

[0057] According to the stress result strategy table, the ecological flow balance stress level can be divided into three levels: deep, general, and low. For example, for any stress indicator, if the possibility level of the stress indicator is between 3-5, and the correlation level of the degree of existence is 5; or, the possibility level is between 4-5, and the correlation level of the degree of existence is 4; or, the possibility level is 5, and the correlation level of the degree of existence is 3, then the balance stress level of the stress indicator is deep stress.

[0058] Exemplarily, if the existence possibility level of the coercion indicator is between 1-2, and the correlation level of the existence degree is level 5; or, the existence possibility level is between 1-3, and the correlation level of the existence degree is level 4; or, the existence possibility level is between 2-4, and the correlation level of the existence degree is level 3; or, the existence possibility level is between 3-5, and the correlation level of the existence degree is level 2; or, the existence possibility level is between 4-5, and the correlation level of the existence degree is level 1, then the balanced coercion level of the coercion indicator is general coercion.

[0059] Exemplarily, if the existence possibility level of the coercion indicator is between 1-3, and the correlation level of the existence degree is level 1; or, the existence possibility level is between 1-2, and the correlation level of the existence degree is level 2; or, the existence possibility level is 1, and the correlation level of the existence degree is level 3, then the balanced coercion level of the coercion indicator is low coercion.

[0060] In order to standardize and integrate the coercion indicators, it can be performed based on the pre-deployed coercion update log. Accordingly, in some optional embodiments, standardizing and integrating the coercion indicators based on the pre-deployed coercion update log may include:

[0061] Based on the pre-deployed coercion update log, the correlation level and the existence possibility level of the existence degree corresponding to each coercion indicator are determined.

[0062] Based on the correlation and existence possibility of the existence degree corresponding to each stress indicator, the equilibrium stress level corresponding to each stress indicator is determined.

[0063] Based on the equilibrium stress level corresponding to each stress indicator, each stress indicator is standardized and integrated.

[0064] In this embodiment, for example, assuming that the correlation level of the corresponding existence degree of any one of the coercion indicators is level 5 and the existence possibility level is level 5, then the equilibrium coercion level corresponding to the coercion indicator is deep coercion. Quantify the coercion level, that is, use the numbers 1, 2, and 3 to represent the coercion level, then low coercion = 1, general coercion = 2, deep coercion = 3, and accordingly, the coercion indicator can be described as 3. Based on this example, the coercion indicators are standardized and integrated. It should be understood that the various definitions or examples in this embodiment are only for comprehensible explanations and do not constitute any substantial limitation to this solution. In actual use, the description values ​​corresponding to the coercion levels can be set based on actual needs.

[0065] In this embodiment, the monitoring and updating model for ecological flow data is divided into two parts. The first part is the process of deploying the monitoring and updating model structure for ecological flow data as described in step 130, and the second part is the data correction part, that is, the process described in step 140. Step 130 and step 140 are described below in conjunction with other relevant embodiments.

[0066] Step 130: Based on the stress indicators and the dependencies between the stress indicators, deploy a monitoring and updating model structure for ecological flow data.

[0067] In some optional embodiments, step 130 deploys a monitoring and updating model structure for ecological flow data based on the stress indicators and the dependencies between the stress indicators, which may include:

[0068] Based on the coercion indicator, deploy the coercion indicator dataset.

[0069] Each stress indicator is used as a network element variable, and the dependency relationship between the stress indicators is determined, and the network element variables are connected to output multiple queues.

[0070] The satisfaction between each queue and the stress indicator dataset is calculated, and the queue with the largest satisfaction is used as the monitoring update model structure for ecological flow data.

[0071] In the LSTM model algorithm structure of the monitoring and updating method for ecological flow data provided by the embodiment of the present invention, the monitoring and updating model structure for ecological flow data can be an LSTM model algorithm structure. There is a dependency association between the stress indicators associated with the ecological flow balance, that is, when a certain stress indicator appears, it can be associated with another stress indicator. Each stress indicator is used as a network element variable, and multiple stress indicators with dependent associations are connected with arrows, so a queue or mesh structure can be formed. The mesh structure is a directed acyclic graph structure consistent with the LSTM model algorithm. Therefore, the mesh structure can be used as an LSTM model algorithm structure.

[0072] Aquatic biological disturbance can lead to water quality disasters, reduced ecological flow, lower water level and flow loss; aquatic biological quality can lead to water quality disasters and lower water level; human pollution can be associated with water quality disasters, reduced ecological flow and lower water level; poor dam repair can be associated with uneven deformation, water quality disasters and reduced ecological flow; uneven deformation can be associated with water quality disasters; vegetation reduction can be associated with uneven deformation and water quality disasters; local siltation can be associated with precipitation disasters and flow changes; flow changes can be associated with water quality disasters, precipitation disasters and abnormal water flow storage; heavy rainfall can be associated with flow changes calification, local siltation, water quality disasters, precipitation disasters, reduced ecological flow and abnormal water flow storage; artificial pumping and drainage is associated with flow changes, abnormal water flow storage and a drop in water level; surface drainage is associated with precipitation disasters, abnormal water flow storage and a drop in water level; flow loss is associated with abnormal water flow storage, water quality disasters, reduced ecological flow and a drop in water level; water quality disasters are associated with reduced ecological flow; precipitation disasters are associated with reduced ecological flow and water quality disasters; abnormal water flow storage is associated with reduced ecological flow, a drop in water level and water quality disasters; incorrect water use is associated with uneven deformation and water quality disasters.

[0073] The LSTM network structure learning for monitoring and updating ecological flow data is to find the network structure that best fits the data set by learning from expert scores of stress indicators or actual operation and maintenance data.

[0074] Step 140: Determine a monitoring and updating model for ecological flow data based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data.

[0075] In some optional embodiments, based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data, a monitoring and updating model for ecological flow data is determined, including:

[0076] Based on the monitoring of ecological flow data, the model structure is updated to determine the probability distribution corresponding to each stress indicator.

[0077] The monitoring and updating model structure for ecological flow data is taken as the parameter learning object, and the standardized and integrated stress indicators are imported into the monitoring and updating model structure for ecological flow data. Parameter learning is used to correct the monitoring and updating model structure for ecological flow data to determine the monitoring and updating model for ecological flow data.

[0078] In this embodiment, based on the principle of uniform distribution, the initial scenario value of each network element is set to 1 / 3. The established monitoring and updating model structure for ecological flow data, that is, the LSTM model algorithm structure, is used as the parameter learning object, and the standardized and integrated stress index data in step 120 is imported into the LSTM model algorithm structure. At the same time, the imported standardized and integrated stress index data and the established LSTM model algorithm structure are matched, so that the parameter learning results are output according to three scenarios: deep stress scenario, general stress scenario and balance scenario. After the standardized and integrated stress index data is imported, the EM algorithm is applied for parameter learning. By modifying the logical relationship between some stress indicators, optimizing and correcting the LSTM model algorithm structure, the final output of the monitoring and updating model for ecological flow data, that is, the LSTM model algorithm, is more scientific and accurate.

[0079] Specifically, taking a network element as an example, applying the EM algorithm to perform parameter learning may include:

[0080] The target network element is determined as “ecological traffic reduction”, and the starting point of the parameters is set randomly, and finally the conditional possibility distribution of each evaluation index network element is output.

[0081] The EM algorithm is an optimization algorithm that iteratively searches for parameters in a likelihood model and performs maximum likelihood estimation.

[0082] Step 150: Collect the balance stress data of the ecological flow to be updated, and output the balance stress update result of the ecological flow to be updated based on the balance stress data and the monitoring update model for the ecological flow data.

[0083] In some embodiments, the balance stress data of the ecological flow to be updated is a stress index associated with the balance of the ecological flow to be updated and / or an abnormal type of the ecological flow to be updated. Then, step 150 outputs the balance stress update result of the ecological flow to be updated based on the balance stress data and the monitoring update model for the ecological flow data, which may include:

[0084] Based on the pre-deployed stress update log, the balance stress data is standardized and integrated, and the standardized and integrated balance stress data is output.

[0085] The standardized and integrated stress indicators associated with the ecological flow balance to be updated are sent to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated; among them, the balance stress scenario includes deep stress scenario, general stress scenario and balance scenario.

[0086] And / or, the abnormal type of the ecological flow to be updated after standardized integration is sent to the LSTM model algorithm to determine the coercion indicator of the abnormal type of the ecological flow to be updated.

[0087] In this embodiment, since the quantities used in the LSTM model algorithm are standardized and integrated, when using the model, the input quantities also need to be standardized and integrated. The specific integration process can refer to the other related embodiments mentioned above and will not be repeated here.

[0088] Based on the different updating purposes of the balance stress of a certain ecological flow, the LSTM network model can be used to update the ecological flow balance stress level for balance anomalies that have not yet existed; the LSTM network model can also be used to investigate and update the reasons for the existence of existing balance anomalies.

[0089] If the balance stress data of the ecological flow to be updated is a stress indicator associated with the balance of the ecological flow to be updated, the balance stress scenario of the ecological flow to be updated can be determined by the following steps:

[0090] The standardized and integrated stress indicators associated with the ecological flow balance to be updated are sent to the LSTM model algorithm, and based on the hypothesis deductive analysis, the possibility distribution of the equilibrium stress scenario corresponding to each stress indicator is output.

[0091] If the probability value of the deep stress scene in the balanced stress scene corresponding to any stress indicator exceeds the preset stress threshold, it is determined that the stress indicator is in the deep stress scene.

[0092] If the possibility values ​​of the deep stress scenes in the balanced stress scenes corresponding to the stress indicators do not exceed the preset stress thresholds, it is determined that the stress indicators are not in the deep stress scenes.

[0093] In the hypothesis-deductive analysis LSTM model algorithm of the monitoring and updating method for ecological flow data provided in the embodiment of the present invention, the preset stress threshold can be 25%, 30%, 35%, etc., which is not limited here. Exemplarily, if the probability value corresponding to the deep stress scenario in any one of the stress indicators exceeds 30%, then it is determined that the stress indicator is in a deep stress scenario, and accordingly, the balanced stress scenario of the ecological flow to be updated is a deep stress scenario; if the probability value corresponding to the deep stress scenario in any one of the stress indicators does not exceed 30%, then it is determined that each stress indicator is not in a deep stress scenario, and accordingly, the balanced stress scenario of the ecological flow to be updated is a balanced scenario.

[0094] The preset coercion threshold is set to 30%. Among them, the probability values ​​of deep coercion scenarios corresponding to the reduction of ecological flow due to human pollution, poor dam repair, local siltation, flow and water changes, heavy rainfall, water level drop, precipitation disasters, and litigation disasters are all greater than 30%. Then these coercion indicators are in deep coercion scenarios. Correspondingly, the balance coercion scenario of the ecological flow is a deep coercion scenario. It is necessary to analyze the causes of coercion in a timely manner and take certain control measures. When the probability of the obtained deep coercion scenario does not exceed 30%, it is considered that the coercion level of the network element is within an acceptable range and only needs to continue monitoring, such as aquatic biological quality, aquatic biological disturbance, surface drainage, artificial pumping and drainage, etc.

[0095] In some optional embodiments, sending the standardized and integrated abnormal type of the ecological flow to be updated to the LSTM model algorithm, and determining whether the ecological flow to be updated has a coercion indicator of the abnormal type may include:

[0096] Based on the abnormal type of ecological traffic to be updated after standardized integration, the possibility distribution of this abnormal type is adjusted to a deep stress scenario in the output network element of the LSTM model algorithm. Based on inductive analysis, the possibility distribution of the balanced stress scenario corresponding to each stress indicator is determined, and the stress indicator whose probability value of the deep stress scenario exceeds the preset stress threshold is taken as the stress indicator of the existence of this abnormal type.

[0097] In the inductive analysis LSTM model algorithm of the monitoring and updating method for ecological flow data provided in an embodiment of the present invention, corresponding to the stress indicators, the abnormal types may also include incorrect water use, vegetation reduction, poor dam repair, flow water changes, heavy rainfall, surface drainage, artificial pumping and drainage, flow loss, aquatic biological disturbance, aquatic biological quality, human behavior pollution, uneven deformation, local siltation, water level drop, water quality disasters, precipitation disasters and water flow accumulation abnormalities, etc., which are not limited here.

[0098] Exemplarily, when the abnormal type of the ecological flow to be updated is the reduction of ecological flow, the possibility value of the deep coercion scenario corresponding to the reduction of ecological flow can be adjusted to exceed the preset coercion threshold in the output network element of the LSTM model algorithm. Since the reduction of ecological flow already exists, the possibility value of the deep coercion scenario can be adjusted to 100%. Based on the inductive analysis, the coercion indicators that cause the reduction of ecological flow are determined. For example, if the possibility value of the deep coercion scenario corresponding to human pollution, poor dam repair, local siltation, flow and water changes, heavy rainfall, water quality disasters, precipitation disasters, abnormal water flow accumulation, and water level drop is greater than the preset coercion threshold of 30%, it can be determined that these coercion indicators are most significantly associated with the reduction of ecological flow, and these coercion indicators need to be investigated in detail.

[0099] In summary, compared with the traditional method, the embodiment of the present invention takes into account the dependency between various stress indicators, the correlation level and the existence possibility level of each stress indicator when deploying the monitoring and updating model for ecological flow data, avoiding the arbitrariness of subjective updates, ensuring the scientific, objective and convenient update, and improving the reliability and accuracy of the update results. The output update results provide reference data for the research work on ecological flow balance stress.

[0100] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0101] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0102] Figure 2 The schematic diagram of the structure of the monitoring and updating device for ecological flow data provided by an embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0103] like Figure 2 As shown, the monitoring and updating device 5 for ecological flow data includes:

[0104] A collection unit 51, for collecting stress indicators related to ecological flow balance;

[0105] The integration unit 52 is used to standardize and integrate the coercion indicators based on the pre-deployed coercion update log, and output the standardized and integrated coercion indicators; wherein the pre-deployed coercion update log is used to illustrate the correlation level and existence possibility level of the existence degree of each coercion indicator;

[0106] A deployment unit 53, configured to deploy a monitoring and updating model structure for ecological flow data based on the stress indicators and the dependency relationships between the stress indicators;

[0107] The deployment unit 53 is further used to determine the monitoring and updating model for ecological flow data based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data;

[0108] The updating unit 54 is used to collect the balance stress data of the ecological flow to be updated, and output the balance stress update result of the ecological flow to be updated based on the balance stress data and the monitoring update model oriented to the ecological flow data.

[0109] In a possible implementation, the monitoring and updating model for ecological flow data is an LSTM model algorithm;

[0110] The balance stress data of the ecological flow to be updated are stress indicators associated with the balance of the ecological flow to be updated and / or abnormal types of the ecological flow to be updated;

[0111] The updating unit 54 is specifically used for:

[0112] Based on the pre-deployed stress update log, the balance stress data is standardized and integrated, and the standardized and integrated balance stress data is output;

[0113] Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated; wherein the balance stress scenario includes a deep stress scenario, a general stress scenario and a balance scenario;

[0114] And / or, the abnormal type of the ecological flow to be updated after standardized integration is sent to the LSTM model algorithm to determine the stress indicator of the abnormal type of the ecological flow to be updated.

[0115] In a possible implementation, the updating unit 54 is specifically configured to:

[0116] Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm, and outputting the possibility distribution of the equilibrium stress scenario corresponding to each stress indicator based on hypothesis deduction analysis;

[0117] If the probability value of the deep stress scene in the balanced stress scene corresponding to any stress indicator exceeds the preset stress threshold, it is determined that the stress indicator is in the deep stress scene;

[0118] If the possibility values ​​of the deep stress scenes in the balanced stress scenes corresponding to the stress indicators do not exceed the preset stress thresholds, it is determined that the stress indicators are not in the deep stress scenes.

[0119] In a possible implementation, the updating unit 54 is specifically configured to:

[0120] Based on the abnormal type of the ecological traffic to be updated after standardized integration, in the output network element of the LSTM model algorithm, the possibility distribution of the abnormal type is adjusted to a deep stress scenario, and based on inductive analysis, the possibility distribution of the balanced stress scenario corresponding to each stress indicator is determined, and the stress indicator whose possibility value of the deep stress scenario exceeds the preset stress threshold is used as a stress indicator of the existence of this abnormal type.

[0121] In a possible implementation, the integration unit 52 is specifically configured to:

[0122] Based on the pre-deployed coercion update log, determine the correlation level and existence possibility level of the existence degree corresponding to each coercion indicator;

[0123] Based on the correlation and existence possibility of the existence degree corresponding to each stress indicator, the equilibrium stress level corresponding to each stress indicator is determined;

[0124] Based on the equilibrium stress level corresponding to each stress indicator, each stress indicator is standardized and integrated.

[0125] In a possible implementation, the deployment unit 53 is specifically configured to:

[0126] Based on the coercion indicator, deploy the coercion indicator dataset;

[0127] Each stress indicator is used as a network element variable, and the dependency relationship between the stress indicators is determined, and the network element variables are connected to output multiple queues;

[0128] The satisfaction between each queue and the stress indicator dataset is calculated, and the queue with the largest satisfaction is used as the monitoring update model structure for ecological flow data.

[0129] In a possible implementation, the deployment unit 53 is specifically configured to:

[0130] Based on the monitoring of ecological flow data, the model structure is updated to determine the probability distribution corresponding to each stress indicator;

[0131] The monitoring and updating model structure for ecological flow data is taken as the parameter learning object, and the standardized and integrated stress indicators are imported into the monitoring and updating model structure for ecological flow data. Parameter learning is used to correct the monitoring and updating model structure for ecological flow data to determine the monitoring and updating model for ecological flow data.

[0132] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0133] In addition, each functional unit in each embodiment of the present application can be integrated into an integrated unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0134] The specific implementation methods of the invention are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the invention is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A monitoring and updating method for ecological flow data, characterized in that: include: Collect stress indicators related to ecological flow balance, including one or more of the following: improper water use, poor dam repair, flow and water changes, heavy rainfall, and abnormal aquatic plants and animals; Based on the pre-deployed coercion update log, the coercion indicators are standardized and integrated, and the standardized and integrated coercion indicators are output; wherein the pre-deployed coercion update log is used to illustrate the correlation level and existence possibility level of the existence degree of each coercion indicator; Based on the stress indicators and the dependencies between them, a monitoring and updating model structure for ecological flow data is deployed; Determining a monitoring and updating model for ecological flow data based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data; Collecting the balance stress data of the ecological flow to be updated, and outputting the balance stress update result of the ecological flow to be updated based on the balance stress data and the monitoring and updating model for the ecological flow data; Based on the stress indicators and the dependencies between them, a monitoring and updating model structure for ecological flow data is deployed, including: Based on the stress indicator, deploy a stress indicator dataset; Each stress indicator is used as a network element variable, and the dependency relationship between the stress indicators is determined, and the network element variables are connected to output multiple queues; Calculate the satisfaction of each queue with the stress indicator data set, and use the queue with the largest satisfaction as the monitoring update model structure for ecological flow data; Based on the standardized integrated stress indicators and the monitoring and updating model structure for ecological flow data, a monitoring and updating model for ecological flow data is determined, including: Determine the probability distribution corresponding to each stress indicator based on the monitoring and updating model structure for ecological flow data; The monitoring and updating model structure for ecological flow data is used as a parameter learning object, and the standardized and integrated stress indicators are imported into the monitoring and updating model structure for ecological flow data, and the monitoring and updating model structure for ecological flow data is modified by parameter learning to determine the monitoring and updating model for ecological flow data; The monitoring and updating model for ecological flow data is an LSTM model algorithm; The balance stress data of the ecological flow to be updated are stress indicators associated with the balance of the ecological flow to be updated and / or abnormal types of the ecological flow to be updated; Based on the pre-deployed stress update log, the balance stress data is standardized and integrated, and the standardized and integrated balance stress data is output; Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated; wherein the balance stress scenario includes a deep stress scenario, a general stress scenario and a balance scenario; And / or, the abnormal type of the ecological flow to be updated after standardized integration is sent to the LSTM model algorithm to determine the coercion indicator of the abnormal type of the ecological flow to be updated.

2. The ecological flow data monitoring and updating method according to claim 1, characterized in that: The standardized and integrated stress index associated with the ecological flow balance to be updated is sent to the LSTM model algorithm to determine the balance stress scenario of the ecological flow to be updated, including: Sending the standardized and integrated stress indicators associated with the ecological flow balance to be updated to the LSTM model algorithm, and outputting the possibility distribution of the equilibrium stress scenario corresponding to each stress indicator based on hypothesis deduction analysis; If the probability value of the deep stress scene in the balanced stress scene corresponding to any stress indicator exceeds the preset stress threshold, it is determined that the stress indicator is in the deep stress scene; If the possibility values ​​of the deep stress scenes in the balanced stress scenes corresponding to the stress indicators do not exceed the preset stress thresholds, it is determined that the stress indicators are not in the deep stress scenes.

3. The ecological flow data monitoring and updating method according to claim 2, characterized in that: The sending of the standardized and integrated abnormal type of the to-be-updated ecological flow to the LSTM model algorithm to determine the stress indicator of the abnormal type of the to-be-updated ecological flow includes: Based on the abnormal type of the ecological traffic to be updated after standardized integration, in the output network element of the LSTM model algorithm, the possibility distribution of the abnormal type is adjusted to a deep stress scenario, and based on inductive analysis, the possibility distribution of the balanced stress scenario corresponding to each stress indicator is determined, and the stress indicator whose possibility value of the deep stress scenario exceeds the preset stress threshold is used as a stress indicator of the existence of this abnormal type.

4. The monitoring and updating method for ecological flow data according to claim 1, characterized in that: The stress update log based on the pre-deployment is used to standardize and integrate the stress indicators, including: Determine, based on the pre-deployed coercion update log, the correlation level and the existence possibility level of the existence degree corresponding to each of the coercion indicators; Determining a balanced stress level corresponding to each stress indicator based on the correlation and existence possibility of the existence degree corresponding to each stress indicator; Based on the equilibrium stress level corresponding to each stress indicator, each stress indicator is standardized and integrated.

5. A monitoring and updating device for ecological flow data, characterized in that: The device is used to monitor and update ecological flow data, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor includes a collection unit, an integration unit, a deployment unit, and an update unit; the computer program executed by the processor is used to execute a monitoring and updating method for ecological flow data as described in any one of claims 1 to 4.

Images