A river water ecosystem health evaluation method, system, medium and product

Abstract

The application relates to the technical field of data analysis, in particular to a river water ecosystem health evaluation method and system, medium and product, which comprises the following steps: acquiring a plurality of criterion indexes and criterion index data corresponding to each criterion layer of a to-be-evaluated river; obtaining an index score value corresponding to each criterion index based on a preset index scoring neural network model and the criterion index data; acquiring preset index weights and preset criterion layer weights corresponding to the to-be-evaluated river; determining a river health comprehensive score corresponding to the to-be-evaluated river based on the preset index weights, the preset criterion layer weights and the index score values corresponding to the to-be-evaluated river; and determining the health condition grade of the to-be-evaluated river based on the river health comprehensive score. The application facilitates improving the accuracy of health evaluation of the river water ecosystem.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a river water ecosystem health evaluation method, system, medium and product. BACKGROUND

[0002] With the rapid advancement of industrialization and urbanization in China, the river water ecosystem is facing increasingly severe challenges. Water resources are in short supply, water quality is deteriorating, and hydrological rhythms are disorderly. Problems such as water resource allocation imbalance, upstream water reduction, river siltation shrinkage, and even river closure are frequent. This not only destroys the continuity of the river water ecosystem, but also poses a serious threat to biodiversity. As the core carrier of water resource circulation, the health of the river water ecosystem is directly related to regional ecological security, sustainable use of water resources, and the quality of the human living environment. Therefore, scientific and accurate health evaluation of the river water ecosystem is the key prerequisite for identifying ecological problems, developing restoration strategies, and ensuring the ecological function of the river. It has important practical significance and application value for promoting ecological civilization construction and achieving sustainable development.

[0003] The existing technology relies too much on artificial experience in the selection and weight determination of multiple index comprehensive evaluation, and is easily affected by subjective preference, which may lead to deviation of the evaluation result from the actual health status of the river. In addition, the existing method generally uses general standards or regional experience thresholds for health evaluation, without fully considering the differences in hydrological characteristics and ecological needs of different rivers. Therefore, there is a defect of low accuracy in health evaluation. SUMMARY

[0004] In order to improve the accuracy of health evaluation of the river water ecosystem, the present application provides a river water ecosystem health evaluation method, system, medium and product.

[0005] In a first aspect, the present application provides a river water ecosystem health evaluation method, which adopts the following technical solution:

[0006] A river water ecosystem health evaluation method, comprising:

[0007] Obtaining a plurality of criterion indicators and criterion indicator data corresponding to each criterion layer of the river to be evaluated;

[0008] Based on the preset index scoring neural network model and the criterion indicator data, an index scoring value corresponding to each criterion indicator is obtained;

[0009] Obtaining a preset index weight and a preset criterion layer weight corresponding to the river to be evaluated, and determining a river health comprehensive score corresponding to the river to be evaluated based on the preset index weight, the preset criterion layer weight, and each index scoring value corresponding to the river to be evaluated.

[0010] determine the health level of the river to be evaluated based on the comprehensive score of the river health.

[0011] By adopting the technical solution, the health of the river to be evaluated is evaluated by comprehensively considering the criterion index data of each criterion layer, avoiding the one-sidedness of health evaluation caused by relying on a single factor, reducing the interference of a single factor on the health evaluation result, ensuring the integrity of the evaluation perspective, and using the preset index score neural network to score each criterion index, using the learning and fitting ability of the neural network model to complex data patterns, facilitating the deep mining of the potential nonlinear mapping relationship between each criterion index data and the corresponding index score value, and getting rid of the limitations of traditional manual scoring on experience dependence, thereby facilitating the improvement of the accuracy of the scoring result. On this basis, by integrating all the index score values with high accuracy and combining the scientific preset index weight and criterion layer weight for comprehensive calculation, the real contribution of each criterion index can be effectively integrated into the river health comprehensive score, avoiding the influence of local criterion index data error or unreasonable weight on the overall health evaluation result, thereby facilitating the improvement of the accuracy of the health evaluation of the river water ecosystem.

[0012] In a possible implementation manner, the index score value corresponding to each criterion index is obtained based on the preset index score neural network model and the criterion index data, including:

[0013] identifying the criterion index features corresponding to each criterion index data, determining the index score strategy corresponding to each criterion index feature from the preset index score neural network model based on each criterion index feature;

[0014] determining the index score value corresponding to each criterion index based on each criterion index data and the corresponding index score strategy.

[0015] By adopting the technical solution, the criterion index features contained in different criterion index data are identified, and the corresponding exclusive scoring strategy is assigned to each criterion index feature, which facilitates the avoidance of scoring deviation caused by mismatching of the scoring strategy, thereby facilitating the improvement of the accuracy of determining the index score value corresponding to each criterion index.

[0016] In a possible implementation manner, the river health comprehensive score corresponding to the river to be evaluated is determined based on the preset index weight, the preset criterion layer weight, and the index score value corresponding to the river to be evaluated, including:

[0017] determine a river health comprehensive score of the river health according to the preset index weight, the preset criterion layer weight, an index score value corresponding to each criterion index, and a first river health comprehensive score formula, wherein the first river health comprehensive score formula is:

[0018] wherein:

[0019] RHI i representing a river health comprehensive score corresponding to the i th river to be evaluated;

[0020] ZB nw representing a preset index weight of the n th index of the criterion layer;

[0021] ZB nr representing an index score value of the n th index of the criterion layer;

[0022] YMB mw representing a preset criterion layer weight of the m th criterion layer.

[0023] By adopting the above technical solution, a specific implementation manner for determining a river health comprehensive score corresponding to a river to be evaluated is provided.

[0024] In a possible implementation manner, when the river to be evaluated comprises a segmented river, the method further comprises:

[0025] identifying a number of segments of the segmented river corresponding to the river to be evaluated, a segment length of each segmented river, and a segment health comprehensive score of each segmented river;

[0026] determining a river health comprehensive score corresponding to the river to be evaluated comprising the segmented river based on the number of segments, the segment length of each segmented river, and the segment health comprehensive score of each segmented river, and a second river health comprehensive score formula, wherein the second river health comprehensive score formula is:

[0027] wherein:

[0028] RHI represents a river health comprehensive score corresponding to the river to be evaluated comprising the segmented river;

[0029] RHI i representing a segment health comprehensive score of the i th segmented river;

[0030] W i representing the i th segmented river and the segment length;

[0031] R s representing the number of segments of the segmented river.

[0032] By adopting the technical scheme, when the river to be evaluated comprises segmented rivers, the data of each segmented river can be independently checked and scored through segmented integration processing, and then the influence of local data errors on the overall result can be weakened through weighted integration, so as to facilitate improving the accuracy of determining the river health comprehensive score of the river to be evaluated.

[0033] In a possible implementation manner, the process of determining the preset index weight comprises:

[0034] performing subjective weight calculation based on the index score values of the respective criterion indexes to obtain a plurality of subjective index weights corresponding to the respective criterion indexes;

[0035] performing objective weight calculation based on the index score values of the respective criterion indexes to obtain a plurality of objective index weights corresponding to the respective criterion indexes;

[0036] performing weight fitting on the plurality of subjective index weights and the plurality of objective index weights corresponding to the respective criterion indexes to obtain a preset index weight corresponding to each criterion index.

[0037] By adopting the technical scheme, the subjective weight calculation is usually based on the experience judgment of the evaluator on the importance of the criterion index, which facilitates embodying the inherent properties of the river water ecological system and the actual management needs, and the objective weight calculation facilitates reflecting the differentiation of the criterion index on the health state of the river to be evaluated in the actual health evaluation, so as to avoid the limitations of the single subjective weight depending on experience and possibly deviating from the actual data, or the single objective weight neglecting ecological logic and possibly being disturbed by abnormal data, and facilitate making the preset index weight conform to ecological cognition while fitting the data essence.

[0038] In a possible implementation manner, the weight fitting on the plurality of subjective index weights and the plurality of objective index weights corresponding to the respective criterion indexes to obtain a preset index weight corresponding to each criterion index comprises:

[0039] inputting the plurality of subjective index weights and the plurality of objective index weights corresponding to the respective criterion indexes as input data into a preset first neural network model and a preset second neural network model respectively for weight fitting to obtain first fitting weights and second fitting weights, wherein the preset first neural network model outputs the first fitting weights, and the preset second neural network model outputs the second fitting weights;

[0040] determining, based on performance indexes of the preset first neural network model and the preset second neural network model, a first weight contribution proportion corresponding to the preset first neural network model and a second weight contribution proportion corresponding to the preset second neural network model;

[0041] Determine a preset index weight corresponding to the criterion index based on the first fitting weight, the second fitting weight, the first weight contribution ratio, and the second weight contribution ratio.

[0042] By adopting the above technical solutions, the internal relationship between the subjective and objective weights is analyzed from different angles through parallel processing of the two models, so that the fitting effect is improved. In addition, by identifying and analyzing the performance indicators of the two neural network models and dynamically determining the respective weight contribution ratios, the accuracy of the weight fitting result is further improved.

[0043] In a second aspect, the present application provides an evaluation system, which adopts the following technical solutions:

[0044] An evaluation system, comprising:

[0045] At least one processor;

[0046] A memory;

[0047] At least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to implement the river water ecosystem health evaluation method.

[0048] In a third aspect, the present application provides a computer readable storage medium, which adopts the following technical solutions:

[0049] A computer readable storage medium, comprising a computer program capable of being loaded and executed by a processor to implement the river water ecosystem health evaluation method.

[0050] In a fourth aspect, the present application provides a computer program product, which adopts the following technical solutions:

[0051] A computer program product, comprising a computer program, which is executed by a processor to implement the river water ecosystem health evaluation method.

[0052] In summary, the present application has at least one of the following beneficial technical effects:

[0053] The health of the river to be evaluated is evaluated by comprehensively considering the criterion index data of the river to be evaluated in each criterion layer, avoiding the one-sidedness of the health evaluation caused by relying on a single factor, reducing the interference of a single factor on the health evaluation result, ensuring the integrity of the evaluation perspective, and assigning scores to each criterion index by using a preset index scoring neural network. With the learning and fitting ability of the neural network model to complex data patterns, it is convenient to deeply mine the potential nonlinear mapping relationship between each criterion index data and the corresponding index score value, get rid of the limitations of traditional manual scoring on experience dependence, and thus improve the accuracy of the scoring result. On this basis, by integrating all the index score values with high accuracy and combining scientific preset index weights and criterion layer weights for comprehensive calculation, the real contribution of each criterion index can be effectively integrated into the river health comprehensive score, avoiding the influence of local criterion index data error or unreasonable weight on the overall health evaluation result, thereby improving the accuracy of the health evaluation of the river water ecosystem.

[0054] The subjective weight calculation is usually based on the experience judgment of the evaluator on the importance of the criterion index, which is convenient for reflecting the inherent properties and actual management needs of the river water ecosystem. However, the objective weight calculation is convenient for reflecting the distinction of the criterion index to the health status of the river to be evaluated in the actual health evaluation. By combining the two, it is convenient to avoid the limitations of a single subjective weight relying on experience, possibly deviating from the actual data, or a single objective weight ignoring ecological logic, possibly being disturbed by abnormal data, so as to make the preset index weight conform to the ecological cognition while being consistent with the data essence. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a flowchart of a river water ecosystem health evaluation method in an embodiment of the present application;

[0056] Figure 2 is a flowchart of determining a preset index weight in an embodiment of the present application;

[0057] Figure 3 is a structural diagram of an evaluation system in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be described in detail in combination with the accompanying drawings. Figures 1 to 3 The present application will be further described in detail.

[0059] Those skilled in the art can make modifications to the present embodiment without creative contribution after reading the present specification according to the needs, but as long as it is within the scope of the claims of the present application, it is protected by the patent law.

[0060] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0061] It should be noted that, in the optional embodiments of the present application, the data related to the object information and the like needs to be obtained with the permission or consent of the object when the embodiments of the present application are applied to specific products or technologies, and the collection, use and processing of the related data need to comply with the relevant laws, regulations and standards of the country and region. That is, if the embodiments of the present application involve data related to the object, the data needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant department, and the compliance with the relevant laws, regulations and standards of the country and region. If the embodiments involve personal information, the consent of the individual needs to be obtained for the acquisition of all personal information, and the separate consent of the information subject needs to be obtained for the acquisition of sensitive information, and the embodiments also need to be implemented with the authorization and consent of the object.

[0062] Specifically, the present application provides a river water ecological system health evaluation method, which is executed by an evaluation system. The evaluation system can be a server or a terminal device. The server can be a stand-alone physical server, a server cluster composed of multiple physical servers or a distributed system, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the present application is not limited thereto.

[0063] Reference Figure 1 , Figure 1 is a flowchart of a river water ecological system health evaluation method in the embodiments of the present application. The method comprises steps S110-S140, wherein:

[0064] Step S110: Obtain multiple criteria indicators and criteria indicator data corresponding to each criteria layer of the river to be evaluated.

[0065] Specifically, the river to be evaluated is a river that needs to be evaluated for health, and can be uploaded to the evaluation system by relevant staff according to actual health evaluation needs. In the multi-index comprehensive evaluation system of river water ecosystem health evaluation, the criterion layer is one of the core levels of the evaluation framework. Due to the great differences in natural environment and social influence of rivers in different regions, the health of river water ecosystem is easily affected by hydrology, water quality, biology, topography, human activities and other factors. Therefore, directly evaluating the health with a large number of scattered indexes may lead to confusion in the evaluation logic. Therefore, a large number of indexes can be divided into different criterion layers, so that the evaluation framework is clearer and the levels are more distinct. In the embodiment of the application, the criterion layer includes a "basin" criterion layer, a "water" criterion layer, a biological criterion layer, and a social service function criterion layer. The "basin" criterion layer includes river longitudinal connectivity index, bank stability, bank vegetation coverage, standardized construction rate of river discharge outlet, layout rationality of river discharge outlet, and "four disorder" condition of river channel, a total of 6 criterion indexes; the "water" criterion layer includes ecological flow satisfaction degree, water quality degree, and water self-purification capacity, a total of 3 criterion indexes; the biological criterion layer includes fish retention index and water bird condition, a total of 2 criterion indexes; the social service criterion layer includes dike flood control standard, cross-river building flood control standard, bank management index, and public satisfaction, a total of 4 criterion indexes. In addition, the criterion indexes can also be navigation guarantee rate, macrobenthic invertebrate biological integrity, etc. The specific criterion indexes can be set by the relevant technical personnel according to the actual health evaluation needs of the river to be evaluated.

[0066] The criterion index data is used to describe or summarize the content of the corresponding criterion index, and is a direct reflection of the actual state, characteristics or performance of each criterion index. It can be collected or detected by relevant staff through field monitoring, laboratory analysis, historical data sorting, questionnaire survey, etc., and then uploaded to the evaluation system. For example, the criterion index data corresponding to the bank vegetation coverage criterion index in the "basin" criterion layer can be remote sensing images containing the bank vegetation of the river to be evaluated, or field measurement data of the bank vegetation of the river to be evaluated; the criterion index data corresponding to the bank management index criterion index in the social service criterion layer can be "the total length of the planned river channel is 18.90km, after the guide line is drawn, the area between the two banks of the guide line is the designed flood control area, during which the water area, sandbar, beach, flood control area, dike protection area, and the guide line extending horizontally outward by 5-10m are classified as the river management range". The specific form of the criterion index data is not limited in the embodiment of the application.

[0067] Step S120: Based on the pre-set index scoring neural network model and each criterion index data, the index scoring value corresponding to each criterion index is obtained.

[0068] Specifically, the preset index scoring neural network model includes a plurality of scoring strategies of criterion index data, different criterion indexes correspond to different scoring manners, and the personalized scoring manner facilitates improving the fitting degree between the scoring result and the actual situation. Further, in order to facilitate avoiding scoring deviation caused by the mismatching of the scoring strategy, based on the preset index scoring neural network model and the criterion index data, an index scoring value corresponding to each criterion index is obtained, which can specifically include:

[0069] identifying criterion index features corresponding to each criterion index data, determining an index scoring strategy corresponding to each criterion index feature from the preset index scoring neural network model based on each criterion index feature; determining an index scoring value corresponding to each criterion index based on each criterion index data and the corresponding index scoring strategy.

[0070] Specifically, the scoring strategy of the "basin" criterion layer in the preset index scoring neural network model can include:

[0071] a1, river longitudinal connectivity index scoring: the corresponding barrage type and barrage quantity can be identified from the criterion index data corresponding to the river longitudinal connectivity index according to the preset feature recognition algorithm, and the longitudinal connectivity building quantity is determined based thereon, the longitudinal connectivity building is a building provided with fish passing facilities, can normally pass through, and can meet the biological ecological flow demand in the river to be evaluated, and then scoring is performed according to the longitudinal connectivity building quantity and the river longitudinal connectivity index scoring standard threshold table, wherein the river longitudinal connectivity index scoring standard threshold table includes the scoring value corresponding to different longitudinal connectivity building quantities per 100 km, for example, if the river to be evaluated contains 1 longitudinal connectivity building per 100 km, the corresponding river longitudinal connectivity index score is 20. The river longitudinal connectivity index scoring standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0072] a2, river bank stability scoring: the corresponding bank slope angle, bank slope vegetation coverage, bank slope height, bank slope matrix and river bank erosion condition can be identified from the criterion index data corresponding to the river bank stability according to the preset feature recognition algorithm, and then the corresponding river bank stability scoring is determined based on the river bank stability scoring standard threshold table, wherein the river bank stability scoring standard threshold table includes the river bank stability scoring corresponding to different combinations of bank slope angle, bank slope vegetation coverage, bank slope height, bank slope matrix and river bank erosion condition, for example, when the bank slope angle is less than or equal to 15°, the bank slope vegetation coverage is greater than or equal to 75%, the bank slope height is less than or equal to 1m, the bank slope matrix is bedrock, and there is no erosion sign on the river bank, the corresponding river bank stability scoring is 100. The river bank stability scoring standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0073] a3, shore vegetation coverage score: according to the preset feature recognition algorithm, the proportion of the vertical projection area of the natural and artificial vegetation in the riparian zone to the area of the riparian zone, i.e., the shore vegetation coverage, can be recognized from the corresponding criterion index data of the shore vegetation coverage, and then the corresponding shore vegetation coverage score can be determined according to the shore vegetation coverage score standard threshold table, wherein the shore vegetation coverage score standard threshold table contains different shore vegetation coverage scores corresponding to different shore vegetation coverages, for example, when the shore vegetation coverage is 50%-75%, the corresponding shore vegetation coverage score can be determined as 75 points. The shore vegetation coverage score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0074] a4, standardized construction of river pollution discharge outlet score: according to the preset feature recognition algorithm, the number of river pollution discharge outlets that have been standardized and the total number of river pollution discharge outlets can be recognized from the corresponding criterion index data of the standardized construction of river pollution discharge outlets, the standardized construction rate of river pollution discharge outlets can be obtained based on the standardized construction rate formula of river pollution discharge outlets, the corresponding grade of the standardized construction rate of river pollution discharge outlets can be determined, and finally the corresponding standardized construction rate of river pollution discharge outlets can be determined according to the grade of the standardized construction rate of river pollution discharge outlets and the standardized construction rate of river pollution discharge outlets score standard threshold table, wherein the standardized construction rate formula of river pollution discharge outlets is R G =N i / N*100, wherein R G is the standardized construction rate of river pollution discharge outlets, N i is the number of river pollution discharge outlets that have been standardized, and N is the total number of river pollution discharge outlets. The standardized construction rate of river pollution discharge outlets score standard threshold table contains different river pollution discharge outlet standardized construction rate grades corresponding to river pollution discharge outlet standardized construction scores, for example, when the standardized construction rate of river pollution discharge outlets is excellent, the corresponding river pollution discharge outlet standardized construction score can be determined as 100 points. The standardized construction rate of river pollution discharge outlets score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0075] a5, the rationality degree of the layout of the river sewage outlet is scored: the setting position, the setting quantity, and the water quality corresponding to the setting position of the river sewage outlet are identified from the criterion index data corresponding to the rationality degree of the layout of the river sewage outlet according to a preset feature recognition algorithm, and the corresponding rationality degree of the layout of the river sewage outlet is determined based on a rationality degree of the layout of the river sewage outlet scoring standard threshold table, wherein the rationality degree of the layout of the river sewage outlet scoring standard threshold table contains a plurality of parameter combinations of the setting position, the setting quantity, and the water quality corresponding to the setting position of the river sewage outlet corresponding to the rationality degree of the layout of the river sewage outlet scoring, for example, when the river sewage outlet setting condition is that there is no river sewage outlet in the drinking water source and the secondary protection area; there is no sewage outlet 1km upstream of the water intake; the length of the sewage belt formed by the sewage is less than 1km, or the width is less than 1 / 4 of the river width, the corresponding rationality degree of the layout of the river sewage outlet scoring can be determined as 50 points. The rationality degree of the layout of the river sewage outlet scoring standard threshold table can be determined by the relevant technical personnel according to the historical experimental data and uploaded to the evaluation system in advance.

[0076] a6, the river "four chaos" condition scoring: the degrees of chaos mining, chaos occupation, chaos stacking, and chaos building are identified from the criterion index data corresponding to the river "four chaos" condition according to a preset feature recognition algorithm, and the corresponding river "four chaos" condition scoring is determined based on a river "four chaos" condition scoring standard threshold table, wherein the river section without "four chaos" condition is scored as 100 points. When the river "four chaos" condition is deducted, the severity should be considered, and the deduction is stopped. For example, the river "four chaos" condition is chaos mining condition, which is a general problem, deducting 5 points; the chaos occupation condition is a more serious problem, deducting 25 points; the chaos stacking condition is a major problem, deducting 50 points; the chaos building condition is a general problem, deducting 5 points, the total deduction is 85 points, and the corresponding river "four chaos" condition scoring is 15 points. The river "four chaos" condition scoring standard threshold table can be determined by the relevant technical personnel according to the historical experimental data and uploaded to the evaluation system in advance.

[0077] The scoring strategy of the "water" criterion layer in the preset index scoring neural network model can include:

[0078] b1, ecological flow satisfaction degree score: the preset feature recognition algorithm can be used to identify the river type of the river to be evaluated from the criterion index data corresponding to the ecological flow satisfaction degree. The river type includes perennial rivers and seasonal rivers. For perennial rivers, the minimum daily flow ratio in the dry season and the minimum daily flow ratio in the wet season can be determined by the minimum daily flow ratio score method. Then, the corresponding ecological flow satisfaction degree score is determined based on the ecological flow satisfaction degree score standard threshold table. The ecological flow satisfaction degree score standard threshold table contains different ecological flow satisfaction degree scores corresponding to different minimum daily flow ratios in the dry season or in the wet season. For example, when the minimum daily flow ratio in the wet season is 40%, the corresponding ecological flow satisfaction degree score is 80. For seasonal rivers, the runoff length / water surface area retention rate can be determined by the runoff length / water surface area retention rate score method. Then, the corresponding ecological flow satisfaction degree score is determined based on the runoff length retention rate score standard threshold table. The runoff length retention rate score standard threshold table contains different runoff length retention rates corresponding to different runoff length retention rates. For example, when the runoff length retention rate is 85%, the corresponding ecological flow satisfaction degree score is 75. The ecological flow satisfaction degree score standard threshold table and the runoff length retention rate score standard threshold table can be determined by relevant technical personnel based on historical experimental data and uploaded to the evaluation system in advance.

[0079] b2, water quality degree score: the preset feature recognition algorithm can be used to identify the corresponding water quality category from the criterion index data corresponding to the water quality degree. Then, the corresponding water quality degree score is determined based on the water quality degree score standard threshold table. The water quality degree score standard threshold table contains different water quality degree scores corresponding to different water quality degrees. For example, when the water quality degree is secondary, the corresponding water quality degree score is 90. The water quality degree score standard threshold table can be determined by relevant technical personnel based on historical experimental data and uploaded to the evaluation system in advance.

[0080] b3, water self-purification ability score: the preset feature recognition algorithm can be used to identify the corresponding dissolved oxygen concentration from the criterion index data corresponding to the water self-purification ability. Then, the corresponding water self-purification ability score is determined based on the water self-purification ability score standard threshold table. The water self-purification ability score standard threshold table contains different water self-purification ability scores corresponding to different dissolved oxygen concentrations. For example, when the dissolved oxygen concentration is greater than or equal to 90%, the corresponding water self-purification ability score is 100. The water self-purification ability score standard threshold table can be determined by relevant technical personnel based on historical experimental data and uploaded to the evaluation system in advance.

[0081] The scoring strategy for the biological criterion layer in the preset index scoring neural network model can include:

[0082] c1, fish retention index score: according to the preset feature recognition algorithm, the number of fish species obtained by the river survey to be evaluated (excluding exotic species) and the historical number of fish species of the river to be evaluated can be identified from the criterion index data corresponding to the fish retention index, and the corresponding fish retention index is determined according to the fish retention index calculation formula, and finally the fish retention index score is determined based on the fish retention index score standard threshold table, wherein the fish retention index calculation formula is FOEI = FO / FE*100, wherein FOEI is the fish retention index, FO is the number of fish species obtained by the river survey to be evaluated, and FE is the historical number of fish species of the river to be evaluated. The fish retention index score standard threshold table contains the fish retention index score corresponding to different fish retention indexes, for example, when the fish retention index is 75%, the corresponding fish retention index score is 60 points. The fish retention index score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0083] c2, water bird condition score: according to the preset feature recognition algorithm, the species, number, and presence or absence of rare birds of the bird can be identified from the criterion index data corresponding to the water bird condition, and the corresponding water bird condition score is determined based on the bird habitat condition score standard threshold table, wherein the bird habitat condition score standard threshold table contains the water bird condition score corresponding to different combinations of the species, number, and presence or absence of rare birds of the bird, for example, when the species, number, and presence of rare birds are high, the corresponding water bird condition score is 95 points. The bird habitat condition score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0084] The scoring strategy of the preset index scoring neural network model for the social service function criterion layer can include:

[0085] d1, dike flood control standard reaching degree score: according to the preset feature recognition algorithm, the proportion of the length of the dike reaching the flood control standard to the total length of the dike, i.e. the flood control standard reaching rate, can be identified from the criterion index data corresponding to the dike flood control standard reaching degree, and the corresponding dike flood control standard reaching degree score is determined based on the dike flood control standard reaching degree score standard threshold table, wherein the dike flood control standard reaching degree score standard threshold table contains the dike flood control standard reaching degree score corresponding to different flood control standard reaching rates, for example, when the flood control standard reaching rate is greater than or equal to 95%, the corresponding dike flood control standard reaching degree score is 100 points. The dike flood control standard reaching degree score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0086] d2, flood control standard degree of the river-crossing building is given score: the flood control standard degree of the river-crossing building can be identified from the corresponding criterion index data of the flood control standard degree of the river-crossing building according to a preset feature recognition algorithm, and the corresponding flood control standard degree of the river-crossing building is determined based on a flood control standard degree of the river-crossing building score standard threshold table, wherein the flood control standard degree of the river-crossing building score standard threshold table contains the corresponding embankment flood control standard degree score of different flood control standard degrees of the river-crossing building, for example, when the flood control standard degree of the river-crossing building is greater than or equal to 95%, the corresponding flood control standard degree of the river-crossing building score is 100 points. The flood control standard degree of the river-crossing building score standard threshold table can be determined by relevant technical personnel according to historical experimental data and uploaded to the evaluation system in advance.

[0087] d3, shore management index score: the corresponding total length of the shore, the length of the developed and utilized shore, and the length of the utilized shore protected perfectly can be identified from the corresponding criterion index data of the shore management index according to a preset feature recognition algorithm, and the corresponding shore management and utilization index is calculated based on the calculation formula of the shore management and utilization index, wherein the calculation formula of the shore management and utilization index is , R u is the shore management and utilization index, L n is the total length of the shore, L u is the length of the developed and utilized shore, L 0 is the length of the utilized shore protected perfectly, and the shore management and utilization index score value = shore management and utilization index x 100.

[0088] d4, public satisfaction score: the average satisfaction degree of the public to the river environment, water quality and quantity, and water-related landscape can be identified from the corresponding criterion index data of the public satisfaction according to a preset feature recognition algorithm, and then it is determined based on the average satisfaction degree and the public satisfaction index score standard threshold table, wherein different average satisfaction degree intervals correspond to different public satisfaction scores in the public satisfaction index score standard threshold table, for example, the average satisfaction degree 85 falls into the average satisfaction degree interval corresponding to the public satisfaction score of 80 points.

[0089] The above method can obtain the index score value corresponding to each criterion index.

[0090] Step S130: obtaining the preset index weight and preset criterion layer weight corresponding to the river to be evaluated, and determining the river health comprehensive score corresponding to the river to be evaluated based on the preset index weight, preset criterion layer weight and each index score value corresponding to the river to be evaluated.

[0091] Specifically, the preset index weight and the preset criterion layer weight corresponding to different to-be-evaluated rivers are different, and can be acquired from the evaluation system based on the river identifier corresponding to the to-be-evaluated river, wherein the preset criterion layer weight is used to represent the importance of different criterion layers in the health evaluation process, and the preset index weight is used to represent the importance of each criterion index in the criterion layer. The embodiment of the present application provides a specific implementation manner of determining the river health comprehensive score corresponding to the to-be-evaluated river, that is, according to the preset index weight, the preset criterion layer weight, the index score value corresponding to each criterion index, and the first river health comprehensive score formula, the river health comprehensive score is determined, wherein the first river health comprehensive score formula is: wherein:

[0092] RHI i representing the river health comprehensive score corresponding to the to-be-evaluated river;

[0093] ZB nw representing the preset index weight of the n th index of the criterion layer;

[0094] ZB nr representing the index score value of the n th index of the criterion layer;

[0095] YMB mw representing the preset criterion layer weight of the m th criterion layer.

[0096] The calculation principle is to determine the criterion layer score corresponding to each criterion layer according to the preset index weight and the index score value corresponding to each criterion layer, to determine the score of each criterion layer according to the preset criterion layer weight and the criterion layer score, and finally to calculate the sum of the scores of each criterion layer corresponding to the criterion layer to obtain the river health comprehensive score of the to-be-evaluated river.

[0097] Step S140: determining the health condition level of the to-be-evaluated river based on the river health comprehensive score.

[0098] Specifically, different river health comprehensive scores correspond to different health status levels. When the river health comprehensive score is in the interval [90, 100), the corresponding health status level of the river can be determined as very healthy; when the river health comprehensive score is in the interval [75, 90), the corresponding health status level of the river can be determined as healthy; when the river health comprehensive score is in the interval [60, 75), the corresponding health status level of the river can be determined as sub-healthy; when the river health comprehensive score is in the interval [40, 60), the corresponding health status level of the river can be determined as unhealthy; and when the river health comprehensive score is in the interval [0, 40), the corresponding health status level of the river can be determined as poor. In order to intuitively view the health status level of the river to be evaluated, different colors can be used for representation. For example, blue, green, yellow, orange, and red can be used to represent very healthy, healthy, sub-healthy, unhealthy, and poor, respectively. The specific colors corresponding to different health status levels are not limited in the embodiments of the present application and can be set by relevant staff according to actual needs.

[0099] For the embodiments of the present application, the health of the river to be evaluated is evaluated by comprehensively considering the criterion index data of each criterion layer, avoiding the one-sidedness of health evaluation caused by relying on a single factor, reducing the interference of a single factor on the health evaluation result, ensuring the integrity of the evaluation perspective. In addition, each criterion index is scored by using the preset index scoring neural network, and the learning and fitting ability of the neural network model for complex data patterns is used to facilitate the deep mining of the potential nonlinear mapping relationship between each criterion index data and the corresponding index score value, and to get rid of the limitations of experience dependence of traditional manual scoring, thereby facilitating the improvement of the accuracy of the scoring result. On this basis, by integrating all the index scoring values with high accuracy and combining the scientific preset index weight and criterion layer weight for comprehensive calculation, the real contribution of each criterion index can be effectively integrated into the river health comprehensive score, avoiding the influence of local criterion index data error or unreasonable weight on the overall health evaluation result, thereby facilitating the improvement of the accuracy of the health evaluation of the river water ecosystem.

[0100] Further, when the river to be evaluated includes a segmented river, the health of each segmented river can be evaluated in the above manner, and the following can also be performed:

[0101] The number of segments of the segmented river corresponding to the river to be evaluated, the segment length of each segmented river, and the segment health comprehensive score of each segmented river are identified; and the river health comprehensive score corresponding to the river to be evaluated including the segmented river is determined based on the number of segments, the segment length of each segmented river, the segment health comprehensive score of each segmented river, and a second river health comprehensive score formula, where the second river health comprehensive score formula is:

[0102] wherein:

[0103] The RHI represents a river health comprehensive score of the river to be evaluated which comprises segmented rivers;

[0104] RHI i The segmented health comprehensive score represents the i-th segmented river;

[0105] W i The i-th segmented river and the segmented length are represented;

[0106] R s The number of segments of the segmented river is represented.

[0107] Specifically, the segmented health comprehensive score of each segmented river can be calculated by using the method provided in the above embodiment, and the number of segments and the segmented length of all the segmented rivers are identified, and each segmented health comprehensive score, the number of segments and the segmented length are introduced into the second river health comprehensive score formula to obtain the final river health comprehensive score of the river to be evaluated which comprises segmented rivers. Through the segmented integration processing, the data of each segmented river can be independently checked and scored, and the influence of local data errors on the overall result is weakened through weighted integration, so as to facilitate the accuracy of determining the river health comprehensive score of the river to be evaluated.

[0108] Further, the application provides a way of determining the preset index weight corresponding to each criterion index, which can specifically include steps S210-S230, as shown in the following table: Figure 2 The table shows that:

[0109] Step S210: Subjective weight calculation is performed based on the index score value of each criterion index to obtain a plurality of subjective index weights corresponding to each criterion index.

[0110] Specifically, three subjective weight calculation methods are provided in the embodiment of the application, which are expert scoring method, analytic hierarchy process and priority graph method.

[0111] Regarding the expert scoring method: the importance evaluation results of experts in the relevant field on each criterion layer and each criterion index can be obtained through questionnaire survey, which can specifically be that all criterion layers and each criterion index are listed, and a plurality of experts in the relevant field are asked to score the relative importance of each criterion layer and each criterion index. Finally, the scores of each criterion layer and each criterion index are normalized to obtain the first subjective weight determined by the expert scoring method, which comprises the first subjective weight of each criterion layer and the first subjective weight of each criterion index.

[0112] Regarding the analytic hierarchy process: the analytic hierarchy process takes a complex multi-objective decision-making problem as a system, decomposes the target into multiple criteria and multiple indexes, calculates the hierarchical single ranking (weight) and total ranking through a qualitative index fuzzy quantification method, and then optimizes the decision-making result of the system method. Specifically, the relative importance of each criterion layer and each criterion index can be determined by using the 1-9 point scale method through pairwise comparison with the help of expert scoring method, a judgment matrix is constructed, the eigenvalue and eigenvector of the judgment matrix are calculated, the eigenvector is normalized, and the second subjective weight determined by the analytic hierarchy process is obtained. The second subjective weight includes the second subjective weight of each criterion layer and the second subjective weight of each criterion index. Since the judgment matrix is determined with the help of expert scoring method, there may be a certain subjectivity, and logical errors may occur in the subsequent calculation process. At this time, the judgment matrix can be subjected to consistency check. If the data in the judgment matrix passes the consistency check, it means that the judgment matrix will not have logical errors in the subsequent calculation process. The CR value can be used to check the consistency of the judgment matrix. Specifically, the consistency index can be determined according to , wherein, CI is the consistency index, λ max is the maximum eigenvalue of the judgment matrix, and n is the matrix order of the judgment matrix. The consistency ratio can be determined according to , wherein, CR is the consistency ratio, RI is the average random consistency index, and when CR <0.1, it is determined that the judgment matrix passes the consistency check, otherwise it is determined that the judgment matrix does not pass the consistency check. The way of consistency check is not limited in the embodiments of the present application.

[0113] Regarding the priority graph method: the priority graph method analyzes the importance of each index by constructing a matrix graph. The index assignment values of the four rivers are input, and the specific steps can be: first, calculate the average value of each index, and count 1 point for the relatively larger value, 0 point for the relatively smaller value, and 0.5 point for the completely equal value; construct a matrix according to the calculated average value, and the elements in the matrix represent the relative importance between the indexes; normalize the data in the matrix, and finally obtain the third subjective weight determined by the priority graph method, which includes the third subjective weight of each criterion layer and the third subjective weight of each criterion index.

[0114] Step S220: Based on the index assignment values of each criterion index, the objective weight calculation is performed to obtain a plurality of objective index weights corresponding to each criterion index.

[0115] Specifically, the present application also provides four objective weight calculation methods, which are information weight method, CRITIC weight method, principal component analysis method and entropy weight method.

[0116] Regarding the information content weight method: the information content weight method, also known as the coefficient of variation method, calculates the weight through the coefficient of variation of the data. The larger the coefficient of variation, the more information it carries, so the weight will also increase accordingly. The specific steps can include: first, standardize the criterion index data to ensure that criterion indexes of different magnitudes can be compared; calculate the coefficient of variation of the standardized criterion index, which is the ratio of the standard deviation to the average value; and finally, according to the coefficient of variation of each criterion index, calculate the first objective weight determined by the information content weight method, which includes the first objective weight of each criterion layer and the first objective weight of each criterion index.

[0117] Regarding the CRITIC weight method: the CRITIC weight method is an objective weighting method based on the volatility of data and the conflict between criterion indexes. Its core idea is to determine the weight of each criterion layer and each criterion index by calculating the contrast intensity and conflict of the data of criterion indexes. The specific steps can be: calculate the standard deviation to measure the volatility of the data of criterion indexes. The larger the standard deviation, the greater the volatility of the data of criterion indexes, which should be given a higher weight. Then calculate the correlation coefficient to measure the conflict between criterion indexes. The larger the correlation coefficient, the smaller the conflict between criterion indexes, and the weight should be correspondingly reduced. Finally, normalize the calculation results of the contrast intensity and the conflict to obtain the second objective weight determined by the CRITIC weight method, which includes the second objective weight of each criterion layer and the second objective weight of each criterion index.

[0118] Regarding the principal component analysis method: principal component analysis is a mathematical transformation that reduces the dimensionality of criterion index data and extracts new variables that can explain most of the variation. These new variables are called principal components. The weights of these principal components are calculated by mathematics and do not depend on human subjective judgment, so they have high objectivity. The specific steps can include: first, standardize the criterion index data so that the mean of each feature is 0 and the variance is 1. Then, based on the standardized data, calculate the covariance matrix. Then, calculate the eigenvalues and eigenvectors from the covariance matrix. The eigenvectors represent the principal components, and the eigenvalues represent the variance contribution rate of the principal components. Finally, multiply the variance contribution rate of each principal component by the load coefficient of the principal component on the original variable, and normalize the result to obtain the third objective weight determined by the principal component analysis method, which includes the third objective weight of each criterion layer and the third objective weight of each criterion index.

[0119] Regarding the entropy weight method: the entropy weight method is a method based on the concept of information entropy to determine the weights of criteria and indexes. Based on the entropy weight method, the fourth objective weight is determined, which includes the fourth objective weight of each criterion layer and the fourth objective weight of each criterion index.

[0120] Step S230: weight fitting is performed on the plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion index to obtain a preset index weight corresponding to each criterion index.

[0121] Specifically, since the expert scoring method, the analytic hierarchy process and the priority graph method rely on human judgment, although professional experience and actual needs can be reflected, they may be affected by subjective preference, knowledge limitation or regional tendency, and there are problems of too strong subjectivity and insufficient result stability. The information weight method, the CRITIC weight method, principal component analysis and the entropy weight method are completely based on data feature analysis, although human bias can be avoided, but the actual business logic or professional experience may be ignored. Therefore, by fitting the two types of weights, the influence of subjective experience and objective data can be balanced, the rationality of professional judgment is retained, the objective law reflected by data is included, and the weight deviation caused by a single method is avoided. The plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion index can be fitted using a BP neural network, and the plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion index can also be fitted using an MLP model. The specific fitting process is not limited in the embodiments of the present application.

[0122] However, according to the historical fitting data, it can be known that the BP neural network has too good effect in the training process, the model is too dependent on the details and noise in the training data, and the model is very sensitive to the change of the training data. Once the training data changes, the performance of the model may be greatly reduced. The MLP is a parameterized model, and needs enough data to learn the mapping relationship between the input and the output. If the sample size is too small, the model is prone to underfitting, that is, it may not be able to capture the core pattern, resulting in poor fitting effect. Therefore, in order to further improve the accuracy of the weight fitting result, when fitting the plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion index to obtain the preset index weight corresponding to the criterion index, the following steps can be specifically included:

[0123] The plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion index are input as input data into a preset first neural network model and a preset second neural network model for weight fitting to obtain a first fitting weight and a second fitting weight. The preset first neural network model outputs the first fitting weight, and the preset second neural network model outputs the second fitting weight. Based on the performance indicators of the preset first neural network model and the preset second neural network model, a first weight contribution proportion corresponding to the preset first neural network model and a second weight contribution proportion corresponding to the preset second neural network model are determined. Based on the first fitting weight, the second fitting weight, the first weight contribution proportion and the second weight contribution proportion, a preset index weight corresponding to the criterion index is determined.

[0124] Specifically, for any criterion index, the preset first neural network model can be a BP neural network model, the preset second neural network model can be an MLP model, the performance index corresponding to the preset neural network model is used to quantitatively evaluate the fitting effect, stability and reliability of the model, the weight contribution ratio of different performance indexes is different, the performance index can be an error index or a correlation index, and the specific performance index can be determined by relevant personnel according to actual needs. When the performance index is a correlation index, a first difference between the correlation index of the preset first neural network model and 1 and a second difference between the correlation index of the preset second neural network model and 1 can be determined first, and the first weight contribution ratio and the second weight contribution ratio are determined according to the first difference and the second difference. The smaller the first difference or the second difference is, the closer the corresponding correlation index is to 1, which also indicates that the linear correlation between the fitting weight output by the corresponding neural network model and the subjective index weight and the objective index weight is stronger, and the neural network model has stronger explanation ability for the weight data. That is, the smaller the difference is, the greater the weight contribution ratio is.

[0125] Based on the performance indexes of the preset first neural network model and the preset second neural network model, the first weight contribution ratio and the second weight contribution ratio can be determined. Finally, the weight summation calculation is performed based on the first fitting weight, the second fitting weight, the first weight contribution ratio and the second weight contribution ratio, and the preset index weight of the criterion index can be obtained. The performance indexes of the preset first neural network model and the preset second neural network model can be uploaded to the evaluation system by relevant personnel in advance. There is a weight ratio corresponding relationship between the performance index and the weight contribution ratio. The weight ratio corresponding relationship can be determined by relevant personnel according to historical experimental data and uploaded to the evaluation system. For example, when the criterion index A obtains a first fitting weight of 0.06 under the fitting simulation of the preset first neural network model, a second fitting weight of 0.07 under the fitting simulation of the preset second neural network model, a first weight contribution ratio of 0.4 corresponding to the performance index of the preset first neural network model, and a second weight contribution ratio of 0.6 corresponding to the performance index of the preset second neural network model, the preset index weight of the criterion index A can be calculated as 0.06*0.4+0.07*0.6=0.066. The preset index weight corresponding to all criterion indexes can be obtained by using the above method.

[0126] An evaluation system is provided in the embodiments of the present application, as shown in Figure 3 , and Figure 3The evaluation system 300 shown includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, through a bus 302. Optionally, the evaluation system 300 can also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the evaluation system 300 does not constitute a limitation on the embodiments of the present application.

[0127] The processor 301 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure of the present application. The processor 301 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0128] The bus 302 can include a path for transmitting information between the above-mentioned components. The bus 302 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 302 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 only one line is used, but it does not mean that there is only one bus or one type of bus.

[0129] The memory 303 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0130] The memory 303 is configured to store application program codes for implementing the solutions of the present application, and the processor 301 is configured to control the execution of the application program codes. The processor 301 is configured to execute the application program codes stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0131] The evaluation system includes, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Personal Computers), PMPs (Portable Multimedia Players), and vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The evaluation system can also be a server or the like. Figure 3 The evaluation system shown is only an example, and should not limit the functions and use range of the embodiments of the present application.

[0132] The embodiments of the present application provide a computer readable storage medium, which stores a computer program. When the computer program is run on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0133] The embodiments of the present application provide a computer program product, which includes a computer program. When the computer program is executed by a processor, the method in any of the foregoing embodiments is implemented.

[0134] It should be understood that although the steps in the flowcharts of the drawings are shown in a sequential order following the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated otherwise herein, the execution of the steps is not strictly limited to the order indicated by the arrows, and can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of which is not necessarily sequential, but can be round-robin or alternating with at least some of the other steps or sub-steps or stages of other steps.

[0135] The above only describes some embodiments of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A method for assessing the health of a riverine aquatic ecosystem, characterized by, The method comprises the following steps: obtaining a plurality of criterion indicators and criterion indicator data corresponding to each criterion layer of a river to be evaluated; obtaining an index score value corresponding to each criterion indicator based on a preset index scoring neural network model and the criterion indicator data; obtaining a preset index weight and a preset criterion layer weight corresponding to the river to be evaluated, and determining a river health comprehensive score corresponding to the river to be evaluated based on the preset index weight, the preset criterion layer weight, and the index score value corresponding to each criterion indicator, wherein the preset criterion layer weight corresponding to the river to be evaluated is obtained based on a river identifier corresponding to the river to be evaluated; determining the health status grade of the river to be evaluated based on the river health comprehensive score; wherein the determination process of the preset index weight comprises: performing subjective weight calculation based on the index score value of each criterion indicator to obtain a plurality of subjective index weights corresponding to each criterion indicator; performing objective weight calculation based on the index score value of each criterion indicator to obtain a plurality of objective index weights corresponding to each criterion indicator; performing weight fitting on the plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion indicator to obtain a preset index weight corresponding to each criterion indicator; wherein the weight fitting on the plurality of subjective index weights and the plurality of objective index weights corresponding to each criterion indicator to obtain a preset index weight corresponding to the criterion indicator comprises: inputting the plurality of subjective index weights and the plurality of objective index weights corresponding to the criterion indicator as input data into a preset first neural network model and a preset second neural network model respectively for weight fitting to obtain a first fitted weight and a second fitted weight, wherein the preset first neural network model outputs the first fitted weight, and the preset second neural network model outputs the second fitted weight; determining a first weight contribution proportion corresponding to the preset first neural network model and a second weight contribution proportion corresponding to the preset second neural network model based on the performance indicators of the preset first neural network model and the preset second neural network model; determining the preset index weight corresponding to the criterion indicator based on the first fitted weight, the second fitted weight, the first weight contribution proportion, and the second weight contribution proportion.

2. The method for evaluating the health of river water ecosystem according to claim 1, characterized in that, The method comprises the following steps: identifying criterion indicator features corresponding to each criterion indicator data, and determining an index scoring strategy corresponding to each criterion indicator feature from the preset index scoring neural network model based on each criterion indicator feature; determining an index score value corresponding to each criterion indicator based on each criterion indicator data and the corresponding index scoring strategy.

3. The method for evaluating the health of river water ecosystem according to claim 1, characterized in that, The method comprises the following steps: According to the preset index weight, the preset criterion layer weight, the index score value corresponding to each criterion index, and a first river health comprehensive score formula, a river health comprehensive score is determined, wherein the first river health comprehensive score formula is: wherein: RHI i characterizes the comprehensive score of the river health corresponding to the ith river to be evaluated; ZB nw a preset index weight of the nth index of the characterization criterion layer; ZB nr an index score value characterizing the nth index of the criteria layer; YMB mw a preset criterion layer weight characterizing the mth criterion layer.

4. The method for evaluating the health of a river water ecosystem according to claim 3, wherein, when the river to be evaluated includes segmented rivers, the method further comprises the following steps: identifying the number of segments, the length of each segment, and the segment health comprehensive score of the segmented rivers corresponding to the river to be evaluated; determining a river health comprehensive score corresponding to the to-be-evaluated river containing the segmented river based on the number of segments, the segment length of each segmented river, the segment health comprehensive score, and a second river health comprehensive score formula, wherein the second river health comprehensive score formula is: wherein: The RHI represents a river health comprehensive score corresponding to the to-be-evaluated river containing the segmented river; RHI i a composite score characterizing the health of the ith segmented river; W i characterizing the ith segment river and the segment length; R s The number of segments characterizing the segmented river is characterized.

5. An evaluation system, characterized by The evaluation system comprises: at least one processor; a memory; at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to execute the river water ecosystem health evaluation method in any one of claims 1-4.

6. A computer-readable storage medium, characterized in that, comprising: a computer program stored in the memory and capable of being loaded and executed by the processor to perform the river water ecosystem health evaluation method in any one of claims 1-4.

7. A computer program product, characterised in that, comprising a computer program, which, when executed by the processor, implements the steps of the river water ecosystem health evaluation method in any one of claims 1-4.

Images