Intelligent evaluation method and system for seaweed carrying capacity based on data analysis

Abstract

The application relates to the technical field of data processing, and discloses a seaweed bearable capacity intelligent evaluation method and system based on data analysis. The method comprises the following steps: collecting environment parameters of a seaweed cultivation area through a multilevel sensor, and obtaining a standardized data set by using an outlier detection algorithm; dividing the cultivation area into environment layered sub-areas by using a clustering algorithm; training the relationship between the environment and the bearable capacity by using a machine learning algorithm to obtain regionalized prediction data; calculating three-dimensional indexes of ecology, economy and society to establish a sustainability evaluation index system; and constructing a neural network model to output the seaweed bearable capacity intelligent evaluation value. The application solves the problems of insufficient multidimensional data processing capability and missing regional differentiation evaluation in the existing seaweed bearable capacity evaluation method, and improves the accuracy and intelligent level of seaweed bearable capacity evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a seaweed bearable capacity intelligent evaluation method and system based on data analysis. BACKGROUND

[0002] With the rapid development of marine economy, seaweed cultivation as an important marine industry is expanding in scale. Traditional seaweed bearable capacity evaluation methods mainly rely on manual sampling and laboratory analysis. By regularly collecting seawater samples and seaweed samples for physicochemical analysis, combined with a simple biological model, the maximum cultivation capacity of the sea area is calculated. Some existing evaluation methods based on data analysis have appeared. A single environmental parameter monitoring and linear regression model are used to predict the seaweed carrying capacity. By collecting basic environmental data such as water temperature and salinity through sensors, a statistical method is used to establish a relationship model between environmental factors and seaweed growth.

[0003] However, the existing data analysis method has significant shortcomings. First, the data collection lacks systematicness and real-time nature, and can only obtain environmental information in a local area, making it difficult to reflect the spatial heterogeneity and temporal dynamics of the marine environment. Second, the existing method mainly uses a linear modeling approach, which cannot effectively capture the complex nonlinear relationship between environmental factors and seaweed growth in the marine ecosystem, and lacks consideration of the differences in environmental characteristics of different sea areas. Third, the traditional evaluation method only focuses on biological carrying capacity, ignoring the comprehensive impact of seaweed cultivation on ecological environment, economic benefits and social sustainability.

[0004] Based on the above analysis, it can be found that the fundamental problem of the existing technology is the lack of intelligent processing capability of multi-dimensional environmental data and regionalized difference evaluation mechanism. Due to the complexity of the marine environment and the multi-factor dependence of seaweed growth, a simple linear model cannot accurately predict the carrying capacity, and the lack of consideration of sustainability dimensions makes it difficult to guide long-term cultivation planning. More importantly, the existing method cannot effectively integrate ecological, economic and social indicators in multiple dimensions, and lacks an intelligent nonlinear aggregation algorithm to handle the complex interaction between multi-dimensional indicators, which directly affects the accuracy and practicality of seaweed bearable capacity evaluation. SUMMARY

[0005] The present application provides a seaweed bearable capacity intelligent evaluation method and system based on data analysis, which solves the problems of insufficient multi-dimensional data processing capability and lack of regional difference evaluation in existing seaweed bearable capacity evaluation methods, and improves the accuracy and intelligent level of seaweed bearable capacity evaluation.

[0006] In a first aspect, the present application provides a seaweed bearable capacity intelligent evaluation method based on data analysis, which comprises:

[0007] The S1 step is to collect water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration of the seaweed cultivation area through a multi-level sensor network, to remove abnormal data points by using an outlier detection algorithm, and to obtain a standardized marine environment data set.

[0008] The S2 step is to input the standardized marine environment data set into a clustering algorithm, to divide the cultivation area into a plurality of sub-areas according to environmental parameter similarity, and to form an environmental layered sub-area.

[0009] The S3 step is to collect seaweed growth rate and biomass data for each of the environmental layered sub-areas, to train the relationship between environmental parameters and seaweed carrying capacity by using a machine learning algorithm, and to obtain regionalized carrying capacity prediction data.

[0010] The S4 step is to calculate an ecological environment impact index, an economic benefit index and a social sustainability index respectively, to determine the weight coefficients of each dimension by using an analytic hierarchy process, and to establish a multi-dimensional sustainability evaluation index system.

[0011] The S5 step is to construct a neural network model, to take the multi-dimensional sustainability evaluation index system and the regionalized carrying capacity prediction data as input features, and to output an intelligent evaluation value of seaweed carrying capacity.

[0012] In a second aspect, the present application provides an intelligent evaluation system of seaweed carrying capacity based on data analysis, which comprises:

[0013] The removing module is configured to collect water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration of the seaweed cultivation area through a multi-level sensor network, to remove abnormal data points by using an outlier detection algorithm, and to obtain a standardized marine environment data set.

[0014] The input module is configured to input the standardized marine environment data set into a clustering algorithm, to divide the cultivation area into a plurality of sub-areas according to environmental parameter similarity, and to form an environmental layered sub-area.

[0015] The training module is configured to collect seaweed growth rate and biomass data for each of the environmental layered sub-areas, to train the relationship between environmental parameters and seaweed carrying capacity by using a machine learning algorithm, and to obtain regionalized carrying capacity prediction data.

[0016] The establishing module is configured to calculate an ecological environment impact index, an economic benefit index and a social sustainability index respectively, to determine the weight coefficients of each dimension by using an analytic hierarchy process, and to establish a multi-dimensional sustainability evaluation index system.

[0017] An output module is configured to build a neural network model, take the multi-dimensional sustainability evaluation index system and the regionalized carrying capacity prediction data as input features, and output an intelligent evaluation value of seaweed carrying capacity.

[0018] In a third aspect, an intelligent seaweed carrying capacity evaluation device based on data analysis is provided, which comprises a memory and at least one processor, and the memory stores instructions; the at least one processor invokes the instructions in the memory to enable the intelligent seaweed carrying capacity evaluation device based on data analysis to perform the intelligent seaweed carrying capacity evaluation method based on data analysis.

[0019] In a fourth aspect, a computer readable storage medium is provided, which stores instructions, and when the instructions are run on a computer, the computer performs the intelligent seaweed carrying capacity evaluation method based on data analysis.

[0020] In the technical scheme provided in the present application, the multi-level sensor network is used to collect the multi-dimensional environmental parameters such as water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration in the seaweed cultivation area, and the abnormal value detection algorithm is used to eliminate abnormal data points, so as to ensure the comprehensiveness and accuracy of data collection and provide a high-quality data basis for subsequent analysis. The clustering algorithm is used to divide the cultivation area into environmental layered sub-areas according to the similarity of environmental parameters, effectively solving the problem of spatial heterogeneity of marine environment, so that different environmental characteristic areas can be accurately evaluated in a differentiated manner. The relationship between environmental parameters and seaweed carrying capacity is trained by using a machine learning algorithm, and an intelligent prediction model is constructed, which can better capture the complex nonlinear relationship between environmental factors and seaweed growth compared with the traditional linear modeling method. The multi-dimensional sustainability evaluation index system established covers ecological environmental impact index, economic benefit index and social sustainability index, and the analytic hierarchy process is used to determine the weight coefficients of each dimension, so as to realize the change from single biological carrying capacity to comprehensive sustainability evaluation, and provide a more comprehensive decision basis for scientific planning of seaweed cultivation.

[0021] The construction of the neural network model fully utilizes its technical advantages in processing multi-dimensional nonlinear data. The multi-dimensional sustainability evaluation index system and the regionalized carrying capacity prediction data are taken as input features. Through the nonlinear transformation and weight learning of multiple neurons, the complex interaction mode and potential correlation between indicators can be automatically identified. Compared with the traditional linear weighted aggregation method, the neural network algorithm can effectively handle the pair-wise interaction and high-order nonlinear relationship between indicators, avoiding the information loss and evaluation deviation caused by simple weighted summation. Especially in the specific application field of intelligent evaluation of seaweed carrying capacity, the adaptive learning ability of neural network enables it to automatically adjust the model parameters according to the environmental characteristics and historical data of different sea areas, realize personalized carrying capacity prediction, and significantly improve the accuracy and reliability of the evaluation results. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor based on these drawings.

[0023] Figure 1 An embodiment schematic diagram of the intelligent evaluation method of seaweed carrying capacity based on data analysis in the embodiments of the present application;

[0024] Figure 2 A clustering effect evaluation result display diagram under different K values in the embodiments of the present application;

[0025] Figure 3 A multi-dimensional index system visualization scheme flowchart for seaweed aquaculture sustainability evaluation in the embodiments of the present application;

[0026] Figure 4 An embodiment schematic diagram of the intelligent evaluation system of seaweed carrying capacity based on data analysis in the embodiments of the present application;

[0027] Figure 5 A structural schematic block diagram of the intelligent evaluation equipment of seaweed carrying capacity based on data analysis in the embodiments of the present application. DETAILED DESCRIPTION

[0028] The embodiment of the present application provides a seaweed bearable capacity intelligent evaluation method and system based on data analysis. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0029] For ease of understanding, the specific process of the embodiment of the present application is described below. Please refer to Figure 1 One embodiment of the seaweed bearable capacity intelligent evaluation method based on data analysis in the embodiment of the present application comprises the following steps.

[0030] S1, collecting water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration of the seaweed cultivation area through a multi-level sensor network, and removing abnormal data points by using an outlier detection algorithm to obtain a standardized marine environment data set;

[0031] S2, inputting the standardized marine environment data set into a clustering algorithm, dividing the cultivation area into a plurality of sub-regions according to environmental parameter similarity, and forming an environmental layered sub-region;

[0032] S3, collecting seaweed growth rate and biomass data for each environmental layered sub-region, training the relationship between environmental parameters and seaweed bearing capacity by using a machine learning algorithm, and obtaining regionalized bearing capacity prediction data;

[0033] S4, calculating an ecological environment impact index, an economic benefit index and a social sustainability index respectively, determining the weight coefficients of each dimension by using an analytic hierarchy process, and establishing a multi-dimensional sustainability evaluation index system;

[0034] S5, constructing a neural network model, taking the multi-dimensional sustainability evaluation index system and the regionalized bearing capacity prediction data as input features, and outputting a seaweed bearable capacity intelligent evaluation value.

[0035] It can be understood that the execution subject of the present application can be a seaweed bearable capacity intelligent evaluation system based on data analysis, and can also be a terminal or a server, which is not limited here. The embodiment of the present application takes a server as an execution subject for example.

[0036] Specifically, the water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration in the seaweed cultivation area are collected by a multi-level sensor network, and the abnormal data points are removed by using an outlier detection algorithm to obtain a standardized marine environment data set. Then, the standardized marine environment data set is input into a clustering algorithm, and the cultivation area is divided into several sub-areas according to the similarity of environmental parameters, forming an environmental layered sub-area. On this basis, the seaweed growth rate and biomass data are collected for each environmental layered sub-area, and a machine learning algorithm is used to train the relationship between environmental parameters and seaweed carrying capacity to obtain regionalized carrying capacity prediction data. Then, the ecological environment impact index, economic benefit index and social sustainability index are calculated respectively, the weight coefficients of each dimension are determined by using the analytic hierarchy process, and a multi-dimensional sustainability evaluation index system is established. A neural network model is constructed, and the multi-dimensional sustainability evaluation index system and the regionalized carrying capacity prediction data are used as input features to output an intelligent evaluation value of seaweed carrying capacity.

[0037] In a specific embodiment, the S1 step further comprises:

[0038] The water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration are collected by a sensor array, and the collected data are sorted according to the time stamp to obtain time-series environmental parameter data.

[0039] Based on the 3σ criterion, the mean and standard deviation of the time-series environmental parameter data are calculated, and the data points deviating from the mean by more than three times the standard deviation are removed to obtain cleaned environmental parameter data.

[0040] The cleaned environmental parameter data is processed by using the minimum-maximum normalization algorithm to convert the numerical range to the interval of 0 to 1 to obtain normalized environmental parameter data.

[0041] The normalized environmental parameter data is subjected to integrity test, and the data integrity rate and accuracy error rate are calculated to obtain a standardized marine environment data set.

[0042] Specifically, the water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration are collected by the sensor array, and the collected data is sorted according to the timestamp to obtain the time-series environmental parameter data. On this basis, the mean and standard deviation of the time-series environmental parameter data are calculated based on the 3σ criterion, and the data points deviating from the mean by more than three times the standard deviation are removed to obtain the cleaned environmental parameter data. After cleaning the environmental parameter data, the minimum-maximum normalization algorithm is used to process the data, and the numerical range is converted to the interval of 0 to 1 to obtain the normalized environmental parameter data. After data normalization, the normalized environmental parameter data is subjected to integrity test, and the data integrity rate and accuracy error rate are calculated to obtain the standardized marine environmental data set. This process effectively ensures the accuracy and consistency of the data, providing a reliable data basis for subsequent analysis and modeling. Through this step, the collected environmental data not only has time series, but also effectively removes outliers, thereby improving the accuracy of subsequent analysis and prediction results.

[0043] For example, the water temperature data of the seaweed cultivation area can be collected by the sensor array, and assuming that the water temperature data of a certain area at a certain time point is: 22.5°C, 23.1°C, 22.8°C, 23.0°C, 25.5°C, 22.7°C, 23.3°C. Since the 5th data point (25.5°C) deviates significantly from other data points, after calculation based on the 3σ criterion, it is found that the deviation of this data point exceeds three times the standard deviation, so it will be removed, and the cleaned water temperature data set is obtained.

[0044] When the normalized data is subjected to integrity test, it is assumed that some data points of the light intensity data are not collected, resulting in data missing. At this time, by calculating the data integrity rate, it is assumed that the result is 85%, i.e. the proportion of data missing is 15%, and by calculating the accuracy error rate, it is found that the error rate of this data set is 3.5%. These test results will further affect the construction of the standardized data set.

[0045] In a specific embodiment, the S2 step further comprises:

[0046] The standardized marine environmental data set is subjected to principal component analysis processing, and the principal components with a cumulative contribution rate of 85% are extracted to obtain the dimension-reduced feature vector;

[0047] The dimension-reduced feature vector is input into the K-means++ clustering algorithm, and the optimal clustering number is determined by the silhouette coefficient method and the elbow rule to obtain the clustering parameter configuration;

[0048] Based on the clustering parameter configuration, the cultivation area is subjected to iterative clustering calculation, and the Euclidean distance is used as the similarity measurement standard to obtain the regional clustering result;

[0049] The effectiveness of the regional clustering results is verified, the coefficient of variation of environmental parameters in each sub-region and the ratio of within-cluster sum of squares to between-cluster sum of squares are calculated, and the environmental hierarchical sub-regions are obtained.

[0050] Specifically, the standardized marine environmental data set is subjected to principal component analysis, the variance contribution rate of each principal component is calculated, the principal components with a cumulative contribution rate of 85% are extracted, and the dimension reduction feature vectors are obtained. This step effectively reduces the dimension of the data while retaining the main environmental information, reducing the computational complexity. Subsequently, these dimension-reduced feature vectors are input into the K-means++ clustering algorithm, and the optimal number of clusters is determined by the silhouette coefficient method and the elbow rule. The silhouette coefficient method calculates the similarity of each data point with other points in the same cluster and the similarity with the nearest different cluster to evaluate the clustering quality. The elbow rule determines the optimal number of clusters by plotting the relationship curve between different numbers of clusters and within-cluster sum of squares. When the number of clusters is determined, the breeding area is iteratively clustered based on these clustering parameters, and the Euclidean distance is used as the similarity measurement standard to obtain multiple regional clustering results with similar environments. Next, the effectiveness of the regional clustering results is verified by calculating the coefficient of variation of environmental parameters in each sub-region to judge the consistency within the sub-region. At the same time, the ratio of within-cluster sum of squares to between-cluster sum of squares is calculated to evaluate the effectiveness of the clustering, ensuring that the environmental characteristics of each sub-region have significant differences, and the within-cluster difference is small and the between-cluster difference is large. Through these verifications, the environmental hierarchical sub-regions are obtained, providing clear regional division for subsequent seaweed growth prediction and carrying capacity evaluation.

[0051] In the principal component analysis of the standardized marine environmental data set, it is assumed that the original data set contains multiple environmental parameters such as water temperature, salinity, and dissolved oxygen concentration. Through calculation, the variance contribution rate of water temperature is 30%, the salinity is 25%, and the dissolved oxygen concentration is 20%, and other parameters have smaller contributions. The principal components with a cumulative contribution rate of 85% are extracted, and the three principal components of water temperature, salinity, and dissolved oxygen concentration are retained to obtain the dimension-reduced feature vectors. In this way, the originally multi-dimensional environmental data is compressed into three dimensions. In the application of the K-means++ clustering algorithm, it is assumed that during the clustering process, the scores under different numbers of clusters are calculated by the silhouette coefficient method, and the results show that when the number of clusters is 4, the silhouette coefficient is the largest, indicating the best clustering effect. Then, the elbow rule is used to plot the relationship curve between the number of clusters and the within-cluster sum of squares, and it is found that the inflection point appears at K=4, so the number of clusters is selected as 4. This clustering parameter configuration provides a basis for subsequent regional division.

[0052] For example, when performing regional clustering calculation, Euclidean distance is used as the similarity measurement standard. It is assumed that, in a specific region, the Euclidean distance of parameters such as water temperature, salinity, and dissolved oxygen concentration obtained through calculation is small, indicating that the environmental conditions of the region have high similarity with other regions. Therefore, these regions are divided into the same cluster, forming a sub-region with similar environmental characteristics.

[0053] In a specific embodiment, the execution step inputs the dimension-reduced feature vector into the K-means++ clustering algorithm, and the process of determining the optimal clustering number through the silhouette coefficient method and the elbow rule to obtain the clustering parameter configuration can specifically include the following steps:

[0054] The value range of the clustering number K is set to 3 to 8, and clustering calculation of different K values is performed on the dimension-reduced feature vector to obtain multiple clustering schemes;

[0055] The silhouette coefficient matrix is obtained by calculating the average distance of each sample point in each clustering scheme to other points in the same class and to the nearest points in the different class based on the silhouette coefficient method;

[0056] The intra-class sum of squares of each clustering scheme is calculated using the elbow rule, a relationship curve of K value and intra-class sum of squares is drawn, the inflection point position is determined, and a candidate optimal K value is obtained;

[0057] The silhouette coefficient matrix and the candidate optimal K value are comprehensively evaluated, the K value with the maximum silhouette coefficient and located near the inflection point is selected as the final clustering number, and the clustering parameter configuration is obtained.

[0058] Specifically, the value range of the clustering number K is set to 3 to 8, and first, multiple clustering calculations of different K values are performed on the dimension-reduced feature vector to obtain multiple clustering schemes. Each scheme is executed according to different K values, from K=3 to K=8, and the results of each clustering scheme are calculated. In the calculation process, the silhouette coefficient method is used to evaluate the average distance of each sample point to other points in the same class and to the nearest points in the different class, and the silhouette coefficient matrix is obtained. The larger the silhouette coefficient, the better the clustering effect, so by comparing the silhouette coefficient values under different K values, the clustering quality of each scheme can be effectively evaluated. At the same time, the elbow rule is used to calculate the intra-class sum of squares of each clustering scheme, and a relationship curve of K value and intra-class sum of squares is drawn. By observing the change of the curve, the inflection point position corresponding to the K value is determined, and the candidate optimal K value is the K value at the inflection point, i.e., the place where the intra-class sum of squares decreases most. At this time, the candidate optimal K value has been basically determined. Finally, the silhouette coefficient matrix and the candidate optimal K value are comprehensively evaluated, the K value with the maximum silhouette coefficient and located near the inflection point is selected as the final clustering number, and the quality and accuracy of the clustering are ensured. Through these steps, the best clustering parameter configuration is finally obtained, providing accurate basis for subsequent environmental layering and data analysis.

[0059] In an embodiment, the value range of the cluster number K is set to 3 to 8, assuming that the feature vectors after dimension reduction are environmental data of the seaweed cultivation area, including water temperature, salinity, and dissolved oxygen concentration, etc. After K-means++ clustering calculation, multiple clustering schemes are obtained. For example, when K = 3, the clustering result shows that the cultivation area is divided into three main categories, with obvious differences in water temperature and salinity, and less variation in dissolved oxygen concentration among different categories; when K = 5, the clustering result shows more subdivided areas, and the environmental characteristics of each area are more uniform.

[0060] In another embodiment, the silhouette coefficient of each clustering scheme is calculated based on the silhouette coefficient method. Assuming that when K = 4, the silhouette coefficient of a certain sub-area is 0.85, indicating that the clustering effect of this area is good, and when K = 6, the silhouette coefficient drops to 0.68, indicating that the clustering effect has declined. By comparing the silhouette coefficients under different K values, it can be clearly judged which K value corresponds to the best clustering effect. Referring to Figure 2 , this figure shows the clustering effect evaluation under different K values.

[0061] In a specific embodiment, the S3 step further comprises:

[0062] Deploying biomass monitoring equipment in each environmental stratification sub-area to collect seaweed growth rate and biomass density data, and classifying and marking the collected data according to the sub-area number to obtain regional seaweed biological data;

[0063] Associating and matching the regional seaweed biological data with the environmental parameter data of the corresponding sub-area to construct an environmental parameter-biomass paired data set and obtain a training sample data set;

[0064] Dividing the training sample data set into a training set and a test set according to a 7:3 ratio, training the training set using a random forest algorithm to obtain an environmental-carrying capacity mapping model;

[0065] Based on the environmental-carrying capacity mapping model, carrying capacity prediction calculation is performed on each environmental stratification sub-area, and the maximum carrying capacity value of each sub-area is output to obtain regionalized carrying capacity prediction data.

[0066] Specifically, biomass monitoring devices are deployed for each environmental stratification sub-region, and the seaweed growth rate and biomass density data of each sub-region are collected. For example, in a certain sub-region, the monitoring device records environmental data such as water temperature, light intensity, etc., while measuring the seaweed growth rate as 0.15 cm / day and the biomass density as 5 g / m². The collected data is classified and labeled according to the sub-region number, and the seaweed biological data set of each sub-region is obtained. These regional seaweed biological data are matched with the environmental parameter data of the corresponding sub-region to construct a paired data set between environmental parameters and biomass, thereby obtaining a training sample data set. These data provide a basis for subsequent model training. The training sample data set is divided into a training set and a test set according to a 7:3 ratio, and 70% of the data is selected for model training, and 30% of the data is used to test and verify the accuracy of the model. A random forest algorithm is used to train the model on the training set, and through the integration of multiple decision trees during the training process, a mapping model between the environment and the seaweed carrying capacity is obtained. This model can effectively capture the influence of environmental parameters on seaweed growth and predict the carrying capacity under different environmental conditions. Based on this mapping model, the carrying capacity of each environmental stratification sub-region is predicted and calculated, and the maximum carrying capacity value of each sub-region is obtained. For example, in a certain sub-region, the model predicts the maximum carrying capacity as 3000 kg / ha, and the carrying capacity of other sub-regions is also calculated and output accordingly, obtaining regionalized carrying capacity prediction data, which provides a scientific basis for further cultivation optimization.

[0067] For example, when deploying biomass monitoring devices, assume that in a certain environmental stratification sub-region, the device records water temperature as 24°C, light intensity as 1500 lux, seaweed growth rate as 0.2 cm / day, and biomass density as 8 g / m². According to the collected data, after classifying and labeling these data, the seaweed biological data of this sub-region will be obtained. Subsequently, these data are matched with the environmental parameters of this sub-region, such as water temperature, dissolved oxygen concentration, and salinity, etc., to construct a paired data set between environmental parameters and biomass. For example, the environmental parameters in a certain data set are water temperature 25°C, dissolved oxygen concentration 8 mg / L, and biomass density 7 g / m², which are used as input for training the model.

[0068] In another embodiment, assume that the training sample dataset contains 1000 sample points, 70% of which are used for training and the remaining 30% as a test set. The training set contains biomass data under different environmental parameters, such as water temperature from 22°C to 30°C and salinity from 10 ppt to 35 ppt, with a size of 700 data points. The random forest algorithm is used to train the model on the training set, and by integrating multiple decision trees, the relationship between environmental parameters and biomass density is fitted, and finally a mapping model of environment-carrying capacity is obtained. Based on this model, assume that for a certain sub-region, the maximum carrying capacity is predicted to be 3500 kg / ha, which reflects the maximum amount of seaweed growth that the region can support under current environmental conditions. In this way, the carrying capacity prediction data of each sub-region is obtained.

[0069] In a specific embodiment, the S4 step further comprises:

[0070] Based on the ratio of water quality pollutant concentration to environmental quality standard limit value, the ecological environment impact index is calculated, the Shannon-Wiener index is used to evaluate the biodiversity impact, and the ecological environment dimension index is obtained;

[0071] According to the product of seaweed yield and market price minus production cost, the unit area output value is calculated, and based on the ratio of annual net income to total investment, the investment return rate is calculated, and the economic benefit dimension index is obtained;

[0072] The employment contribution degree is calculated by the ratio of the number of jobs created to the total number of employed people in the region, and the input-output multiplier method is used to calculate the industry driving effect, and the social sustainability dimension index is obtained;

[0073] The ecological environment dimension index, economic benefit dimension index and social sustainability dimension index are input into the analytic hierarchy process for weight distribution calculation, a judgment matrix is constructed and consistency check is performed, a multi-dimensional sustainability evaluation index system is obtained, and reference Figure 3 The figure shows a multi-dimensional index system visualization scheme process for seaweed cultivation sustainability assessment.

[0074] Specifically, the ecological environmental impact index is calculated based on the ratio of water quality pollutant concentration to environmental quality standard limit value. For example, assuming that the water quality pollutant concentration of a certain seaweed cultivation area is 30 mg / L, and the environmental quality standard limit value is 40 mg / L, the ratio is 0.75. According to this ratio, the ecological environmental impact index of the area is 0.75, indicating that the water quality of the area is close to the standard limit value. Subsequently, the Shannon-Wiener index is used to evaluate the biodiversity impact, assuming that the biodiversity index of the area is 3.2, and the comprehensive index of the ecological environmental dimension is further calculated, and finally the evaluation value of the ecological environmental dimension of the area is obtained. In the economic benefit dimension evaluation, the unit area output value is calculated according to the product of seaweed yield and market price minus production cost. For example, the annual yield of seaweed in a certain area is 5000 kg / ha, the market price is 10 yuan / kg, and the production cost is 3000 yuan / ha, the unit area output value is 5000 x 10 - 3000 = 20000 yuan / ha. Then, the investment return rate is calculated based on the ratio of annual net income to total investment. Assuming that the annual net income of the area is 10000 yuan, and the total investment is 50000 yuan, the investment return rate is 10000 / 50000 = 0.2, indicating that the investment return rate is 20%. These data are combined to obtain the evaluation value of the economic benefit dimension of the area.

[0075] In the social sustainability dimension evaluation, the employment contribution degree is calculated by the ratio of the number of created employment positions to the total number of employed population in the area. For example, a certain area has created 50 employment positions, and the total employed population is 1000 people, then the employment contribution degree is 50 / 1000 = 0.05. In addition, the input-output multiplier method is used to calculate the industrial driving effect, assuming that the industrial driving effect of the area is 1.8, combined with the employment contribution degree to obtain the evaluation value of the social sustainability dimension of the area. The ecological environmental dimension index, economic benefit dimension index and social sustainability dimension index are input into the analytic hierarchy process for weight distribution calculation. By constructing a judgment matrix and conducting consistency test, the weight coefficients of each dimension are obtained, and finally a multi-dimensional sustainability evaluation index system is formed, ensuring the rationality and consistency of each index in the overall evaluation. This comprehensive evaluation system provides a comprehensive decision basis for the sustainable development of seaweed cultivation.

[0076] In a specific embodiment, the S5 step further comprises:

[0077] A three-layer feedforward neural network architecture is designed, and each dimension index in the multi-dimensional sustainability evaluation index system is used as an input layer node. The number of hidden layer neurons is set to obtain the neural network structure configuration;

[0078] The multi-dimensional sustainability evaluation index system and regionalized carrying capacity prediction data are combined to construct a neural network training dataset, and an interaction feature item between indexes is added as a supplementary input to obtain an extended feature dataset;

[0079] The Adam optimization algorithm is used to train the neural network, and the learning rate and batch size parameters are set. The network weights and biases are updated through the backpropagation algorithm to obtain a trained neural network model.

[0080] The sustainability index data of the region to be evaluated is input into the trained neural network model for forward calculation, and the output layer activation function is processed to obtain an intelligent evaluation value of the seaweed carrying capacity.

[0081] Specifically, a three-layer feedforward neural network architecture is designed, and each dimension index in the multi-dimensional sustainability evaluation index system is used as an input layer node. The number of hidden layer neurons is set to obtain a neural network structure configuration. Assuming that the input layer is: where each represents a different dimension evaluation index, and is the dimension number of the input feature. The number of hidden layer neurons is set to , and the output layer is the intelligent evaluation value of the seaweed carrying capacity.

[0082] The multi-dimensional sustainability evaluation index system and regionalized carrying capacity prediction data are combined to construct a neural network training dataset, and an interaction feature item between indexes is added as a supplementary input to obtain an extended feature dataset where is the interaction feature item between indexes. The input dimension of the neural network increases from to .

[0083] In the training process, the Adam optimization algorithm is used to train the neural network, and the optimization goal is to minimize the loss . The loss function is commonly expressed as mean square error (MSE), i.e.: where is the number of training samples, is the actual value of the training sample, is the model prediction value. The Adam optimization algorithm adjusts the network weights and biases to minimize the loss function. In each iteration, the weights and biases are updated using the backpropagation algorithm, and the formula is: where is the current weight, is the updated weight, is the learning rate, and the first moment estimation and the second moment estimation, respectively, is a minimum value, which is used to prevent division by zero error.

[0084] After the training is completed, the sustainability index data of the region to be evaluated is input into the trained neural network model for forward calculation. The process of forward propagation can be represented as: , , , , , wherein, is the weighted input of each layer, is the extended input data set, is the weight matrix of the neural network, corresponding to the weight from the input layer to the first hidden layer, from the first hidden layer to the second hidden layer, and from the second hidden layer to the output layer, respectively. is the bias term, corresponding to the bias of each layer, is the activation value of each layer, is the activation function, and the intelligent evaluation value of the seaweed carrying capacity is obtained through the output layer activation function , which represents the output of the sustainability index of the region to be evaluated in the neural network model.

[0085] Taking the seaweed carrying capacity evaluation as an example, a three-layer feedforward neural network architecture is designed. The input layer includes various dimensions of the sustainability evaluation index system, such as water temperature, light intensity, and nutrient salt concentration, etc. It is assumed that these indicators are input features. The number of hidden layer neurons is set to 128, and the best neural network structure configuration is obtained through experimental tuning. The training data set is combined with the regional carrying capacity prediction data by adding the interaction feature items between the indicators to obtain the extended feature data set, so that the input dimension of the neural network is improved. The Adam optimization algorithm is used to train the neural network, and the model is optimized by minimizing the mean square error (MSE) loss function. During the training process, the weights and biases of the network are constantly updated through the backpropagation algorithm until the loss function converges. After the training is completed, the sustainability index data of the region to be evaluated is input into the trained neural network model for forward calculation, and the model is processed through the output layer activation function to finally obtain the intelligent evaluation value of the seaweed carrying capacity, which is the prediction result of the seaweed carrying capacity of the region to be evaluated.

[0086] The above describes the seaweed carrying capacity intelligent evaluation method based on data analysis in the embodiments of the present application, and the following describes the seaweed carrying capacity intelligent evaluation system based on data analysis in the embodiments of the present application. Please refer to Figure 4 ​In the embodiment of the present application, one embodiment of the seaweed bearable capacity intelligent evaluation system based on data analysis includes:

[0087] The culling module is configured to collect water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient salt concentration in the seaweed cultivation area through the multi-level sensor network, and to obtain a standardized marine environment data set by using an outlier detection algorithm to remove abnormal data points.

[0088] The input module is configured to input the standardized marine environment data set into a clustering algorithm, divide the cultivation area into a plurality of sub-regions according to environmental parameter similarity, and form environment layered sub-regions.

[0089] The training module is configured to collect seaweed growth rate and biomass data for each of the environment layered sub-regions, train the relationship between environmental parameters and seaweed bearable capacity by using a machine learning algorithm, and obtain regionalized bearable capacity prediction data.

[0090] The establishment module is configured to calculate an ecological environment impact index, an economic benefit index and a social sustainability index respectively, determine the weight coefficients of each dimension by using an analytic hierarchy process, and establish a multi-dimensional sustainability evaluation index system.

[0091] The output module is configured to construct a neural network model, take the multi-dimensional sustainability evaluation index system and the regionalized bearable capacity prediction data as input features, and output a seaweed bearable capacity intelligent evaluation value.

[0092] The above Figure 4 The seaweed bearable capacity intelligent evaluation system based on data analysis in the embodiment of the present application is described in detail from the perspective of modular functional entities, and the seaweed bearable capacity intelligent evaluation device based on data analysis in the embodiment of the present application is described in detail from the perspective of hardware processing.

[0093] Referring to Figure 5 In the embodiment of the present application, a seaweed bearable capacity intelligent evaluation device based on data analysis is also provided. Figure 5The data analysis based seaweed loadable capacity intelligent evaluation device is shown in the figure. The data analysis based seaweed loadable capacity intelligent evaluation device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the computer designed processor is used to provide computing and control capability. The memory of the data analysis based seaweed loadable capacity intelligent evaluation device includes non-volatile storage medium and internal memory. The non-volatile storage medium stores operating system, computer program and database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the data analysis based seaweed loadable capacity intelligent evaluation device is used to store the corresponding data in this embodiment. The network interface of the data analysis based seaweed loadable capacity intelligent evaluation device is used to communicate with the external terminal through the network connection. The computer program is executed by the processor to realize the above method.

[0094] Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the application, and does not constitute a limitation on the data analysis based seaweed loadable capacity intelligent evaluation device to which the scheme of the application is applied.

[0095] The application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium, and can also be a volatile computer readable storage medium. The computer readable storage medium stores instructions, and when the instructions run on the computer, the computer executes the steps of the data analysis based seaweed loadable capacity intelligent evaluation method.

[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, system and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

[0097] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a data analysis based seaweed loadable capacity intelligent evaluation device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0098] The above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data-driven intelligent assessment method for the carrying capacity of seaweed, characterized in that, The method includes: Step S1: Collect water temperature, salinity, dissolved oxygen concentration, light intensity, and nutrient concentration in the seaweed cultivation area through a multi-level sensor network, and use an outlier detection algorithm to remove outlier data points to obtain a standardized marine environment dataset. Step S2: Input the standardized marine environment dataset into a clustering algorithm to divide the aquaculture area into several sub-regions according to the similarity of environmental parameters, forming environmental stratified sub-regions; Step S3 involves collecting algal growth rate and biomass data for each of the aforementioned environmental stratification sub-regions, and using machine learning algorithms to train the relationship between environmental parameters and algal carrying capacity to obtain regionalized carrying capacity prediction data. This includes: deploying biomass monitoring equipment in each of the aforementioned environmental stratification sub-regions to collect algal growth rate and biomass density data; classifying and labeling the collected data according to sub-region numbers to obtain regional algal biological data; associating and matching the regional algal biological data with the corresponding sub-region environmental parameter data to construct a paired dataset of environmental parameters and biomass to obtain a training sample dataset; dividing the training sample dataset into a training set and a test set at a ratio of 7:3, and using a random forest algorithm to train the model on the training set to obtain an environment-carrying capacity mapping model; and performing carrying capacity prediction calculations for each environmental stratification sub-region based on the environment-carrying capacity mapping model, outputting the maximum carrying capacity value of each sub-region, and obtaining regionalized carrying capacity prediction data. Step S4 involves calculating the ecological environment impact index, economic benefit index, and social sustainability index separately. The analytic hierarchy process (AHP) is used to determine the weight coefficients for each dimension, establishing a multi-dimensional sustainability assessment index system. This includes: calculating the ecological environment impact index based on the ratio of water pollutant concentration to environmental quality standard limits; assessing biodiversity impact using the Shannon-Wiener index to obtain ecological environment dimension indicators; calculating the output value per unit area by subtracting production costs from the product of seaweed production and market price; calculating the return on investment based on the ratio of annual net income to total investment to obtain economic benefit dimension indicators; calculating the employment contribution by the ratio of job creation to the total regional employment population; and calculating the industrial driving effect using the input-output multiplier method to obtain social sustainability dimension indicators; and inputting the ecological environment dimension indicators, economic benefit dimension indicators, and social sustainability dimension indicators into the AHP for weight allocation calculation, constructing a judgment matrix, and performing consistency checks to obtain the multi-dimensional sustainability assessment index system. Step S5: Construct a neural network model, using the multi-dimensional sustainability assessment index system and the regional carrying capacity prediction data as input features, and output an intelligent assessment value of the seaweed carrying capacity.

2. The intelligent assessment method for seaweed carrying capacity based on data analysis according to claim 1, characterized in that, Step S1 further includes: Water temperature, salinity, dissolved oxygen concentration, light intensity, and nutrient concentration are collected by a sensor array. The collected data are sorted according to timestamps to obtain time-series environmental parameter data. The mean and standard deviation of the time-series environmental parameter data are calculated based on the 3σ criterion. Data points that deviate from the mean by more than three times the standard deviation are removed to obtain the cleaned environmental parameter data. The environmental parameter data after cleaning is processed using a minimum-maximum normalization algorithm to convert the numerical range to the interval between 0 and 1, thereby obtaining normalized environmental parameter data. The normalized environmental parameter data are subjected to integrity checks, and the data integrity rate and accuracy error rate are calculated to obtain a standardized marine environmental dataset.

3. The intelligent assessment method for seaweed carrying capacity based on data analysis according to claim 1, characterized in that, Step S2 further includes: Principal component analysis was performed on the standardized marine environment dataset to extract the principal components with a cumulative contribution rate of 85%, resulting in a dimensionality-reduced feature vector. The reduced feature vector is input into the K-means++ clustering algorithm, and the optimal number of clusters is determined by the silhouette coefficient method and the elbow rule to obtain the clustering parameter configuration. Based on the clustering parameter configuration, iterative clustering calculations are performed on the aquaculture area, and Euclidean distance is used as the similarity metric to obtain the regional clustering results. The validity of the clustering results is verified by calculating the coefficient of variation of environmental parameters and the ratio of intra-class sum of squares to inter-class sum of squares in each sub-region, thus obtaining the environmental stratification sub-regions.

4. The intelligent assessment method for seaweed carrying capacity based on data analysis according to claim 3, characterized in that, The step involves inputting the dimensionality-reduced feature vector into the K-means++ clustering algorithm, determining the optimal number of clusters using the silhouette coefficient method and the elbow rule, and obtaining the clustering parameter configuration, including: The number of clusters K is set to range from 3 to 8. Clustering calculations with different K values ​​are performed on the dimensionality-reduced feature vectors to obtain multiple clustering schemes. Based on the silhouette coefficient method, the average distance from each sample point to other points of the same category and the average distance to the nearest point of a different category in each of the clustering schemes are calculated to obtain the silhouette coefficient matrix. The elbow rule is used to calculate the sum of squares within each clustering scheme, and the relationship curve between the K value and the sum of squares within each cluster is plotted. The inflection point of the curve is determined to obtain the candidate optimal K value. The silhouette coefficient matrix and the candidate optimal K values ​​are comprehensively evaluated, and the K value with the largest silhouette coefficient and located near the inflection point is selected as the final number of clusters, thus obtaining the clustering parameter configuration.

5. The intelligent assessment method for seaweed carrying capacity based on data analysis according to claim 1, characterized in that, Step S5 further includes: A three-layer feedforward neural network architecture is designed, with each dimension index in the multi-dimensional sustainability assessment index system as the input layer node, and the number of hidden layer neurons is set to obtain the neural network structure configuration. The multi-dimensional sustainability assessment index system and the regional carrying capacity prediction data are combined to construct a neural network training dataset. Interaction feature terms between the indicators are added as supplementary inputs to obtain an extended feature dataset. The neural network is trained using the Adam optimization algorithm. The learning rate and batch size parameters are set, and the network weights and biases are updated through the backpropagation algorithm to obtain the trained neural network model. The sustainability index data of the area to be evaluated is input into the trained neural network model for forward calculation. The data is then processed by the output layer activation function to obtain the intelligent evaluation value of the seaweed carrying capacity.

6. A data analysis-based intelligent assessment system for seaweed carrying capacity, characterized in that, For implementing the intelligent assessment method for seaweed carrying capacity based on data analysis as described in any one of claims 1 to 5, the intelligent assessment system for seaweed carrying capacity based on data analysis comprises: The elimination module is used to collect water temperature, salinity, dissolved oxygen concentration, light intensity and nutrient concentration in the seaweed cultivation area through a multi-level sensor network, and use an outlier detection algorithm to eliminate outlier data points to obtain a standardized marine environment dataset. The input module is used to input the standardized marine environment dataset into a clustering algorithm to divide the aquaculture area into several sub-regions according to the similarity of environmental parameters, forming environmental stratified sub-regions; The training module is used to collect algal growth rate and biomass data for each of the aforementioned environmental stratification sub-regions, and to train the relationship between environmental parameters and algal carrying capacity using machine learning algorithms to obtain regionalized carrying capacity prediction data. This includes: deploying biomass monitoring equipment in each of the aforementioned environmental stratification sub-regions to collect algal growth rate and biomass density data; classifying and labeling the collected data according to sub-region numbers to obtain regional algal biological data; associating and matching the regional algal biological data with the corresponding sub-region environmental parameter data to construct a paired dataset of environmental parameters and biomass, obtaining a training sample dataset; dividing the training sample dataset into a training set and a test set at a 7:3 ratio, and using a random forest algorithm to train the model on the training set to obtain an environment-carrying capacity mapping model; and calculating the carrying capacity prediction for each environmental stratification sub-region based on the environment-carrying capacity mapping model, outputting the maximum carrying capacity value for each sub-region, and obtaining regionalized carrying capacity prediction data. A module is established to calculate the ecological environment impact index, economic benefit index, and social sustainability index respectively. The analytic hierarchy process (AHP) is used to determine the weight coefficients of each dimension, establishing a multi-dimensional sustainability assessment index system. This includes: calculating the ecological environment impact index based on the ratio of water pollutant concentration to environmental quality standard limits, and assessing biodiversity impact using the Shannon-Wiener index to obtain ecological environment dimension indicators; calculating the output value per unit area by subtracting production costs from the product of seaweed production and market price, and calculating the return on investment based on the ratio of annual net income to total investment to obtain economic benefit dimension indicators; calculating the employment contribution by the ratio of job creation to the total regional employment population, and calculating the industrial driving effect using the input-output multiplier method to obtain social sustainability dimension indicators; inputting the ecological environment dimension indicators, economic benefit dimension indicators, and social sustainability dimension indicators into the AHP for weight allocation calculation, constructing a judgment matrix, and performing consistency checks to obtain the multi-dimensional sustainability assessment index system. The output module is used to construct a neural network model, taking the multi-dimensional sustainability assessment index system and the regional carrying capacity prediction data as input features, and outputting an intelligent assessment value of the seaweed carrying capacity.

7. A data analysis-based intelligent assessment device for seaweed carrying capacity, characterized in that, It includes a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the intelligent assessment method for seaweed carrying capacity based on data analysis as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to execute the intelligent assessment method for seaweed carrying capacity based on data analysis as described in any one of claims 1 to 5.

Images