Aquaculture intelligent management system based on big data and artificial intelligence

Abstract

The invention provides an intelligent aquaculture management system based on big data and artificial intelligence, relates to the technical field of intelligent aquaculture management, and aims to generate an aquaculture monitoring system, perform regional aquaculture monitoring, further perform comprehensive aquaculture monitoring, train an aquaculture effect prediction model and calculate regional aquaculture coefficients of a preset region. Obtaining an area standard deviation coefficient of the preset area, and obtaining abnormal breeding area information; calculating a dimension breeding coefficient of the dimension of the preset area, further obtaining a dimension standard deviation coefficient of the corresponding dimension, and obtaining abnormal breeding dimension information; adjusting the abnormal breeding dimension of the abnormal breeding area according to the preset dimension adjusting parameters, predicting the effect through the breeding effect prediction model, and further judging whether the adjustment is continued or not. By integrating big data and an artificial intelligence technology, real-time monitoring of a breeding environment, accurate prediction of a breeding effect and automatic identification and adjustment of an abnormal breeding state are realized.

Description

TECHNICAL FIELD

[0001] The application provides an intelligent management system for aquaculture based on big data and artificial intelligence, and relates to the technical field of intelligent management of aquaculture. BACKGROUND

[0002] Traditional aquaculture management often relies on the experience and intuition of breeders, lacking scientific data support and accurate decision-making basis. This management method is not only inefficient, but also easily affected by environmental changes and diseases, etc., leading to increased breeding risks. Existing aquaculture monitoring systems can only monitor and alarm a single environmental parameter, and cannot comprehensively consider the mutual influence between multiple parameters and the comprehensive influence on breeding effect. In addition, these systems often lack intelligent prediction and decision-making functions, and cannot automatically adjust breeding strategies according to real-time monitoring data. SUMMARY

[0003] The application provides an intelligent management system for aquaculture based on big data and artificial intelligence to solve the above problems:

[0004] The application provides an intelligent management system for aquaculture based on big data and artificial intelligence, which comprises:

[0005] The system area prediction module is used to obtain multiple preset area information of rivers and streams, generate an aquaculture monitoring system, conduct regional aquaculture monitoring, and then conduct comprehensive aquaculture monitoring, train a breeding effect prediction model, and output breeding effect prediction data.

[0006] The abnormal area analysis module is used to calculate the regional aquaculture coefficient of the preset area, and then obtain the regional standard deviation coefficient of the preset area, mark the regional discrete state of the preset area according to the regional standard deviation coefficient, and then obtain the abnormal aquaculture area information.

[0007] The abnormal dimension analysis module is used to calculate the dimension aquaculture coefficient of the dimension of the preset area, and then obtain the dimension standard deviation coefficient of the corresponding dimension, mark the dimension discrete state of the corresponding dimension according to the dimension standard deviation coefficient, and then obtain the abnormal aquaculture dimension information.

[0008] The regional dimension adjustment module is used to adjust the abnormal aquaculture dimension of the abnormal aquaculture area by a preset dimension adjustment parameter, predict the effect by the breeding effect prediction model, and then determine whether to continue adjusting.

[0009] Further, the system area prediction module comprises:

[0010] The regional integration module is used to obtain multiple preset area information of rivers or streams, combine the preset area information, and obtain regional integration information.

[0011] a system area monitoring module configured to generate an aquaculture monitoring system according to the area integration information, and monitor each preset area according to the aquaculture monitoring system to obtain area aquaculture monitoring data;

[0012] combine the area aquaculture monitoring data to generate comprehensive aquaculture monitoring data;

[0013] a model effect prediction module configured to obtain historical comprehensive aquaculture monitoring data, and train an aquaculture effect prediction model;

[0014] predict the aquaculture effect of real-time comprehensive aquaculture monitoring data by using the aquaculture effect prediction model, and obtain aquaculture effect prediction data.

[0015] Further, the abnormal area analysis module comprises:

[0016] an area aquaculture calculation module configured to calculate an area aquaculture coefficient of each preset area according to the area aquaculture monitoring data of each preset area;

[0017] an area standard deviation calculation module configured to calculate an area standard deviation coefficient of each area aquaculture coefficient according to the area aquaculture coefficient;

[0018] an abnormal area obtaining module configured to compare the area standard deviation coefficient with a preset area standard coefficient threshold, and obtain an area dispersion comparison result of the preset area;

[0019] label the area dispersion state of the preset area corresponding to the area standard deviation coefficient according to the area dispersion comparison result, and obtain area dispersion state labeling information;

[0020] obtain abnormal aquaculture area information according to the area dispersion state labeling information.

[0021] Further, the abnormal dimension analysis module comprises:

[0022] a dimension aquaculture calculation module configured to obtain a preset aquaculture monitoring dimension, and collect multi-dimensional aquaculture monitoring data of the preset area according to the preset aquaculture monitoring dimension to obtain dimension aquaculture monitoring data of each preset area in each dimension;

[0023] calculate a dimension aquaculture coefficient of each dimension of each preset area according to the dimension aquaculture monitoring data of each dimension of each preset area;

[0024] a dimension standard deviation calculation module configured to calculate a dimension standard deviation coefficient of each dimension aquaculture coefficient of each preset area according to the dimension aquaculture coefficient;

[0025] An abnormal dimension obtaining module is configured to compare the dimension standard deviation coefficient with a preset dimension standard coefficient threshold to obtain a dimension dispersion comparison result of the preset area;

[0026] According to the dimension dispersion comparison result, the dimension standard deviation coefficient corresponding to the preset area is marked to obtain dimension dispersion state marking information;

[0027] According to the dimension dispersion state marking information, abnormal breeding dimension information is obtained.

[0028] Further, the area dimension adjusting module comprises:

[0029] An adjusting parameter obtaining module is configured to obtain abnormal breeding dimension information of the abnormal breeding area, and obtain a preset dimension adjusting parameter according to the abnormal breeding dimension information of the preset breeding area;

[0030] An adjusting effect prediction module is configured to input the preset dimension parameter into a breeding effect prediction model to obtain predicted breeding effect data of the corresponding abnormal breeding dimension of the corresponding abnormal breeding area;

[0031] The predicted breeding effect data is compared with a preset effect threshold to obtain an effect comparison result;

[0032] An adjusting parameter replacement module is configured to trigger replacement of the preset dimension adjusting parameter when the predicted breeding effect data is less than the preset effect threshold, until the predicted breeding effect data is greater than or equal to the preset effect threshold, and the replacement of the preset dimension adjusting parameter is stopped.

[0033] Further, the management method comprises:

[0034] S1, obtain a plurality of preset area information of rivers and streams, generate an aquaculture monitoring system, perform area breeding monitoring, and further perform comprehensive breeding monitoring, train a breeding effect prediction model, and output breeding effect prediction data;

[0035] S2, calculate the area breeding coefficient of the preset area, and further obtain the area standard deviation coefficient of the preset area, mark the area dispersion state of the preset area according to the area standard deviation coefficient, and further obtain abnormal breeding area information;

[0036] S3, calculate the dimension breeding coefficient of the dimension of the preset area, and further obtain the dimension standard deviation coefficient of the corresponding dimension, mark the dimension dispersion state of the corresponding dimension according to the dimension standard deviation coefficient, and further obtain abnormal breeding dimension information;

[0037] S4, adjust the abnormal breeding dimension of the abnormal breeding area with a preset dimension adjusting parameter, predict the effect through the breeding effect prediction model, and further determine whether to continue adjusting.

[0038] Further, S1 includes:

[0039] Obtain information on multiple preset areas of a river or stream, and combine the preset area information to obtain integrated area information;

[0040] An aquaculture monitoring system is generated based on the regional integrated information. Aquaculture monitoring is then conducted in each preset area based on the aquaculture monitoring system to obtain regional aquaculture monitoring data.

[0041] The aquaculture monitoring data of the aforementioned region are combined to generate comprehensive aquaculture monitoring data;

[0042] Acquire historical comprehensive aquaculture monitoring data and train aquaculture performance prediction model;

[0043] The aquaculture effect prediction model is used to predict the aquaculture effect based on real-time comprehensive aquaculture monitoring data, thereby obtaining aquaculture effect prediction data.

[0044] Further, S2 includes:

[0045] Calculate the regional aquaculture coefficient for each preset region based on the regional aquaculture monitoring data of each preset region;

[0046] Calculate the regional standard deviation coefficient of the aquaculture coefficient for each region based on the aforementioned regional aquaculture coefficient;

[0047] The regional standard deviation coefficient is compared with a preset regional standard coefficient threshold to obtain the regional discrete comparison result of the preset region.

[0048] Based on the regional discrete comparison results, the preset region corresponding to the regional standard deviation coefficient is labeled with regional discrete state to obtain regional discrete state labeling information.

[0049] Information on abnormal aquaculture areas is obtained based on the discrete state labeling information of the region.

[0050] Further, S3 includes:

[0051] Obtain preset aquaculture monitoring dimensions, and collect multi-dimensional aquaculture monitoring data for preset areas based on the preset aquaculture monitoring dimensions to obtain multi-dimensional aquaculture monitoring data for each preset area;

[0052] The dimensional aquaculture coefficient for each dimension of each preset area is calculated based on the dimensional aquaculture monitoring data of multiple dimensions for each preset area.

[0053] Calculate the dimensional standard deviation coefficient of each dimensional aquaculture coefficient for each preset region based on the dimensional aquaculture coefficient;

[0054] The dimension standard deviation coefficient is compared with a preset dimension standard coefficient threshold to obtain a dimension dispersion comparison result of the preset area;

[0055] According to the dimension dispersion comparison result, the dimension dispersion state of the preset area corresponding to the dimension standard deviation coefficient is labeled to obtain dimension dispersion state labeling information;

[0056] According to the dimension dispersion state labeling information, abnormal breeding dimension information is obtained.

[0057] Further, the S4 comprises:

[0058] Abnormal breeding dimension information of the abnormal breeding area is obtained, and preset dimension adjustment parameters are obtained according to the abnormal breeding dimension information of the preset breeding area;

[0059] The preset dimension parameters are input into a breeding effect prediction model to obtain prediction breeding effect data of the corresponding abnormal breeding dimension corresponding to the abnormal breeding area;

[0060] The prediction breeding effect data is compared with a preset effect threshold to obtain an effect comparison result;

[0061] When the prediction breeding effect data is less than the preset effect threshold, the replacement of the preset dimension adjustment parameters is triggered until the prediction breeding effect data is greater than or equal to the preset effect threshold, and the replacement of the preset dimension adjustment parameters is stopped.

[0062] The present application has the following advantages: by monitoring and predicting the breeding effect in real time, the system can timely find and solve the problems existing in the breeding process, thereby improving the breeding efficiency and yield, and can identify and warn the abnormal breeding area and dimension, helping the breeder to take timely measures to reduce the breeding risk caused by environmental mutation or improper management. Through accurate data analysis and prediction, the breeder can be guided to reasonably allocate feed, water resources and other production materials, and realize the optimal allocation of resources. The breeding environment can be monitored in real time, the breeding effect can be comprehensively evaluated, the abnormal breeding area and dimension can be automatically identified, and the breeder can be guided to make accurate adjustment according to the prediction result. Such a system not only can improve the breeding efficiency and reduce the breeding risk, but also can optimize the resource allocation and promote the sustainable development of aquaculture industry. By integrating big data and artificial intelligence technology, real-time monitoring of the breeding environment, accurate prediction of the breeding effect, and automatic identification and adjustment of the abnormal breeding state are realized. BRIEF DESCRIPTION OF DRAWINGS

[0063] Figure 1 It is a schematic diagram of an intelligent management method for aquaculture based on big data and artificial intelligence;

[0064] Figure 2 It is a schematic diagram of an aquaculture monitoring system. DETAILED DESCRIPTION

[0065] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application.

[0066] In one embodiment of the present application, the present application provides an intelligent management system for aquaculture based on big data and artificial intelligence, the management system comprises:

[0067] The system area prediction module is used for acquiring a plurality of preset area information of rivers and streams, generating an aquaculture monitoring system, performing regional aquaculture monitoring, and then performing comprehensive aquaculture monitoring, training an aquaculture effect prediction model, and outputting aquaculture effect prediction data.

[0068] The abnormal area analysis module is used for calculating the area aquaculture coefficient of the preset area, and then acquiring the area standard deviation coefficient of the preset area, marking the area discrete state of the preset area according to the area standard deviation coefficient, and then acquiring the abnormal aquaculture area information.

[0069] The abnormal dimension analysis module is used for calculating the dimension aquaculture coefficient of the dimension of the preset area, and then acquiring the dimension standard deviation coefficient of the corresponding dimension, marking the dimension discrete state of the corresponding dimension according to the dimension standard deviation coefficient, and then acquiring the abnormal aquaculture dimension information.

[0070] The area dimension adjustment module is used for adjusting the abnormal aquaculture dimension of the abnormal aquaculture area by a preset dimension adjustment parameter, predicting the effect by the aquaculture effect prediction model, and then determining whether to continue adjusting.

[0071] The working principle of the above technical solution is that the key aquaculture environmental parameters such as water quality, temperature, dissolved oxygen, and pH value of multiple preset areas in rivers and streams are acquired by a sensor acquisition method to generate an aquaculture monitoring system. The system can monitor the aquaculture environmental state of these preset areas in real time or periodically. The data collected by the monitoring system is input into an aquaculture effect prediction model. This model is trained based on big data and artificial intelligence methods and can predict the aquaculture effect (such as growth rate, survival rate, etc.) under different aquaculture conditions. The system uses these prediction data to evaluate the current aquaculture conditions and provides a preliminary prediction of the aquaculture effect. The regional aquaculture coefficient of each preset area is calculated. The regional standard deviation coefficient is calculated to measure the dispersion degree (i.e., fluctuation size) of the aquaculture environment in that area. According to the regional standard deviation coefficient, the system labels the preset areas with discrete states such as "stable", "fluctuating", or "abnormal" and identifies abnormal aquaculture areas. Further analysis of the data in each dimension (such as water quality, temperature, etc.) in each preset area is performed to calculate the dimension aquaculture coefficient and the dimension standard deviation coefficient. These dimensions are labeled with discrete states and abnormal aquaculture is identified. For the identified abnormal aquaculture areas and dimensions, the system adjusts according to the preset dimension adjustment parameters (such as increasing or decreasing feed, adjusting water quality, etc.). After adjustment, the system again uses the aquaculture effect prediction model to predict the effect to evaluate the effectiveness of the adjustment. According to the prediction results, the system decides whether to continue adjusting or adjust the adjustment strategy.

[0072] The technical effect of the above technical solution is that by monitoring and predicting the aquaculture effect in real time, the system can timely discover and solve problems in the aquaculture process, thereby improving the aquaculture efficiency and yield, identifying and warning abnormal aquaculture areas and dimensions, helping the aquaculturist to take timely measures to reduce the risk of environmental mutation or improper management. Through precise data analysis and prediction, the aquaculturist can be guided to reasonably allocate feed, water resources, and other production materials to optimize resource allocation. The aquaculture environment can be monitored in real time, the aquaculture effect can be comprehensively evaluated, abnormal aquaculture areas and dimensions can be automatically identified, and the aquaculturist can be guided to make precise adjustments according to the prediction results. Such a system not only improves the aquaculture efficiency and reduces the aquaculture risk, but also optimizes resource allocation and promotes the sustainable development of the aquaculture industry. By integrating big data and artificial intelligence technologies, real-time monitoring of the aquaculture environment, precise prediction of the aquaculture effect, and automatic identification and adjustment of abnormal aquaculture states are achieved.

[0073] In an embodiment of the present application, the system area prediction module comprises:

[0074] The regional integration module is configured to acquire the information of multiple preset areas of the river or stream, combine the preset area information, and obtain regional integration information.

[0075] a system area monitoring module configured to generate an aquaculture monitoring system based on the area integrated information, and monitor each preset area based on the aquaculture monitoring system to obtain area cultivation monitoring data;

[0076] combining the area cultivation monitoring data to generate comprehensive cultivation monitoring data;

[0077] a model effect prediction module configured to obtain historical comprehensive cultivation monitoring data, and train a cultivation effect prediction model;

[0078] predict the cultivation effect of real-time comprehensive cultivation monitoring data by the cultivation effect prediction model to obtain cultivation effect prediction data.

[0079] The working principle of the above technical solution is as follows: basic information of multiple preset areas in a river or stream, such as geographical location, water quality characteristics, and cultivation varieties, is obtained by means of big data and the like, and is integrated together to form area integrated information. Based on the area integrated information, a customized aquaculture monitoring system is generated. This system can set corresponding monitoring parameters and thresholds for each preset area. Through sensor networks, remote monitoring devices, and other means, the cultivation environment data (such as water temperature, dissolved oxygen, pH value, and feed feeding amount) and cultivation biological data (such as growth rate and survival rate) of each preset area are collected in real time or periodically. The area cultivation monitoring data is obtained by combining and comprehensively analyzing the area cultivation monitoring data of each preset area. The comprehensive cultivation monitoring data reflects the overall situation of the entire cultivation area, including the stability of the cultivation environment and the growth of the cultivation organisms. Historical comprehensive cultivation monitoring data is obtained, which includes cultivation environment data, cultivation biological data, and corresponding cultivation effect data (such as final yield and quality) in the past period of time. Using these historical data, a cultivation effect prediction model is trained. This model can predict future cultivation effects based on current cultivation environment data and cultivation biological data. When new comprehensive cultivation monitoring data is generated, the system predicts it through the cultivation effect prediction model to obtain cultivation effect prediction data.

[0080] The technical effect of the above technical solution is that by monitoring and comprehensively analyzing the cultivation data of each preset area in real time, the system can timely discover abnormal situations in the cultivation process. Based on the cultivation effect prediction model, the system can predict the cultivation effects under different cultivation strategies to improve cultivation efficiency and yield. The system can early warn potential risks in the cultivation environment, such as water quality deterioration and disease outbreak, to reduce cultivation losses. Through precise control of the cultivation environment and optimization of the cultivation strategy, the quality and safety of the cultivation products are improved.

[0081] In an embodiment of the present application, the abnormal area analysis module comprises:

[0082] The regional aquaculture calculation module is configured to calculate a regional aquaculture coefficient of each preset region according to regional aquaculture monitoring data of the preset region.

[0083] The calculation formula of the regional aquaculture coefficient is:

[0084]

[0085] wherein QY is the regional aquaculture coefficient, j is the number of types of aquaculture products in the preset region, Y si is the actual yield of the i-th type of aquaculture product, Y avgi is the average yield of all preset regions of the i-th type of aquaculture product, Q i is the preset weight of the i-th type of aquaculture product, m is the number of types of resource inputs in the preset region, I sn is the actual amount of the n-th type of resource input, I avgn is the average amount of all preset regions of the n-th type of resource input, C n is the preset weight of the n-th type of resource input, k is the number of types of environmental impacts in the preset region, E sx is the actual degree data (actual average value) of the x-th type of environmental impact, E avgx is the average degree data of all preset regions of the x-th type of environmental impact, F x is the preset weight of the x-th type of environmental impact.

[0086] The regional standard deviation calculation module is configured to calculate a regional standard deviation coefficient of each regional aquaculture coefficient according to the regional aquaculture coefficient.

[0087] The calculation formula of the regional standard deviation coefficient of the regional aquaculture coefficient is:

[0088]

[0089] wherein α yz is the regional standard deviation coefficient of the regional aquaculture coefficient, x is the regional aquaculture coefficient, u is the average value of all regional aquaculture coefficients, and N is the total number of regional aquaculture coefficients.

[0090] The abnormal region acquisition module is configured to compare the regional standard deviation coefficient with a preset regional standard coefficient threshold to obtain a regional dispersion comparison result of the preset region.

[0091] According to the regional dispersion comparison result, the preset region corresponding to the regional standard deviation coefficient is marked with a regional dispersion state, and regional dispersion state marking information is obtained.

[0092] Abnormal aquaculture region information is obtained according to the regional dispersion state marking information.

[0093] The working principle of the above technical solution is: according to the regional aquaculture monitoring data of each preset region, the regional aquaculture coefficient of the corresponding preset region is calculated, which considers the comparison of the actual yield and the average yield of the aquaculture products, the comparison of the actual amount and the average amount of resource input, and the comparison of the actual degree and the average degree of environmental impact. According to the number of types of aquaculture products, the actual yield and the average yield of each type of aquaculture product, the preset weight of each type of aquaculture product, the number of types of resource input, the actual amount and the average amount of each type of resource input, the preset weight of each type of resource input, the number of types of environmental impact, the actual degree data and the average degree data of each type of environmental impact, and the preset weight of each type of environmental impact, the calculation is carried out. Calculate the regional standard deviation coefficient of each regional aquaculture coefficient, which reflects the dispersion degree of the regional aquaculture coefficient, that is, the difference degree of the aquaculture status of each preset region and the overall average level, considering the average value (u) and the total number of all regional aquaculture coefficients, and calculating by the standard deviation formula. Compare the calculated regional standard deviation coefficient with the preset regional standard coefficient threshold, and mark the preset region corresponding to the regional standard deviation coefficient according to the comparison result. The marking result includes states such as "stable", "fluctuation", "abnormal" and the like, which are used to indicate whether the aquaculture status of the region deviates from the normal level. According to the regional discrete state marking information, the preset region marked as "abnormal" is selected as the abnormal aquaculture region information output.

[0094] The technical effect of the above technical solution is: by calculating the regional aquaculture coefficient and the regional standard deviation coefficient, the system can more finely evaluate the aquaculture status of each preset region. The system automatically identifies and marks the abnormal aquaculture region, and locks, pays attention to and continuously monitors the abnormal aquaculture region, obtains the problems in the aquaculture process such as disease outbreak and water quality deterioration, so as to take effective measures for intervention. By analyzing the regional aquaculture coefficient and the regional standard deviation coefficient, the abnormal region is obtained, and the aquaculture risk is reduced: the risk in the aquaculture process is reduced, and the quality and safety of the aquaculture products are improved.

[0095] In an embodiment of the present application, the abnormal dimension analysis module comprises:

[0096] The dimension aquaculture calculation module is used for acquiring a preset aquaculture monitoring dimension, and performing multi-dimensional aquaculture monitoring data acquisition on the preset region according to the preset aquaculture monitoring dimension to obtain dimension aquaculture monitoring data of multiple dimensions of each preset region.

[0097] The dimension aquaculture coefficient of each dimension of each preset region is calculated according to the dimension aquaculture monitoring data of multiple dimensions of each preset region.

[0098] The calculation formula of the dimension aquaculture coefficient of each dimension of each preset region is:

[0099]

[0100] wherein, WY is a dimension breeding coefficient, Y sio is the actual yield of the i-th kind of breeding product of the o-th dimension in the preset area, Y avgio is the average yield of all preset areas of the i-th kind of breeding product of the o-th dimension, Q io is the preset weight of the i-th kind of breeding product of the o-th dimension in the preset area, I sno is the actual amount of resource input of the n-th kind of the o-th dimension in the preset area, I avgnb is the average amount of all preset areas of the n-th kind of resource input of the o-th dimension, C nb is the preset weight of the n-th kind of resource input of the o-th dimension in the preset area, E sx is the actual degree data (actual average value) of environmental impact of the x-th kind of the o-th dimension in the preset area, E avgx is the average degree data of all preset areas of the x-th kind of environmental impact of the o-th dimension, F x is the preset weight of the x-th kind of environmental impact of the o-th dimension in the preset area;

[0101] a dimension standard deviation calculation module, configured to calculate a dimension standard deviation coefficient of each dimension breeding coefficient of each preset area according to the dimension breeding coefficient;

[0102] The calculation formula of the dimension standard deviation coefficient of each dimension of each preset area is:

[0103]

[0104] wherein, β wd is the dimension standard deviation coefficient of each dimension of each preset area, y is the dimension breeding coefficient, a is the average value of all dimension breeding coefficients, and M is the total number of dimension breeding coefficients of the preset area;

[0105] an abnormal dimension acquisition module, configured to compare the dimension standard deviation coefficient with a preset dimension standard coefficient threshold to obtain a dimension dispersion comparison result of the preset area;

[0106] According to the dimension dispersion comparison result, the preset area corresponding to the dimension standard deviation coefficient is marked in a dimension dispersion state to obtain dimension dispersion state marking information;

[0107] According to the dimension dispersion state marking information, abnormal breeding dimension information is obtained.

[0108] The working principle of the above technical solution is: according to the breeding demand and the monitoring target, the preset breeding monitoring dimension is obtained, which includes equipment, feed, temperature, biological health and the like. For each preset area, multi-dimensional breeding monitoring data collection is carried out according to the preset monitoring dimension. These data include the actual yield of breeding products, the actual amount of resource input, the actual degree of environmental impact data and the like. According to the multi-dimensional dimension breeding monitoring data of each preset area, the dimension breeding coefficient of each dimension of each preset area is calculated. This coefficient is a comprehensive index reflecting the breeding condition of the area in a specific dimension. The formula will consider the comparison of actual yield and average yield, the comparison of actual amount of resource input and average amount, and the comparison of actual degree of environmental impact and average degree, and also consider the preset weight of each factor. The dimension standard deviation coefficient of each dimension breeding coefficient of each preset area is calculated. This coefficient reflects the difference degree of the breeding condition of the area in a specific dimension and the overall average level. The average value and the total number of all dimension breeding coefficients are considered, and the calculation is carried out through the standard deviation formula. The dimension standard deviation coefficient calculated is compared with the preset dimension standard coefficient threshold, and the preset area corresponding to the dimension standard deviation coefficient is marked according to the dimension dispersion state according to the comparison result. The marking result includes states such as "stable", "fluctuation" and "abnormal", which are used to indicate whether the breeding condition of the area in a specific dimension deviates from the normal level. According to the dimension dispersion state marking information, the preset area and the corresponding dimension marked as "abnormal" are selected as the abnormal breeding dimension information output. These information can help the breeder to find and handle the problems in the breeding process in time.

[0109] The technical effect of the above technical solution is: through multi-dimensional breeding monitoring data collection and dimension breeding coefficient calculation, the breeding condition of each preset area in a specific dimension can be more finely evaluated. Abnormal breeding dimensions such as insufficient resource input in a specific dimension and excessive environmental impact can be automatically identified and marked, so that effective measures can be taken for intervention. By analyzing the dimension breeding coefficient and the dimension standard deviation coefficient, the system can provide suggestions for optimizing resource allocation for the breeder, such as adjusting the resource input amount in a specific dimension and improving the breeding environment, so as to improve the breeding efficiency and yield. Reduce the risk in the breeding process, improve the quality and safety of the breeding products, and thus increase the income of the breeder. The implementation of the method promotes the intelligent and data-oriented development of aquaculture, improves the overall scientific and technological level and competitiveness of the breeding industry. At the same time, multi-dimensional monitoring and analysis also provide more comprehensive and in-depth breeding information for the breeder.

[0110] In an embodiment of the present application, the area dimension adjustment module comprises:

[0111] The adjusting parameter acquisition module is configured to acquire abnormal aquaculture dimension information of the abnormal aquaculture area, and acquire preset dimension adjusting parameters according to the abnormal aquaculture dimension information of the preset aquaculture area.

[0112] The adjusting effect prediction module is configured to input the preset dimension parameters into an aquaculture effect prediction model to obtain predicted aquaculture effect data corresponding to the abnormal aquaculture dimension of the abnormal aquaculture area.

[0113] The adjusting effect prediction module is configured to input the preset dimension parameters into an aquaculture effect prediction model to obtain predicted aquaculture effect data corresponding to the abnormal aquaculture dimension of the abnormal aquaculture area.

[0114] The adjusting parameter replacement module is configured to trigger replacement of the preset dimension adjusting parameters when the predicted aquaculture effect data is less than the preset effect threshold, until the predicted aquaculture effect data is greater than or equal to the preset effect threshold, and stop replacement of the preset dimension adjusting parameters.

[0115] The working principle of the above technical solution is as follows: by monitoring and analyzing multi-dimensional data (such as water quality, environmental parameters, biological health, etc.) of the aquaculture area, the abnormal aquaculture area and its corresponding abnormal aquaculture dimension are identified. The abnormal dimension is a key factor leading to poor aquaculture effect or deterioration of biological health condition. For the identified abnormal aquaculture dimension, the system acquires corresponding adjusting parameters from a preset adjusting parameter library. Including the usage amount of water quality adjusting agent, the adjustment range of environmental parameters, the adjustment of feed formula, etc., aiming to improve the abnormal aquaculture condition. The preset dimension adjusting parameters are input into an aquaculture effect prediction model. The model can predict the possible effect of the aquaculture area under different adjusting parameters based on historical data and machine learning algorithm. The model outputs the predicted aquaculture effect data corresponding to the abnormal aquaculture dimension of the abnormal aquaculture area, such as growth rate, survival rate, yield, etc. The predicted aquaculture effect data is compared with the preset effect threshold. If the predicted effect is lower than the threshold, it means that the current adjusting parameter may not be sufficient to improve the abnormal aquaculture condition, and the system will trigger replacement of the preset dimension adjusting parameters. This process may involve multiple iterations, each iteration will adjust the adjusting parameters and input the model for prediction until the predicted aquaculture effect data reaches or exceeds the preset effect threshold. When the predicted aquaculture effect data is greater than or equal to the preset effect threshold, the system stops replacement of the preset dimension adjusting parameters, and outputs the final adjusting parameters and predicted aquaculture effect data.

[0116] The technical effects of the above technical solutions are: the system can accurately identify abnormal breeding areas and their corresponding abnormal breeding dimensions, and adjust them according to preset dimension adjustment parameters. The breeding conditions can be improved quickly, and the breeding efficiency and yield can be improved. Through the application of the breeding effect prediction model, the system can provide intelligent decision support for breeders. Breeders can adjust the adjustment parameters according to the prediction results. The prediction results can be dynamically adjusted to optimize the allocation of breeding resources. Waste can be reduced, resource utilization efficiency can be improved, and breeding costs can be reduced. Through accurate identification and intelligent decision support, the system can significantly improve the breeding efficiency. Breeders can more efficiently use breeding resources, improve the health of the organisms and the growth rate, and thus obtain higher yields and better economic benefits. The method promotes the development of the breeding industry towards intelligent and data-based direction. Through real-time monitoring and analysis of breeding data.

[0117] In one embodiment of the present application, the management method comprises:

[0118] S1, obtain a plurality of preset region information of rivers and streams, generate an aquaculture monitoring system, monitor regional breeding, and then monitor comprehensive breeding, train a breeding effect prediction model, and output breeding effect prediction data;

[0119] S2, calculate the regional breeding coefficient of the preset region, and then obtain the regional standard deviation coefficient of the preset region, label the regional discrete state of the preset region according to the regional standard deviation coefficient, and then obtain abnormal breeding region information;

[0120] S3, calculate the dimension breeding coefficient of the dimension of the preset region, and then obtain the dimension standard deviation coefficient of the corresponding dimension, label the dimension discrete state of the corresponding dimension according to the dimension standard deviation coefficient, and then obtain abnormal breeding dimension information;

[0121] S4, adjust the abnormal breeding dimension of the abnormal breeding region with the preset dimension adjustment parameter, predict the effect through the breeding effect prediction model, and then determine whether to continue adjusting.

[0122] The working principle of the above technical solution is that the key aquaculture environmental parameters such as water quality, temperature, dissolved oxygen, and pH value of multiple preset areas in rivers and streams are acquired by a sensor acquisition method to generate an aquaculture monitoring system. The system can monitor the aquaculture environmental state of these preset areas in real time or periodically. The data collected by the monitoring system is input into an aquaculture effect prediction model. This model is trained based on big data and artificial intelligence methods and can predict the aquaculture effect (such as growth rate, survival rate, etc.) under different aquaculture conditions. The system uses these prediction data to evaluate the current aquaculture conditions and provides a preliminary prediction of the aquaculture effect. The regional aquaculture coefficient of each preset area is calculated. The regional standard deviation coefficient is calculated to measure the dispersion degree (i.e., fluctuation size) of the aquaculture environment in that area. According to the regional standard deviation coefficient, the system labels the preset areas with discrete states such as "stable", "fluctuating", or "abnormal" and identifies abnormal aquaculture areas. Further analysis of the data in each dimension (such as water quality, temperature, etc.) in each preset area is performed to calculate the dimension aquaculture coefficient and the dimension standard deviation coefficient. These dimensions are labeled with discrete states, and abnormal aquaculture is identified. For the identified abnormal aquaculture areas and dimensions, the system adjusts according to the preset dimension adjustment parameters (such as increasing or decreasing feed, adjusting water quality, etc.). After adjustment, the system uses the aquaculture effect prediction model again to predict the effect to evaluate the effectiveness of the adjustment. According to the prediction results, the system decides whether to continue adjusting or adjust the adjustment strategy.

[0123] The technical effect of the above technical solution is that by monitoring and predicting the aquaculture effect in real time, the system can timely discover and solve problems in the aquaculture process, thereby improving the aquaculture efficiency and yield, identifying and warning abnormal aquaculture areas and dimensions, helping the aquaculturist to take timely measures to reduce the risk of environmental mutation or improper management. Through precise data analysis and prediction, the aquaculturist can be guided to reasonably allocate feed, water resources, and other production materials to optimize resource allocation. The aquaculture environment can be monitored in real time, the aquaculture effect can be comprehensively evaluated, abnormal aquaculture areas and dimensions can be automatically identified, and the aquaculturist can be guided to make precise adjustments according to the prediction results. Such a system not only improves the aquaculture efficiency and reduces the aquaculture risk, but also optimizes resource allocation and promotes the sustainable development of the aquaculture industry. By integrating big data and artificial intelligence technologies, real-time monitoring of the aquaculture environment, precise prediction of the aquaculture effect, and automatic identification and adjustment of abnormal aquaculture states are achieved.

[0124] In one embodiment of the present application, the S1 comprises:

[0125] Obtain information of multiple preset areas of a river or stream, combine the preset area information to obtain integrated area information;

[0126] According to the region integration information, a water product cultivation monitoring system is generated, water product cultivation in each preset region is monitored according to the water product cultivation monitoring system, and region cultivation monitoring data is obtained;

[0127] The region cultivation monitoring data is combined to generate comprehensive cultivation monitoring data;

[0128] Historical comprehensive cultivation monitoring data is obtained, and a cultivation effect prediction model is trained;

[0129] The cultivation effect prediction model is used to predict the cultivation effect of real-time cultivation comprehensive cultivation monitoring data, and cultivation effect prediction data is obtained.

[0130] The working principle of the above technical solution is as follows: through big data and other means, the basic information of a plurality of preset regions in a river or stream is obtained, such as geographical position, water quality characteristics, cultivation varieties, etc., which are integrated together to form region integration information. Based on the region integration information, a customized water product cultivation monitoring system is generated. This system can set corresponding monitoring parameters and thresholds for each preset region. Through sensor networks, remote monitoring devices and other means, the cultivation environment data (such as water temperature, dissolved oxygen, pH value, feed feeding amount, etc.) and cultivation biological data (such as growth rate, survival rate, etc.) of each preset region are collected in real time or periodically. The region cultivation monitoring data is obtained by combining and comprehensively analyzing the region cultivation monitoring data of each preset region. The comprehensive cultivation monitoring data reflects the overall situation of the entire cultivation region, including the stability of the cultivation environment and the growth of the cultivation organisms. Historical comprehensive cultivation monitoring data is obtained, which includes cultivation environment data, cultivation biological data and corresponding cultivation effect data (such as final yield, quality, etc.) in the past period of time. Using these historical data, a cultivation effect prediction model is trained. This model can predict the future cultivation effect based on the current cultivation environment data and cultivation biological data. When new comprehensive cultivation monitoring data is generated, the system predicts it through the cultivation effect prediction model to obtain cultivation effect prediction data.

[0131] The technical effect of the above technical solution is that by monitoring and comprehensively analyzing the cultivation data of each preset region in real time, the system can timely discover abnormal situations in the cultivation process. Based on the cultivation effect prediction model, the system can predict the cultivation effect under different cultivation strategies to improve the cultivation efficiency and yield. The system can early warn potential risks in the cultivation environment, such as water quality deterioration and disease outbreak, to reduce cultivation losses. Through precise control of the cultivation environment and optimization of the cultivation strategy, the quality and safety of the cultivation products are improved.

[0132] In an embodiment of the present application, S2 comprises:

[0133] Calculate the regional aquaculture coefficient of the preset region according to the regional aquaculture monitoring data of each preset region;

[0134] The calculation formula of the regional aquaculture coefficient is:

[0135]

[0136] Wherein, QY is the regional aquaculture coefficient, j is the number of types of aquaculture products in the preset region, Y si is the actual yield of the i-th type of aquaculture product, Y avgi is the average yield of all preset regions of the i-th type of aquaculture product, Q i is the preset weight of the i-th type of aquaculture product, m is the number of types of resource inputs in the preset region, I sn is the actual amount of the n-th type of resource input, I avgn is the average amount of all preset regions of the n-th type of resource input, C n is the preset weight of the n-th type of resource input, k is the number of types of environmental impact in the preset region, E sx is the actual degree data (actual average value) of the x-th type of environmental impact, E avgx is the average degree data of all preset regions of the x-th type of environmental impact, F x is the preset weight of the x-th type of environmental impact;

[0137] Calculate the regional standard deviation coefficient of each regional aquaculture coefficient according to the regional aquaculture coefficient;

[0138] The calculation formula of the regional standard deviation coefficient of the regional aquaculture coefficient is:

[0139]

[0140] Wherein, α yz is the regional standard deviation coefficient of the regional aquaculture coefficient, x is the regional aquaculture coefficient, u is the average value of all regional aquaculture coefficients, and N is the total number of regional aquaculture coefficients;

[0141] Compare the regional standard deviation coefficient with the preset regional standard coefficient threshold to obtain the regional dispersion comparison result of the preset region;

[0142] According to the regional dispersion comparison result, the regional dispersion state of the preset region corresponding to the regional standard deviation coefficient is marked to obtain regional dispersion state marking information;

[0143] According to the regional dispersion state marking information, obtain abnormal aquaculture region information.

[0144] The working principle of the above technical solution is that according to the regional aquaculture monitoring data of each preset region, the regional aquaculture coefficient of the corresponding preset region is calculated, which considers the comparison of the actual yield and the average yield of the aquaculture products, the comparison of the actual amount and the average amount of resource input, and the comparison of the actual degree and the average degree of environmental impact. According to the number of types of aquaculture products, the actual yield and the average yield of each type of aquaculture product, the preset weight of each type of aquaculture product, the number of types of resource input, the actual amount and the average amount of each type of resource input, the preset weight of each type of resource input, the number of types of environmental impact, the actual degree data and the average degree data of each type of environmental impact, and the preset weight of each type of environmental impact, the calculation is carried out. Calculate the regional standard deviation coefficient of each regional aquaculture coefficient, which reflects the dispersion degree of the regional aquaculture coefficient, that is, the difference degree of the aquaculture status of each preset region and the overall average level, considering the average value (u) and the total number of all regional aquaculture coefficients, and calculating by the standard deviation formula. The calculated regional standard deviation coefficient is compared with the preset regional standard coefficient threshold, and the preset region corresponding to the regional standard deviation coefficient is marked according to the comparison result. The marking result includes states such as "stable", "fluctuation", "abnormal" and the like, which are used to indicate whether the aquaculture status of the region deviates from the normal level. According to the regional discrete state marking information, the preset region marked as "abnormal" is selected as the abnormal aquaculture region information output.

[0145] The technical effect of the above technical solution is that by calculating the regional aquaculture coefficient and the regional standard deviation coefficient, the system can more finely evaluate the aquaculture status of each preset region. The system automatically identifies and marks the abnormal aquaculture region, and locks, pays attention to and continuously monitors the abnormal aquaculture region, obtains the problems in the aquaculture process such as disease outbreak and water quality deterioration, so as to take effective measures for intervention. By analyzing the regional aquaculture coefficient and the regional standard deviation coefficient, the abnormal region is obtained, and the aquaculture risk is reduced: the risk in the aquaculture process is reduced, and the quality and safety of the aquaculture products are improved.

[0146] In an embodiment of the present application, the S3 comprises:

[0147] A preset aquaculture monitoring dimension is obtained, and multi-dimensional aquaculture monitoring data acquisition is performed on the preset region according to the preset aquaculture monitoring dimension, so as to obtain dimension aquaculture monitoring data of multiple dimensions of each preset region;

[0148] According to the dimension aquaculture monitoring data of multiple dimensions of each preset region, the dimension aquaculture coefficient of each dimension of each preset region is calculated;

[0149] The calculation formula of the dimension aquaculture coefficient of each dimension of each preset region is:

[0150]

[0151] wherein WY is a dimension breeding coefficient, Y sio is the actual yield of the i-th kind of breeding product of the o-th dimension in the preset area, Y avgio is the average yield of all preset areas of the i-th kind of breeding product of the o-th dimension, Q io is the preset weight of the i-th kind of breeding product of the o-th dimension in the preset area, I sno is the actual amount of resource input of the n-th kind of the o-th dimension in the preset area, I avgnb is the average amount of all preset areas of the n-th kind of resource input of the o-th dimension, C nb is the preset weight of the n-th kind of resource input of the o-th dimension in the preset area, E sx is the actual degree data (actual average value) of the x-th kind of environmental impact of the o-th dimension in the preset area, E avgx is the average degree data of all preset areas of the x-th kind of environmental impact of the o-th dimension, F x is the preset weight of the x-th kind of environmental impact of the o-th dimension in the preset area;

[0152] According to the dimension breeding coefficient, a dimension standard deviation coefficient of each dimension breeding coefficient of each preset area is calculated;

[0153] The calculation formula of the dimension standard deviation coefficient of each dimension of each preset area is:

[0154]

[0155] wherein β wd is the dimension standard deviation coefficient of each dimension of each preset area, y is the dimension breeding coefficient, a is the average value of all dimension breeding coefficients, and M is the total number of dimension breeding coefficients of the preset area;

[0156] The dimension standard deviation coefficient is compared with a preset dimension standard coefficient threshold to obtain a dimension dispersion comparison result of the preset area;

[0157] According to the dimension dispersion comparison result, the preset area corresponding to the dimension standard deviation coefficient is marked in a dimension dispersion state to obtain dimension dispersion state marking information;

[0158] According to the dimension dispersion state marking information, abnormal breeding dimension information is obtained.

[0159] The working principle of the above technical solution is as follows: according to the breeding demand and the monitoring target, the preset breeding monitoring dimension is obtained, which includes equipment, feed, temperature, biological health and the like. For each preset area, multi-dimensional breeding monitoring data collection is carried out according to the preset monitoring dimension. These data include the actual yield of breeding products, the actual amount of resource input, the actual degree of environmental impact data and the like. According to the multi-dimensional dimension breeding monitoring data of each preset area, the dimension breeding coefficient of each dimension of each preset area is calculated. This coefficient is a comprehensive index reflecting the breeding condition of the area in a specific dimension. The formula will consider the comparison of the actual yield and the average yield, the comparison of the actual amount of resource input and the average amount, and the comparison of the actual degree of environmental impact and the average degree, and will also consider the preset weight of each factor. The dimension standard deviation coefficient of each dimension breeding coefficient of each preset area is calculated. This coefficient reflects the difference degree of the breeding condition of the area in a specific dimension and the overall average level. The average value and the total number of all dimension breeding coefficients are considered, and the calculation is carried out through the standard deviation formula. The dimension standard deviation coefficient calculated is compared with the preset dimension standard coefficient threshold, and the preset area corresponding to the dimension standard deviation coefficient is marked according to the dimension dispersion state according to the comparison result. The marking result includes states such as "stable", "fluctuation" and "abnormal", which are used to indicate whether the breeding condition of the area in a specific dimension deviates from the normal level. According to the dimension dispersion state marking information, the preset area marked as "abnormal" and the corresponding dimension are selected as the abnormal breeding dimension information output. These information can help the breeder to find and handle the problems in the breeding process in time.

[0160] The technical effect of the above technical solution is as follows: through multi-dimensional breeding monitoring data collection and dimension breeding coefficient calculation, the breeding condition of each preset area in a specific dimension can be evaluated more finely. Abnormal breeding dimensions such as insufficient resource input in a specific dimension and excessive environmental impact can be automatically identified and marked, so that effective measures can be taken for intervention. By analyzing the dimension breeding coefficient and the dimension standard deviation coefficient, the system can provide suggestions for optimizing resource allocation for the breeder, such as adjusting the resource input amount in a specific dimension and improving the breeding environment, so as to improve the breeding efficiency and yield. The risk in the breeding process is reduced, and the quality and safety of the breeding products are improved, so as to increase the income of the breeder. The implementation of the method promotes the intelligent and data-oriented development of aquaculture, improves the overall scientific and technological level and competitiveness of the breeding industry. At the same time, multi-dimensional monitoring and analysis provide more comprehensive and in-depth breeding information for the breeder.

[0161] In an embodiment of the present application, the S4 comprises:

[0162] The abnormal breeding dimension information of the abnormal breeding area is obtained, and the preset dimension adjustment parameter is obtained according to the abnormal breeding dimension information of the preset breeding area.

[0163] inputting the preset dimension parameter into a breeding effect prediction model to obtain prediction breeding effect data of the corresponding abnormal breeding dimension of the abnormal breeding area;

[0164] comparing the prediction breeding effect data with a preset effect threshold to obtain an effect comparison result;

[0165] when the prediction breeding effect data is less than the preset effect threshold, triggering replacement of the preset dimension adjustment parameter until the prediction breeding effect data is greater than or equal to the preset effect threshold, and stopping replacement of the preset dimension adjustment parameter.

[0166] The working principle of the above technical solution is as follows: by monitoring and analyzing multi-dimensional data (such as water quality, environmental parameters, biological health, etc.) of the breeding area, the abnormal breeding area and its corresponding abnormal breeding dimension are identified. Abnormal dimension is the key factor leading to poor breeding effect or deterioration of biological health condition. For the identified abnormal breeding dimension, the system obtains the corresponding adjustment parameter from the preset adjustment parameter library. Including the usage amount of water quality adjusting agent, the adjustment range of environmental parameters, the adjustment of feed formula, etc., aiming to improve the abnormal breeding condition. The preset dimension adjustment parameter is input into the breeding effect prediction model. The model can predict the possible effect of the breeding area under different adjustment parameters based on historical data and machine learning algorithm. The model outputs the prediction breeding effect data of the corresponding abnormal breeding dimension of the abnormal breeding area, such as growth rate, survival rate, yield, etc. The prediction breeding effect data is compared with the preset effect threshold. If the prediction effect is lower than the threshold, it means that the current adjustment parameter may not be enough to improve the abnormal breeding condition, and the system will trigger the replacement of the preset dimension adjustment parameter. This process may involve multiple iterations, each iteration will adjust the adjustment parameter and input the model for prediction until the prediction breeding effect data reaches or exceeds the preset effect threshold. When the prediction breeding effect data is greater than or equal to the preset effect threshold, the system stops the replacement of the preset dimension adjustment parameter, and outputs the final adjustment parameter and prediction breeding effect data.

[0167] The technical effects of the above technical solutions are: the system can accurately identify abnormal breeding areas and their corresponding abnormal breeding dimensions, and adjust them according to the preset dimension adjustment parameters. It can quickly improve the breeding conditions, improve the breeding efficiency and yield. Through the application of the breeding effect prediction model, the system can provide intelligent decision support for breeders. Breeders can adjust the adjustment parameters according to the prediction results. It can dynamically adjust the adjustment parameters according to the prediction results, optimize the allocation of breeding resources. It can reduce waste, improve resource utilization efficiency, and reduce breeding costs. Through accurate identification and intelligent decision support, the system can significantly improve the breeding efficiency. Breeders can more efficiently use breeding resources, improve the health and growth rate of organisms, and thus achieve higher yields and better economic benefits. The method promotes the development of the breeding industry towards intelligence and data. Through real-time monitoring and analysis of breeding data.

[0168] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application also intends to include these modifications and variations.

Claims

1. An intelligent management system for aquaculture based on big data and artificial intelligence, characterized in that, The management system includes: The system's regional prediction module is used to acquire information on multiple preset regions of rivers and streams, generate an aquaculture monitoring system, conduct regional aquaculture monitoring, and then conduct comprehensive aquaculture monitoring, train an aquaculture effect prediction model, and output aquaculture effect prediction data. The abnormal area analysis module is used to calculate the regional aquaculture coefficient of a preset area, and then obtain the regional standard deviation coefficient of the preset area. Based on the regional standard deviation coefficient, the preset area is labeled with a regional discrete state, and then abnormal aquaculture area information is obtained. The abnormal dimension analysis module is used to calculate the dimension aquaculture coefficient of a preset region, and then obtain the dimension standard deviation coefficient of the corresponding dimension. Based on the dimension standard deviation coefficient, the corresponding dimension is labeled with dimension discrete state, and then abnormal aquaculture dimension information is obtained. The regional dimension adjustment module is used to adjust the preset dimension adjustment parameters for abnormal aquaculture dimensions in abnormal aquaculture areas, predict the effect through an aquaculture effect prediction model, and then determine whether to continue adjustment.

2. The intelligent aquaculture management system based on big data and artificial intelligence according to claim 1, characterized in that, The system's regional prediction module includes: The region integration module is used to acquire multiple preset region information of rivers or streams, and combine the preset region information to obtain region integration information; The system area monitoring module is used to generate an aquaculture monitoring system based on the integrated area information, and to conduct aquaculture monitoring in each preset area based on the aquaculture monitoring system to obtain regional aquaculture monitoring data. The aquaculture monitoring data of the aforementioned region are combined to generate comprehensive aquaculture monitoring data; The model performance prediction module is used to acquire historical comprehensive aquaculture monitoring data and train the aquaculture performance prediction model. The aquaculture effect prediction model is used to predict the aquaculture effect based on real-time comprehensive aquaculture monitoring data, thereby obtaining aquaculture effect prediction data.

3. The intelligent aquaculture management system based on big data and artificial intelligence according to claim 1, characterized in that, The abnormal region analysis module includes: The regional aquaculture calculation module is used to calculate the regional aquaculture coefficient of each preset region based on the regional aquaculture monitoring data of each preset region. The regional standard deviation calculation module is used to calculate the regional standard deviation coefficient of each regional aquaculture coefficient based on the regional aquaculture coefficient. The abnormal region acquisition module is used to compare the standard deviation coefficient of the region with a preset region standard coefficient threshold to obtain the regional discrete comparison result of the preset region. Based on the regional discrete comparison results, the preset region corresponding to the regional standard deviation coefficient is labeled with regional discrete state to obtain regional discrete state labeling information. Information on abnormal aquaculture areas is obtained based on the discrete state labeling information of the region.

4. The intelligent aquaculture management system based on big data and artificial intelligence according to claim 1, characterized in that, The anomaly dimension analysis module includes: The dimensional aquaculture calculation module is used to obtain preset aquaculture monitoring dimensions, and to collect multi-dimensional aquaculture monitoring data for preset areas based on the preset aquaculture monitoring dimensions, thereby obtaining multi-dimensional aquaculture monitoring data for each preset area. The dimensional aquaculture coefficient for each dimension of each preset area is calculated based on the dimensional aquaculture monitoring data of multiple dimensions for each preset area. The dimensional standard deviation calculation module is used to calculate the dimensional standard deviation coefficient of each dimensional aquaculture coefficient for each preset area based on the dimensional aquaculture coefficient. An anomaly dimension acquisition module is used to compare the standard deviation coefficient of the dimension with a preset standard coefficient threshold of the dimension to obtain the dimension discrete comparison result of the preset region. Based on the dimensional discrete comparison results, the preset region corresponding to the dimensional standard deviation coefficient is labeled with dimensional discrete state to obtain dimensional discrete state labeling information. Abnormal aquaculture dimension information is obtained based on the discrete state labeling information of the dimensions.

5. The intelligent aquaculture management system based on big data and artificial intelligence according to claim 1, characterized in that, The region dimension adjustment module includes: The adjustment parameter acquisition module is used to acquire abnormal aquaculture dimension information of abnormal aquaculture areas and acquire preset dimension adjustment parameters based on the abnormal aquaculture dimension information of the preset aquaculture areas. The adjustment effect prediction module is used to input the preset dimension parameters into the aquaculture effect prediction model to obtain the predicted aquaculture effect data of the corresponding abnormal aquaculture dimension of the corresponding abnormal aquaculture area. The predicted aquaculture effect data is compared with a preset effect threshold to obtain the effect comparison result; The parameter replacement module is used to trigger the replacement of preset dimension adjustment parameters when the predicted aquaculture effect data is less than the preset effect threshold, until the predicted aquaculture effect data is greater than or equal to the preset effect threshold, at which point the replacement of preset dimension adjustment parameters stops.

6. A management method for implementing the intelligent aquaculture management system based on big data and artificial intelligence as described in claim 1, characterized in that, The management method includes: S1. Obtain information on multiple preset areas of rivers and streams, generate an aquaculture monitoring system, conduct regional aquaculture monitoring, then conduct comprehensive aquaculture monitoring, train an aquaculture effect prediction model, and output aquaculture effect prediction data. S2. Calculate the regional aquaculture coefficient of the preset area, and then obtain the regional standard deviation coefficient of the preset area. Based on the regional standard deviation coefficient, perform regional discrete state labeling on the preset area, and then obtain abnormal aquaculture area information. S3. Calculate the dimensional aquaculture coefficient of the preset region, and then obtain the dimensional standard deviation coefficient of the corresponding dimension. Based on the dimensional standard deviation coefficient, perform dimensional discrete state labeling on the corresponding dimension, and then obtain abnormal aquaculture dimension information. S4. Adjust the preset dimension adjustment parameters for abnormal aquaculture dimensions in abnormal aquaculture areas, predict the effect through the aquaculture effect prediction model, and then determine whether to continue adjustment.

7. The management method of the intelligent aquaculture management system based on big data and artificial intelligence according to claim 6, characterized in that, S1 includes: Obtain information on multiple preset areas of a river or stream, and combine the preset area information to obtain integrated area information; An aquaculture monitoring system is generated based on the regional integrated information. Aquaculture monitoring is then conducted in each preset area based on the aquaculture monitoring system to obtain regional aquaculture monitoring data. The aquaculture monitoring data of the aforementioned region are combined to generate comprehensive aquaculture monitoring data; Acquire historical comprehensive aquaculture monitoring data and train aquaculture performance prediction model; The aquaculture effect prediction model is used to predict the aquaculture effect based on real-time comprehensive aquaculture monitoring data, thereby obtaining aquaculture effect prediction data.

8. The management method of the intelligent aquaculture management system based on big data and artificial intelligence according to claim 6, characterized in that, S2 includes: Calculate the regional aquaculture coefficient for each preset region based on the regional aquaculture monitoring data of each preset region; Calculate the regional standard deviation coefficient of the aquaculture coefficient for each region based on the aforementioned regional aquaculture coefficient; The regional standard deviation coefficient is compared with a preset regional standard coefficient threshold to obtain the regional discrete comparison result of the preset region. Based on the regional discrete comparison results, the preset region corresponding to the regional standard deviation coefficient is labeled with regional discrete state to obtain regional discrete state labeling information. Information on abnormal aquaculture areas is obtained based on the discrete state labeling information of the region.

9. The management method of the intelligent aquaculture management system based on big data and artificial intelligence according to claim 6, characterized in that, S3 includes: Obtain preset aquaculture monitoring dimensions, and collect multi-dimensional aquaculture monitoring data for preset areas based on the preset aquaculture monitoring dimensions to obtain multi-dimensional aquaculture monitoring data for each preset area; The dimensional aquaculture coefficient for each dimension of each preset area is calculated based on the dimensional aquaculture monitoring data of multiple dimensions for each preset area. Calculate the dimensional standard deviation coefficient of each dimensional aquaculture coefficient for each preset region based on the dimensional aquaculture coefficient; The dimensional standard deviation coefficient is compared with a preset dimensional standard coefficient threshold to obtain the dimensional discrepancy comparison result of the preset region. Based on the dimensional discrete comparison results, the preset region corresponding to the dimensional standard deviation coefficient is labeled with dimensional discrete state to obtain dimensional discrete state labeling information. Abnormal aquaculture dimension information is obtained based on the discrete state labeling information of the dimensions.

10. The management method of the intelligent aquaculture management system based on big data and artificial intelligence according to claim 6, characterized in that, S4 includes: Obtain abnormal aquaculture dimension information of abnormal aquaculture areas, and obtain preset dimension adjustment parameters based on the abnormal aquaculture dimension information of the preset aquaculture areas; The preset dimension parameters are input into the aquaculture effect prediction model to obtain the predicted aquaculture effect data for the corresponding abnormal aquaculture dimension in the corresponding abnormal aquaculture area; The predicted aquaculture effect data is compared with a preset effect threshold to obtain the effect comparison result; When the predicted aquaculture effect data is less than the preset effect threshold, the preset dimension adjustment parameter is changed until the predicted aquaculture effect data is greater than or equal to the preset effect threshold, at which point the change of the preset dimension adjustment parameter stops.

Images