Digital fishery comprehensive management platform

Abstract

The invention belongs to the technical field of fishery management, and provides a digital fishery integrated management platform, which comprises a data acquisition module used for acquiring target management data of a fishery; the intelligent control module is used for intelligently controlling the fishing ground by utilizing the fishing ground management equipment according to an analysis result of the target management data; and the management and maintenance module is used for displaying and maintaining the target management data and managing the intelligent control process. By integrating the data acquisition module and the intelligent control and management maintenance module, comprehensive digitization and intelligentization of fishery management can be realized, the intelligent level of fishery management is effectively improved, the labor cost is reduced, the breeding efficiency and environmental protection benefits are improved, and sustainable development of fishery is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of fishery management, and in particular to a digital fishery integrated management platform. Background Art

[0002] Traditional fishery management is highly dependent on manual experience, and there are problems such as delayed data collection, slow disease response, and low resource utilization. Although existing technologies have tried to introduce sensors and basic data platforms, there are still the following significant defects: First, the limitations of data collection and analysis, which focus more on the monitoring of a single water quality parameter and lack real-time tracking and analysis of the growth status of individual fish; for example, traditional methods cannot accurately identify the growth differences of different individuals in a school of fish, resulting in extensive operations such as feeding and drug administration, and serious waste of resources; second, insufficient intelligent control. Although the aquaculture data platform can integrate some environmental data, it has not achieved deep linkage with equipment control; for example, when the water quality is abnormal, manual intervention is required to start the oxygenation or water exchange equipment, and the response efficiency is low and prone to errors; third, environmental protection and sustainability are shortcomings. In the traditional breeding model, tailwater treatment relies on manual detection, pollutant emission control is not accurate, and the abuse of chemical agents leads to an increased environmental burden.

[0003] Therefore, it is necessary to provide a digital integrated fisheries management platform. Summary of the invention

[0004] The present invention provides a digital integrated fishery management platform, which can realize the comprehensive digitization and intelligence of fishery management by integrating three modules: data collection, intelligent control and management and maintenance, effectively improve the intelligence level of fishery management, reduce labor costs, improve breeding efficiency and environmental benefits, and promote the sustainable development of fisheries.

[0005] The present invention provides a digital fishery integrated management platform, comprising:

[0006] Data collection module, used to collect and obtain target management data of the fishery;

[0007] An intelligent control module is used to intelligently control the fishery using fishery management equipment based on the analysis results of the target management data;

[0008] The management and maintenance module is used to display maintenance target management data and manage the intelligent control process.

[0009] Further, the data acquisition module includes a water quality parameter data acquisition unit, a tailwater treatment data acquisition unit, and a fish activity data acquisition unit;

[0010] The water quality parameter data acquisition unit is used to measure the first water quality data of the fish farm based on the water quality meters configured in the feeding ponds and outer ponds of the fish farm; the first water quality data includes but is not limited to dissolved oxygen, dissolved oxygen saturation, ammonia nitrogen, nitrite, temperature, and pH value;

[0011] The tail water treatment data acquisition unit is used to measure the second water quality data of the discharged tail water based on the water quality detection probe; the second water quality data includes but is not limited to total phosphorus, total nitrogen, suspended solids, potassium permanganate, and pH value;

[0012] The fish activity data acquisition unit is used to obtain fish body activity image data by shooting with a camera configured in the feeding pond of the fish farm.

[0013] Further, the intelligent control module includes a target management data analysis unit and an intelligent control implementation unit;

[0014] The target management data analysis unit is used to perform data analysis on the first water quality data, the second water quality data, and the fish body activity image data to obtain a data analysis result;

[0015] The intelligent control implementation unit is used to perform intelligent control on the fish farm by using the fish farm management equipment according to the data analysis result.

[0016] Further, performing data analysis on the first water quality data, the second water quality data, and the fish body activity image data to obtain a data analysis result, including:

[0017] Based on the configured Internet of Things and cloud server, transmit the first water quality data, the second water quality data, and the fish body activity image data to the processor of the cloud server through the LoRa or NB-IoT protocol;

[0018] Based on the processor's data analysis of the first water quality data, the second water quality data, and the fish body activity image data, obtain a data analysis result.

[0019] Further, based on the processor's data analysis of the first water quality data, the second water quality data, and the fish body activity image data, obtain a data analysis result, including:

[0020] Based on the set first water quality data threshold library, perform data comparison on the first water quality data to obtain a first data comparison result, and generate a first water quality analysis result according to the first data comparison result;

[0021] Based on the set second water quality data threshold library, perform data comparison on the second water quality data to obtain a second data comparison result, and generate a second water quality analysis result according to the second data comparison result;

[0022] A fish feature extraction model based on transfer learning extracts the facial texture and body shape features of fish from fish body activity image data; based on a deep convolutional neural network model, according to the facial texture and body shape features of fish, fish growth trend prediction, fish disease type identification, and fish disease development trend prediction are carried out to obtain the analysis results of fish body activity images.

[0023] Furthermore, fish growth trend prediction, fish disease type identification, and fish disease development trend prediction are carried out to obtain the analysis results of fish body activity images, including:

[0024] According to the facial texture and body shape features of fish, coordinate curves of the body length and growth time of different types of fish are fitted and generated, and growth cycle diagrams of different types of fish are generated;

[0025] According to the body shape features of fish, based on a set fish disease system feature library, disease type matching is carried out to obtain the types of fish diseases;

[0026] According to the body shape features of fish in different periods, based on a set long short-term memory network model, the development trend of fish diseases is predicted to obtain the prediction results of the development trend of fish diseases;

[0027] Based on the fish body growth cycle diagram, the types of fish diseases, and the prediction results of the development trend of fish diseases, the analysis results of fish body activity images are generated.

[0028] Furthermore, according to the data analysis results, the fish farm is intelligently controlled by using fish farm management equipment, including:

[0029] Configure fish farm management equipment; the fish farm management equipment includes an automatic feeder, an aerator, and a drug delivery device;

[0030] Based on the fish farm management equipment, according to the data analysis results, the feeding frequency and feeding amount are matched according to the fish growth stage, the power of the aerator is adjusted, and the drug delivery dose is controlled by combining the types of fish diseases and the prediction results of the development trend of fish diseases.

[0031] Furthermore, the management and maintenance module includes a data display unit and a management and maintenance unit;

[0032] A data display unit for building a visualization platform that integrates the Web and mobile ends, realizing real-time display and traceability tracking of fishery data; the real-time display includes showing the three-dimensional distribution map of the fishery; the three-dimensional distribution map is built based on a 3D model and is marked with the devices and data involved in the data acquisition module and the intelligent control module; the traceability tracking includes information tracking of fish individuals and traceability tracking of fishery management data; the information tracking of fish individuals includes: based on the RFID tags or QR code identifiers set on the fish body, tracking the growth cycle and sales flow information of fish individuals; the traceability tracking of fishery management data includes: tracing the breeding record data, breeding cycle data, and warehouse update record data of the fishery.

[0033] A management and maintenance unit for implementing management and maintenance according to the target management data.

[0034] Further, implementing management and maintenance according to the target management data includes:

[0035] Controlling the warning device to give a warning, controlling the configured water pollution collection device to start a malfunction, and generating a maintenance work order to be pushed to the configured management and maintenance terminal;

[0036] Recording the device inspection log and maintenance history and uploading them in real time through the mini-program end;

[0037] Configuring an expert Q&A link interface to provide technical guidance based on the university resources or large database of the link.

[0038] Further, the management and maintenance module further includes an input management intelligent adjustment unit; the input management intelligent adjustment unit is used to analyze the clustering distribution of the fish school, combine the fish school sales data and input data, calculate the sales input ratio, and perform intelligent adjustment of the fishery input according to the comparison between the sales input ratio and the set threshold; specifically including:

[0039] Using the DBSCAN clustering analysis method, based on the fish body activity image data, performing density clustering processing on the fish school distribution in the set water area space of the fishery to identify multiple clustering clusters of several types of fish; among them, the clustering parameters of the DBSCAN clustering analysis method are: the clustering radius is between 0.45N and 0.85N, where N is the larger value of the average body length and average body width of the fish body of the identified fish; the minimum number of samples is set to 1;

[0040] Among the multiple clustering clusters, screening out the clustering clusters greater than the preset density threshold as the target clustering clusters;

[0041] Estimating the number of fish bodies in the target clustering cluster and determining the average growth cycle of the fish school in the target clustering cluster;

[0042] Obtaining the input data of automatic feeding and drug administration according to the average growth cycle.

[0043] Obtain the equipment procurement cost data, labor input cost data, R & D cost data, and operation and maintenance cost data;

[0044] Based on the historical sales data, predict the sales data of the number of fish bodies in the target clustering cluster;

[0045] Combine the input data of automatic feeding and drug administration, equipment procurement cost data, labor input cost data, R & D cost data, operation and maintenance cost data, and sales data to calculate the sales input ratio; the calculation formula is as follows:

[0046]

[0047] In the above formula, E represents the sales input ratio, Q i represents the sales volume of the i-th type of fish, P i represents the market price of the i-th type of fish; n represents the number of fish species in the target clustering cluster; represents the equipment procurement cost; B j represents the quantity of the i-th type of equipment; P j represents the unit price of the i-th type of equipment; represents the fish group feeding cost; F t represents the feeding amount on the t-th day, T represents the feeding cycle; P f represents the unit price of the bait; represents the fish drug input cost, F k represents the drug dosage for the k-th disease, P s represents the unit price of the fish drug, K represents the number of disease occurrences; C 人工 represents the labor input cost, C 研发 represents the R & D cost, C 运维 represents the operation and maintenance cost;

[0048] If the sales input ratio is lower than the preset threshold, then according to the set adjustment strategy, adjust the frequency and quantity of automatic feeding and drug administration; the adjustment strategy is obtained based on an intelligent adjustment model, and the intelligent adjustment model is constructed using a deep learning model, which can automatically optimize the adjustment strategy according to the historical sales input data to ensure the maximization of the aquaculture benefits of the fish farm.

[0049] Compared with the prior art, the present invention has the following advantages and beneficial effects: It realizes the comprehensive digitization of fishery farm management, effectively improving the intelligent level of fishery farming; through the data acquisition module, it can obtain the key data of the fishery farm in real time and accurately, providing a solid foundation for subsequent intelligent control. The intelligent control module automatically adjusts the working state of the fishery farm management equipment according to the data analysis results, thus realizing the precise management of the fishery farm; in addition, the management and maintenance module not only provides powerful data display and traceability functions, but also can implement management and maintenance according to the abnormal data of tail water treatment, further ensuring the stable operation of the fishery farm.

[0050] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written specification and the drawings.

[0051] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0053] Figure 1 is a schematic structural diagram of a digital fishery integrated management platform;

[0054] Figure 2 is a schematic structural diagram of the data acquisition module;

[0055] Figure 3 is a schematic structural diagram of the intelligent control module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0057] The present invention provides a digital fishery integrated management platform, as Figure 1 shown, including:

[0058] A data acquisition module for collecting and obtaining the target management data of the fishery farm;

[0059] An intelligent control module for intelligently controlling the fishery farm by using the fishery farm management equipment according to the analysis results of the target management data;

[0060] A management and maintenance module for displaying and maintaining the target management data and managing the process of intelligent control.

[0061] The working principle of the above technical solution is as follows: In order to implement a digital fishery integrated management platform, the present invention proposes a data acquisition module, which can efficiently and accurately collect various target management data from the fishing ground, such as water quality parameters, fish growth conditions, feed consumption, etc., providing a solid data foundation for subsequent intelligent analysis and control; the intelligent control module relies on advanced data analysis algorithms to deeply mine and analyze the collected data, identify the optimal strategies for fishing ground management, and automatically adjust the working status of fishing ground management equipment to achieve intelligent operations such as precise feeding and water quality regulation, thereby greatly improving the fishing production efficiency and management level; the management and maintenance module undertakes the important responsibilities of data display and management and intelligent control process supervision. Through an intuitive user interface, managers can monitor the fishing ground status in real time, adjust management strategies in a timely manner, and ensure the stable operation of the platform.

[0062] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, the intelligent and refined level of fishery management can be significantly improved; firstly, the efficient operation of the data acquisition module ensures the timeliness and accuracy of various data in the fishing ground, providing strong support for scientific decision-making; secondly, through intelligent analysis and control, the intelligent control module realizes the automation and precision of fishing ground management, effectively avoiding human errors and resource waste in traditional fishery management; finally, the introduction of the management and maintenance module not only facilitates managers to monitor and manage the fishing ground status in real time, but also enhances the stability and reliability of the platform, ensuring the continuity and efficiency of fishing production.

[0063] In one embodiment, as Figure 2 shown, the data acquisition module includes a water quality parameter data acquisition unit, a tail water treatment data acquisition unit, and a fish activity data acquisition unit;

[0064] The water quality parameter data acquisition unit is used to measure the first water quality data of the fishing ground based on water quality meters configured in the feeding ponds and outer ponds of the fishing ground; the first water quality data includes, but is not limited to, dissolved oxygen, dissolved oxygen saturation, ammonia nitrogen, nitrite, temperature, and pH value;

[0065] The tail water treatment data acquisition unit is used to measure the second water quality data of the discharged tail water based on water quality detection probes; the second water quality data includes, but is not limited to, total phosphorus, total nitrogen, suspended solids, potassium permanganate, and pH value;

[0066] The fish activity data acquisition unit is used to obtain fish body activity image data by shooting with a camera configured in the feeding pond of the fishing ground.

[0067] The working principle of the above technical solution is as follows: The data acquisition module realizes the comprehensive monitoring of the fishery environment by integrating various sensors and image acquisition devices; the water quality parameter data acquisition unit uses a high-precision water quality meter to continuously monitor the key water quality indicators in the fishery, such as dissolved oxygen, dissolved oxygen saturation, ammonia nitrogen, nitrite, temperature, and pH value. These data are crucial for evaluating the water quality status of the fishery, predicting the growth trend of fish, and formulating reasonable management strategies; the tail water treatment data acquisition unit focuses on the quality monitoring of the discharged tail water, and accurately measures parameters such as total phosphorus, total nitrogen, suspended solids, potassium permanganate, and pH value through water quality detection probes to ensure that the tail water discharge meets environmental protection requirements and reduces the impact on the environment. For example, an ultraviolet spectrophotometer, a turbidity sensor, and an oxidation-reduction potential probe are integrated to form a multi-parameter monitoring node at the tail water discharge outlet; the fish activity data acquisition unit uses an infrared thermal imaging and visible light dual-spectrum camera to continuously shoot in the feeding area.

[0068] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, these data acquisition units work together to jointly constitute the data foundation of the digital fishery comprehensive management platform, providing strong support for subsequent intelligent analysis and control.

[0069] In one embodiment, as Figure 3 shown, the intelligent control module includes a target management data analysis unit and an intelligent control implementation unit;

[0070] The target management data analysis unit is used to perform data analysis on the first water quality data, the second water quality data, and the fish body activity image data to obtain a data analysis result;

[0071] The intelligent control implementation unit is used to perform intelligent control on the fishery by using fishery management equipment according to the data analysis result.

[0072] The working principle of the above technical solution is as follows: The intelligent control module is the core of the entire digital fishery integrated management platform. Through in-depth analysis of various types of data and intelligent decision-making, it realizes the automated management of the fishing ground. The target management data analysis unit first receives data from the water quality parameter data collection unit, the tail water treatment data collection unit, and the fish activity data collection unit, namely the first water quality data, the second water quality data, and the fish body activity image data. These data cover all aspects of the fishing ground, including water quality conditions, tail water discharge situations, and fish activities. During the data analysis process, the target management data analysis unit will use advanced algorithms and models to deeply mine and comprehensively analyze these data. By comparing historical data, setting thresholds, and trend prediction, etc., the data analysis unit can identify possible problems or potential risks in the fishing ground, such as water quality deterioration, abnormal fish growth, etc. At the same time, it can also propose corresponding management suggestions or control strategies according to the analysis results. The intelligent control implementation unit is responsible for converting the data analysis results into specific control instructions and executing these instructions through the fishing ground management equipment. These equipment may include aerators, feeding machines, water pumps, etc. They can automatically adjust their working states according to the instructions of the intelligent control implementation unit to improve water quality, optimize feeding strategies, or adjust the fishing ground environment.

[0073] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, precise management and efficient operation of the fishing ground can be achieved. It can not only improve the production efficiency and product quality of the fishing ground, but also reduce the operation cost and environmental risk, providing strong support for the sustainable development of fishery.

[0074] In one embodiment, data analysis is performed on the first water quality data, the second water quality data, and the fish body activity image data to obtain data analysis results, including:

[0075] Based on the configured Internet of Things and cloud server, the first water quality data, the second water quality data, and the fish body activity image data are transmitted to the processor of the cloud server through the LoRa or NB-IoT protocol.

[0076] Based on the processor's data analysis of the first water quality data, the second water quality data, and the fish body activity image data, data analysis results are obtained.

[0077] The working principle of the above technical solution is as follows: Through the Internet of Things technology, various sensors and monitoring devices in the fish farm can collect first water quality data (such as dissolved oxygen, pH value, temperature, etc.) and second water quality data (such as key water quality indicators such as ammonia nitrogen and nitrite), as well as image data of fish bodies. These data are first transmitted through low-power wide-area network communication technologies such as LoRa or NB-IoT to ensure that the data can be stably and reliably sent to the cloud server; on the cloud server, a powerful processor deeply analyzes and processes the received data. The processor uses advanced algorithms and models to compare, screen, and comprehensively analyze these data, so as to obtain a comprehensive assessment of the water quality status of the fish farm, the health status of fish, and the operation efficiency of the fish farm. These data analysis results can not only help fish farm managers discover and solve potential problems in a timely manner, but also provide strong support for optimizing fish farm management strategies, improving production efficiency, and product quality. For example: When the suspended solid concentration at the tail water discharge outlet rises abnormally after a heavy rain, the turbidity sensor of the discharge pipeline detects that the turbidity value jumps from 12 to 45 (the exceeding standard threshold is 30), then a data packet containing a timestamp and the exceeding standard parameter is generated and directly transmitted to the cloud server through the 4G network.

[0078] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, real-time transmission and analysis of data can be realized, greatly improving the timeliness and accuracy of data analysis. Through the combination of the Internet of Things and the cloud server, fish farm managers can obtain the latest fish farm data anytime and anywhere, so as to make decisions in a timely manner; in addition, the application of the LoRa or NB-IoT protocol also ensures the stability and reliability of data transmission, avoiding problems such as data loss or transmission delay.

[0079] In one embodiment, based on the processor's data analysis of the first water quality data, the second water quality data, and the fish body activity image data, data analysis results are obtained, including:

[0080] Based on the set first water quality data threshold library, the first water quality data is compared, and the first data comparison result is obtained. According to the first data comparison result, the first water quality analysis result is generated;

[0081] Based on the set second water quality data threshold library, the second water quality data is compared, and the second data comparison result is obtained. According to the second data comparison result, the second water quality analysis result is generated;

[0082] Based on the fish feature extraction model based on transfer learning, for the fish body activity image data, the facial texture and body shape features of the fish are extracted; based on the deep convolutional neural network model, according to the facial texture and body shape features of the fish, fish growth trend prediction, fish disease type identification, and fish disease development trend prediction are carried out to obtain the fish body activity image analysis result.

[0083] The working principle of the above technical solution is as follows: First, the processor compares the collected first water quality data (such as dissolved oxygen, temperature, pH value, etc.) one by one according to the preset first water quality data threshold library to determine whether these data exceed the normal threshold range. If the data exceeds the threshold, corresponding alarm information is generated, and the exceeded water quality parameters and their specific values are listed in detail to form the first water quality analysis result. Then, the processor performs similar data comparison and analysis on the second water quality data (such as ammonia nitrogen, nitrate, heavy metal content, etc.) according to the second water quality data threshold library to generate the second water quality analysis result. This step helps to comprehensively evaluate the safety and health status of the fish farm water quality. When processing the fish body activity image data, the processor first uses the fish feature extraction model of transfer learning to extract the facial texture and body shape features of the fish in the image. These features can reflect the growth status, breed information, and potential disease conditions of the fish. Subsequently, the processor uses the deep convolutional neural network model to predict the growth trend of the fish, identify the types of fish diseases, and predict the development trend of fish diseases according to the extracted fish features. This process can automatically and accurately identify the health status of the fish, providing timely disease warnings and treatment suggestions for the fish farm managers.

[0084] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, through intelligent data processing and analysis technologies, the comprehensive monitoring and management of the fish farm water quality and fish health status are realized, providing a strong guarantee for the sustainable development of the fishery.

[0085] In one embodiment, when predicting the growth trend of fish, identifying the types of fish diseases, and predicting the development trend of fish diseases to obtain the fish body activity image analysis result, it includes:

[0086] According to the facial texture and body shape features of the fish, fitting and generating the coordinate curves of the body lengths of different types of fish and the growth time, and generating the growth cycle diagrams of different types of fish;

[0087] According to the body shape features of the fish, based on the set disease fish system feature library, perform the matching of disease types to obtain the types of fish diseases;

[0088] According to the body shape features of the fish in different periods, based on the set long short-term memory network model, predict the development trend of fish diseases to obtain the prediction result of the development trend of fish diseases;

[0089] Based on the fish body growth cycle diagram, the types of fish diseases, and the prediction result of the development trend of fish diseases, generate the fish body activity image analysis result.

[0090] The working principle of the above technical solution is as follows: Through image processing and data analysis technologies, the growth trend of fish can be accurately captured and analyzed. First, using the facial texture and body shape characteristics of fish, the platform can fit the coordinate curves of the body length and growth time of various fish, which not only shows the growth speed of fish, but also intuitively reveals the growth laws and cycles of various fish by generating growth cycle diagrams of different types of fish bodies; this provides valuable reference information for fish farm managers and helps them formulate reasonable breeding strategies according to the growth stages of fish. In terms of identifying the types of fish diseases, the platform can quickly and accurately match the possible types of fish diseases by comparing the body shape characteristics of fish with the established characteristic library of diseased fish systems; this feature-based matching method not only improves the accuracy of disease identification, but also greatly shortens the identification time, winning valuable time for the timely prevention and treatment of diseases. Among them, to solve the situation where the accuracy of fish body recognition is low due to factors such as light changes and water turbidity, and the spatio-temporal correlation characteristics of group behavior cannot be effectively captured, the present invention proposes a three-dimensional spatio-temporal attention network architecture. By integrating the spatial attention mechanism and the temporal attention mechanism, refined recognition of fish group behavior is achieved. Specifically: An improved ResNet-50 is used as the backbone network, and a deformable convolutional layer is introduced at the Conv4_x stage to capture the deformation feature vector of the fish body; based on the deformation feature vector of the fish body, a spatial attention weight matrix is generated; a Transformer encoder structure is adopted, and relative position encoding is introduced; a typical behavior template library is generated through a clustering algorithm, and a behavior similarity metric is defined; according to the difference in similarity, the abnormal behavior of the fish group is captured. At the same time, the present invention also uses a long short-term memory network model to predict the development trend of fish diseases. This model can learn and simulate the development law of diseases according to the body shape characteristics of fish in different cycles, so as to predict the development trend of fish diseases in the future for a period of time.

[0091] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, through the comprehensive application of technologies such as image processing, data analysis, and machine learning, accurate prediction and analysis of the growth trend, disease types, and disease development trend of fish are realized, the intelligent level of fish farm management is improved, and strong technical support is provided for the sustainable development of the fishery.

[0092] In one embodiment, according to the data analysis results, the fish farm is intelligently controlled using fish farm management equipment, including:

[0093] Configure fish farm management equipment; the fish farm management equipment includes an automatic feeder, an aerator, and a drug delivery device;

[0094] Based on the fishery management equipment, according to the data analysis results, match the feeding frequency and feeding amount based on the fish growth stage, adjust the power of the aerator, and control the drug dosage in combination with the types of fish diseases and the prediction results of the development trend of fish diseases.

[0095] The working principle of the above technical solution is as follows: First, by collecting and analyzing the environmental parameters, fish growth data, and disease monitoring results in the fishery, a comprehensive understanding of the fishery operation status is obtained; Subsequently, based on this data, the optimal fishery management strategy is calculated; For example, at different stages of fish growth, the system will automatically adjust the feeding frequency and feed types of the feeder to ensure that the fish obtain appropriate nutritional supply and promote their healthy growth; At the same time, it will also dynamically adjust the power of the aerator according to the water quality monitoring results to maintain the dissolved oxygen content in the water within an appropriate range and provide a good living environment for the fish; In addition, when fish diseases are detected, the drug delivery device will be immediately activated, and according to the types and severity of the diseases, the drug dosage will be accurately controlled to effectively prevent the spread of diseases and ensure the health and safety of the fish.

[0096] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, through intelligent device control, precise regulation of the fishery environment is achieved; Specifically, according to the fish growth stage and real-time data analysis results, the automatic feeder can match the most suitable feeding frequency and feed amount to ensure that the fish obtain sufficient nutrition without overfeeding; The aerator adjusts the power according to the water quality situation to keep the dissolved oxygen content in the water within an appropriate range; The drug delivery device combines the disease identification results to accurately control the drug dosage, which not only effectively treats diseases but also avoids the abuse of drugs.

[0097] In one embodiment, the management and maintenance module includes a data display unit and a management and maintenance unit;

[0098] The data display unit is used to build a visualization platform integrating the Web end and the mobile end to realize the real-time display and traceability tracking of fishery data; The real-time display includes displaying the three-dimensional distribution map of the fishery; The three-dimensional distribution map is constructed based on a 3D model and is marked with the equipment and data involved in the data acquisition module and the intelligent control module; The traceability tracking includes the information tracking of fish individuals and the traceability tracking of fishery management data; The information tracking of fish individuals includes: Based on the RFID tag or two-dimensional code identification set on the fish body, the information tracking of the growth cycle and sales flow of fish individuals is carried out; The traceability tracking of fishery management data includes: The traceability tracking of the breeding record data, breeding cycle data, and warehouse update record data of the fishery.

[0099] The management and maintenance unit is used to implement management and maintenance according to the target management data.

[0100] The working principle of the above technical solution is as follows: When the data acquisition module or the intelligent control module detects abnormal data during the tail water treatment process, the management and maintenance unit will immediately receive and analyze this data; for example, if it detects that the water quality parameters exceed the preset range, or a certain device fails, the management and maintenance unit will automatically trigger the alarm mechanism and send an alarm message to the management personnel through the visualization platform; the management personnel can view the alarm details immediately through the Web end or the mobile end and quickly locate the problem; in addition, the management and maintenance unit can also automatically generate maintenance tasks or suggestions according to the target management data, and these tasks or suggestions will be directly pushed to the relevant maintenance personnel or teams to ensure that the problem is solved in a timely manner. After the maintenance personnel complete the maintenance work, they can submit a maintenance report through the platform, detailing information such as the maintenance process, replaced parts, and processing results; this information will be traced and stored in the system as part of the fishery management data for future reference and analysis.

[0101] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, through the management and maintenance module, the digital fishery comprehensive management platform not only realizes the real-time display and traceability tracking of fishery data, but also improves the efficiency and accuracy of target management data, providing strong support for the sustainable development of fishery.

[0102] In one embodiment, according to the target management data, the implementation of management and maintenance includes:

[0103] Controlling the warning device to give a warning, and controlling the configured water pollution collection equipment to start a failure, and generating a maintenance work order to be pushed to the configured management and maintenance terminal;

[0104] Recording the equipment inspection log and maintenance history, and uploading it in real time through the applet end;

[0105] Configuring an expert Q&A link interface to provide technical guidance based on the university resources or large database of the link.

[0106] The working principle of the above technical solution is as follows: When the target management data is abnormal, the management and maintenance module will respond quickly. First, by controlling warning devices such as audio-visual alarms, it sends warning signals to on-site personnel to ensure that problems can be discovered in time. At the same time, this module will automatically start the configured water pollution collection equipment to temporarily store the possibly polluted water body and prevent the spread of pollution. Immediately afterwards, according to the type and analysis results of the abnormal data, it will automatically generate a maintenance work order and push it to the configured management and maintenance terminal, such as the mobile phone or computer of the management personnel, through the network, to ensure that maintenance personnel can quickly obtain maintenance information. During the maintenance process, the management and maintenance module will also record the equipment inspection logs and maintenance history. These log information not only includes basic information such as the time, location, and type of the fault, but also includes detailed information such as the response time of the maintenance personnel, maintenance steps, and replaced parts. These information will be uploaded to the digital fishery comprehensive management platform in real time through the applet end, facilitating management personnel to view and analyze at any time. In addition, in order to provide aquaculture technical guidance, the management and maintenance module is also configured with an expert Q&A link interface. This interface can be linked to university resources or large databases. Aquaculture personnel can directly conduct online consultations with aquaculture experts by clicking the link to obtain professional aquaculture suggestions and technical support, which not only improves aquaculture efficiency but also reduces economic losses caused by technical problems.

[0107] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment, the digital fishery comprehensive management platform realizes intelligent management in terms of maintenance, which not only improves the processing efficiency and accuracy but also provides a strong guarantee for the sustainable development of the fishery.

[0108] In one embodiment, the management and maintenance module further includes an intelligent adjustment unit for input management; the intelligent adjustment unit for input management is used to calculate the sales input ratio based on the analysis of the clustering distribution of fish schools, combined with fish school sales data and input data, and perform intelligent adjustment of the fishery input according to the comparison between the sales input ratio and the set threshold; specifically including:

[0109] Using the DBSCAN clustering analysis method, based on fish body activity image data, perform density clustering processing on the fish school distribution in the set water area space of the fishery to identify multiple clustering clusters of several types of fish; among them, the clustering parameters of the DBSCAN clustering analysis method are: the clustering radius is between 0.45N and 0.85N, where N is the larger value of the average body length and average body width of the fish bodies of the identified fish; the minimum sample number is set to 1;

[0110] Among the multiple clustering clusters, screen out the clustering clusters greater than the preset density threshold as the target clustering clusters;

[0111] Estimate the number of fish bodies in the target clustering cluster and determine the average growth cycle of the fish school in the target clustering cluster;

[0112] Obtain the input data for automatic feeding and drug administration based on the average growth cycle;

[0113] Obtain the equipment procurement cost data, labor input cost data, R & D cost data, and operation and maintenance cost data;

[0114] Predict the sales data of the number of fish bodies in the target clustering cluster based on historical sales data;

[0115] Combine the input data for automatic feeding and drug administration, equipment procurement cost data, labor input cost data, R & D cost data, operation and maintenance cost data, and sales data to calculate the sales input ratio; the calculation formula is as follows:

[0116]

[0117] In the above formula, E represents the sales input ratio, Q i represents the sales volume of the i-th type of fish, P i represents the market price of the i-th type of fish; n represents the number of fish species in the target clustering cluster; represents the equipment procurement cost; B j represents the quantity of the i-th type of equipment; P j represents the unit price of the i-th type of equipment; represents the fish feeding cost; F t represents the feeding amount on the t-th day, T represents the feeding cycle; P f represents the unit price of the bait; represents the fish drug input cost, F k represents the drug dosage for the k-th disease, P s represents the unit price of the fish drug, K represents the number of disease occurrences; C 人工 represents the labor input cost, C 研发 represents the R & D cost, C 运维 represents the operation and maintenance cost;

[0118] If the sales input ratio is lower than the preset threshold, then according to the set adjustment strategy, adjust the frequency and quantity of automatic feeding and drug administration; the adjustment strategy is obtained based on an intelligent adjustment model, and the intelligent adjustment model is constructed using a deep learning model, which can automatically optimize the adjustment strategy according to historical sales input data to ensure the maximization of the aquaculture benefits of the fish farm.

[0119] The working principle of the above technical solution is as follows: By collecting the image data of the fish school and using the DBSCAN clustering analysis method to perform density clustering on the fish school, the clustering clusters of different fish species can be accurately identified, and then the distribution of the fish school can be analyzed. For the target clustering clusters with higher density, the system can estimate the number of fish bodies and determine the average growth cycle, which provides a scientific basis for subsequent input management; after obtaining the automatic feeding and drug administration input data, equipment procurement cost data, labor input cost data, R & D cost data, and operation and maintenance cost data, the system combines the historical sales data and can accurately obtain the sales input ratio through complex calculation formulas. This indicator intuitively reflects the economic benefits of the fish farm and provides decision-making support for managers; when the sales input ratio is lower than the preset threshold, the system will automatically trigger an adjustment mechanism to optimize the frequency and quantity of automatic feeding and drug administration according to the intelligent adjustment model. This intelligent adjustment process not only improves the management efficiency but also helps to maximize the aquaculture benefits of the fish farm.

[0120] The beneficial effects of the above technical solution are as follows: By adopting the solution provided in this embodiment and integrating the intelligent adjustment unit for input management, the precise management and intelligent optimization of the fish farm input are realized, bringing a significant improvement in benefits to fishery production.

[0121] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A digital fishery comprehensive management platform, characterized in that, Including: A data acquisition module, which is used to acquire target management data of the fishing ground; An intelligent control module, which is used to intelligently control the fishing ground by using fishing ground management equipment according to the analysis result of the target management data; A management and maintenance module, which is used to display and maintain the target management data and manage the process of intelligent control.

2. The digital fishery comprehensive management platform according to claim 1, characterized in that, The data acquisition module includes a water quality parameter data acquisition unit, a tail water treatment data acquisition unit and a fish activity data acquisition unit; The water quality parameter data acquisition unit is used to measure the first water quality data of the fishing ground based on the water quality meters configured in the feeding ponds and outer ponds of the fishing ground; the first water quality data includes but is not limited to dissolved oxygen, dissolved oxygen saturation, ammonia nitrogen, nitrite, temperature and pH value; The tail water treatment data acquisition unit is used to measure the second water quality data of the discharged tail water based on the water quality detection probe; the second water quality data includes but is not limited to total phosphorus, total nitrogen, suspended solids, potassium permanganate and pH value; The fish activity data acquisition unit is used to obtain fish body activity image data by shooting with a camera configured in the feeding pond of the fishing ground.

3. The digital fishery comprehensive management platform according to claim 2, characterized in that, The intelligent control module includes a target management data analysis unit and an intelligent control implementation unit; The target management data analysis unit is used to perform data analysis on the first water quality data, the second water quality data and the fish body activity image data to obtain an analysis result; The intelligent control implementation unit is used to intelligently control the fishing ground by using fishing ground management equipment according to the analysis result.

4. The digital fishery comprehensive management platform according to claim 3, characterized in that, Performing data analysis on the first water quality data, the second water quality data and the fish body activity image data to obtain an analysis result, including: Based on the configured Internet of Things and cloud server, transmitting the first water quality data, the second water quality data and the fish body activity image data to the processor of the cloud server through the LoRa or NB-IoT protocol; Based on the processor, performing data analysis on the first water quality data, the second water quality data and the fish body activity image data to obtain an analysis result.

5. The digital fishery integrated management platform according to claim 4, wherein, Based on the processor, performing data analysis on the first water quality data, the second water quality data and the fish body activity image data to obtain an analysis result, including: Based on the set first water quality data threshold library, performing data comparison on the first water quality data to obtain a first data comparison result, and generating a first water quality analysis result according to the first data comparison result; Based on the set second water quality data threshold library, performing data comparison on the second water quality data to obtain a second data comparison result, and generating a second water quality analysis result according to the second data comparison result; Based on the fish feature extraction model of transfer learning, extracting the facial texture and body shape features of fish from the fish body activity image data; based on the deep convolutional neural network model, predicting the fish growth trend, identifying the types of fish diseases and predicting the development trend of fish diseases according to the facial texture and body shape features of fish to obtain the fish body activity image analysis result.

6. The digital fishery comprehensive management platform according to claim 5, characterized in that, Predicting the fish growth trend, identifying the types of fish diseases and predicting the development trend of fish diseases to obtain the fish body activity image analysis result, including: According to the facial texture and body shape features of fish, fitting and generating the coordinate curves of the body lengths and growth times of different types of fish, and generating the growth cycle diagrams of different types of fish; According to the body shape characteristics of fish, based on the set characteristic library of diseased fish systems, match the types of diseases to obtain the types of fish body diseases; According to the body shape characteristics of fish in different periods, based on the set long short-term memory network model, predict the development trend of fish body diseases to obtain the prediction result of the development trend of fish body diseases; Based on the fish body growth cycle diagram, the types of fish body diseases, and the prediction result of the development trend of fish body diseases, generate the analysis result of fish body activity images.

7. The digital fishery comprehensive management platform according to claim 6, characterized in that, According to the data analysis result, use the fishery management equipment to perform intelligent control on the fishery, including: Configure the fishery management equipment; the fishery management equipment includes an automatic feeding machine, an aerator, and a drug administration device; Based on the fishery management equipment, according to the data analysis result, match the feeding frequency and feeding amount based on the fish growth stage, adjust the power of the aerator, and control the drug administration dose in combination with the types of fish body diseases and the prediction result of the development trend of fish body diseases.

8. A digital fishery integrated management platform according to claim 1, characterized in that, The management and maintenance module includes a data display unit and a management and maintenance unit; The data display unit is used to construct a visualization platform integrating the Web side and the mobile side to realize the real-time display and traceability tracking of fishery data; The real-time display includes displaying the three-dimensional distribution map of the fishery; the three-dimensional distribution map is constructed based on a 3D model and is marked with the equipment and data involved in the data acquisition module and the intelligent control module; The traceability tracking includes the information tracking of fish individuals and the traceability tracking of fishery management data; the information tracking of fish individuals includes: based on the RFID tag or two-dimensional code identification set on the fish body, perform information tracking on the growth cycle and sales flow of fish individuals; the traceability tracking of fishery management data includes: perform traceability tracking on the breeding record data, breeding cycle data, and warehouse update record data of the fishery; The management and maintenance unit is used to implement management and maintenance according to the target management data.

9. The digital fishery comprehensive management platform according to claim 8, characterized in that, Implementing management and maintenance according to the target management data includes: Controlling the warning device to give a warning, controlling the configured water pollution receiving equipment to start a fault, and generating a maintenance work order to push to the configured management and maintenance terminal; Recording the equipment inspection log and maintenance history and uploading it in real time through the applet side; Configure an expert Q&A link interface to provide technical guidance based on the university resources or large database of the link.

10. A digital fishery comprehensive management platform according to claim 7, characterized in that, The management and maintenance module also includes an input management intelligent adjustment unit; the input management intelligent adjustment unit is used to analyze the clustering distribution of the fish population, combine the fish population sales data and input data, calculate the sales input ratio, and perform intelligent adjustment of the fishery input according to the comparison between the sales input ratio and the set threshold; specifically including: Adopt the DBSCAN clustering analysis method to perform density clustering processing on the fish population distribution in the set water area space of the fishery based on the fish body activity image data to identify multiple clustering clusters of several types of fish; among them, the clustering parameters of the DBSCAN clustering analysis method are: the clustering radius is between 0.45N and 0.85N, where N is the larger value of the average body length and average body width of the fish body of the identified fish; the minimum number of samples is set to 1; Among the multiple clustering clusters, screen out the clustering clusters greater than the preset density threshold as the target clustering clusters; Estimate the number of fish bodies in the target clustering cluster and determine the average growth cycle of the fish school in the target clustering cluster; Obtain the input data for automatic feeding and drug administration based on the average growth cycle; Obtain the equipment procurement cost data, labor input cost data, R & D cost data, and operation and maintenance cost data; Predict the sales data of the number of fish bodies in the target clustering cluster based on the historical sales data; Combine the input data for automatic feeding and drug administration, equipment procurement cost data, labor input cost data, R & D cost data, operation and maintenance cost data, and sales data to calculate the sales input ratio; the calculation formula is as follows: In the above formula, E represents the sales investment ratio, and Q i represents the sales volume of the i-th type of fish, and P i represents the market price of the i-th type of fish; n represents the number of fish species in the target clustering cluster; represents the equipment procurement cost; B j represents the quantity of the i-th type of equipment; P j represents the unit price of the i-th type of equipment; represents the fish feeding cost; F t represents the feeding amount on the t-th day, and T represents the feeding cycle; P f represents the unit price of the bait; represents the fish medicine input cost, and F k represents the dosage of the medicine for the k-th disease, and P s represents the unit price of the fish medicine, and K represents the number of disease occurrences; C 人工 represents the labor input cost, and C 研发 represents the R & D cost, and C 运维 represents the operation and maintenance cost; If the sales input ratio is lower than the preset threshold, then adjust the frequency and quantity of automatic feeding and drug administration according to the set adjustment strategy; the adjustment strategy is obtained based on the intelligent adjustment model, and the intelligent adjustment model is constructed using a deep learning model, which can automatically optimize the adjustment strategy according to the historical sales input data to ensure the maximization of the breeding benefits of the fish farm.

Images