Intelligent observation device for fish hemorrhagic disease and behavioral fever

Abstract

The application discloses an intelligent observation device for fish hemorrhagic disease and behavioral fever, and relates to the field of intelligent agriculture.In the device, the cylinder body comprises multiple cavities and multiple intermediate walls; different cavities are separated by the intermediate walls and grid baffles; the grid baffles are arranged between the intermediate walls; the grid baffles are used for connecting the water bodies of different cavities and isolating the fish in different cavities; a water pump is used for realizing water flow circulation between the cavities; an underwater camera and a sensor group are arranged in the water body of the cylinder body; the sensor group comprises multiple types of sensors; and a control module is used for controlling a temperature heating module to heat according to the data collected by the sensor group.The application can realize observation of the behavioral fever of fish diseases on and in water.

Description

Technical Field

[0001] This application relates to the field of smart agriculture, and in particular to an intelligent observation device for fish hemorrhagic disease and behavioral fever. Background Technology

[0002] Currently, artificial intelligence methods for detecting fish diseases are limited to image capture detection, lacking multi-data acquisition and control platforms that observe fish behavior, controllable flow fields, and temperature fields. Multiple sensors for diseased fish need to be integrated, and simultaneous detection above and below water is difficult to achieve. Furthermore, research on "behavioral fever" in fish diseases requires interconnected tanks at different temperatures, a device that currently lacks such a device. Summary of the Invention

[0003] The purpose of this application is to provide an intelligent observation device for fish hemorrhagic disease and behavioral fever, which can simultaneously observe behavioral fever in fish diseases both above and below water.

[0004] To achieve the above objectives, this application provides the following solution: This application provides an intelligent observation device for fish hemorrhagic disease and behavioral fever, comprising: Power and support components, circulation and flow guidance components, measurement and imaging components, and control module; The power and support components include: a cylinder and a water pump; the circulation and flow guiding components include: a mesh baffle; the measurement and imaging components include: a surface camera, an underwater camera, a sensor group, and a temperature heating module; the measurement and imaging components are used to observe fish hemorrhagic disease and behavioral fever; The cylinder body includes: multiple chambers and multiple intermediate walls; different chambers are separated by the intermediate walls and the mesh baffles; the mesh baffles are disposed between the intermediate walls; the mesh baffles are used to connect the water in different chambers and isolate fish in different chambers; the water pump is used to realize water circulation between the chambers; the underwater camera and the sensor group are both disposed in the water body of the cylinder body; the sensor group includes various types of sensors; the control module is used to control the temperature heating module to heat according to the data collected by the sensor group.

[0005] In one embodiment, the circulation and flow guiding component further includes: a separating mesh block; The separator mesh is located at the connection between the water pump and the cylinder.

[0006] In one embodiment, the number of chambers is two.

[0007] In one embodiment, the sensor group is fixed to the bottom of the cylinder; the underwater camera is fixed to both sides of the intermediate wall; and the temperature heating module is fixed to the intermediate wall and located in the water.

[0008] In one embodiment, the measurement and imaging assembly further includes: a sliding crossbeam plate, a connecting slider, a camera rod, a slide bar, and an elliptical ring slide rail; The sliding crossbeam is mounted on the intermediate wall; the elliptical ring slide rail is connected to the sliding crossbeam; the elliptical ring slide rail moves linearly on the sliding crossbeam; the camera pole is connected to the elliptical ring slide rail; and the underwater camera is mounted on the camera pole.

[0009] In one embodiment, the measurement and imaging assembly further includes: a slider and a connecting slider; The elliptical ring slide rail is connected to the sliding crossbeam plate via the slide rod; the slide rod is used to make the elliptical ring slide rail move linearly on the sliding crossbeam plate; the connecting slider is installed in the track of the elliptical ring slide rail; the camera rod is connected to the elliptical ring slide rail via the connecting slider.

[0010] In one embodiment, a slot is provided in the middle of the sliding crossbeam plate; the slot is connected to the slide rod.

[0011] In one embodiment, the elliptical ring slide rail is provided with an opening; the opening is used for loading and unloading the connecting slider.

[0012] In one embodiment, the sensor group includes a temperature sensor, a pH sensor, and a dissolved oxygen sensor.

[0013] In one embodiment, the control module uses deep reinforcement learning and PID control algorithms to control the temperature heating module to perform heating based on the data collected by the sensor group.

[0014] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides an intelligent observation device for fish hemorrhagic disease and behavioral fever. Different chambers are separated by an intermediate wall and a mesh baffle. Water circulation between the different chambers is achieved by a water pump, which can isolate direct contact between fish in different chambers. The spread of the virus in the fish population can only be through water. Based on a temperature heating module, "fish behavioral fever" can be observed. A surface camera and an underwater camera are set up to conduct observations both above and below water, thereby realizing the simultaneous observation of behavioral fever in fish diseases from both above and below water. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A schematic diagram of the overall structure of an intelligent observation device for fish hemorrhagic disease and behavioral fever; Figure 2 Exploded view of the overall structure of an intelligent observation device for fish hemorrhagic disease and behavioral fever; Figure 3 Diagram of the bottom structure of an intelligent observation device for fish hemorrhagic disease and behavioral fever; Figure 4 A partial structural diagram of the camera stand for an intelligent observation device for fish hemorrhagic disease and behavioral fever; Figure 5 Top view of the sensor section of the intelligent monitoring device for fish hemorrhagic disease and behavioral fever; Figure 6 A comparison chart showing the effects of two temperature control methods in region A; Figure 7 This is a comparison chart showing the effects of two temperature control methods in region B.

[0017] Reference numerals: 1-Cylinder block, 2-Grid baffle, 3-Sliding crossbeam plate, 4-Water pump, 5-Camera rod, 6-Camera group, 7-Sensor group, 8-Nut, 9-Connecting slider, 10-Slide rod, 11-Elliptical ring slide rail, 12-Separating mesh block, 13-Server, 14-Temperature heating module. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] In one exemplary embodiment, such as Figures 1-5 As shown, this application provides an intelligent observation device for fish hemorrhagic disease and behavioral fever, including: a power and support assembly, a circulation and diversion assembly, a measurement and imaging assembly, and a control module.

[0021] The power and support components include: a cylinder 1 and a water pump 4; the circulation and flow guiding components include: a mesh baffle 2; the measurement and imaging components include: a surface camera, an underwater camera, a sensor group 7, and a temperature heating module 14. The surface camera and the underwater camera constitute camera group 6. The measurement and imaging components are used to observe fish hemorrhagic disease and behavioral fever. Camera group 6 is used to collect video data.

[0022] The cylinder 1 includes multiple chambers and multiple intermediate walls; different chambers are separated by the intermediate walls and the mesh baffles 2; the mesh baffles 2 are disposed between the intermediate walls; the mesh baffles 2 are used to connect the water in different chambers and isolate fish in different chambers; the water pump 4 is used to realize water circulation between the chambers; the underwater camera and the sensor group 7 are both disposed in the water of the cylinder 1; the sensor group 7 includes various types of sensors; the control module is used to control the temperature heating module 14 to heat according to the data collected by the sensor group 7.

[0023] In an exemplary embodiment, the circulation and flow guiding assembly further includes: a separating mesh block 12; the separating mesh block 12 is disposed at the connection between the water pump 4 and the cylinder 1. The mesh baffle is a mesh-like baffle, installed in a slot on the middle wall of the cylinder 1, and is detachable. When the cylinder 1 is a two-chamber cylinder, it is used to separate the fish in the two chambers. The separating mesh block 12 is a mesh-like circular plug, embedded in the water pipe connecting the integrated water pump 4 to the channel of the cylinder 1, i.e. Figure 2 The openings on both sides of the middle tank 1 filter impurities and isolate the experimental diseased fish from the water pump.

[0024] In one exemplary embodiment, there are two chambers. The tank body 1 is the main container of the aquarium, an elliptical acrylic tank divided into two chambers separated by a mesh baffle. It is used to hold experimental water and the diseased fish to be observed, while also providing support. The water pump 4 is an integrated water pump installed outside the tank body 1 via a water pipe, enabling water circulation between the two chambers.

[0025] In an exemplary embodiment, the sensor group 7 is fixed to the bottom of the tank 1; the underwater camera is fixed to both sides of the intermediate wall; and the temperature heating module 14 is fixed to the intermediate wall and located in the water. In practical applications, the sensor group 7 includes a temperature sensor, a pH sensor, and a dissolved oxygen sensor. The sensors fixed to the bottom of the tank can be categorized as temperature sensors, pH sensors, and dissolved oxygen sensors according to different needs. The temperature heating module 14 is used to change the temperature inside the tank, and data is collected through the temperature sensor to facilitate analysis of the "behavioral hyperthermia" of fish in variable temperature fields.

[0026] In an exemplary embodiment, the measurement and shooting assembly further includes: a sliding crossbeam plate 3, a connecting slider 9, a camera rod 5, a slide bar 10, and an elliptical ring slide rail 11; the sliding crossbeam plate 3 is disposed on the intermediate wall; the elliptical ring slide rail 11 is connected to the sliding crossbeam plate 3; the elliptical ring slide rail 11 moves linearly on the sliding crossbeam plate 3; the camera rod 5 is connected to the elliptical ring slide rail 11; and the underwater camera is disposed on the camera rod 5.

[0027] In practical applications, the measurement and shooting assembly further includes: a slide rod 10 and a connecting slider 9; the elliptical ring slide rail 11 is connected to the sliding crossbeam plate 3 via the slide rod 10; the slide rod 10 is used to allow the elliptical ring slide rail 11 to move linearly on the sliding crossbeam plate 3; the connecting slider 9 is installed in the track of the elliptical ring slide rail 11; the camera rod 5 is connected to the elliptical ring slide rail 11 via the connecting slider 9. The slide rod 10 is a cylindrical rod that fixes the elliptical ring slide rail 11 to the sliding crossbeam plate 3, thereby enabling manual linear sliding. The connecting slider 9 is installed in the elliptical ring slide rail 11 and can be connected to the camera rod 5 via a nut 8, allowing manual sliding within the elliptical ring slide rail 11 to achieve a high degree of freedom in shooting tasks.

[0028] The sliding crossbeam plate 3 has a slot in the middle, which connects to the slide rod 10. The elliptical ring slide rail 11 has an opening for mounting and dismounting the connecting slider 9. The sliding crossbeam plate 3 is a detachable crossbeam plate, installed at both ends of the middle wall of the cylinder 1. The slot in the center of the plate connects to the slide rod 10, ensuring that the entire elliptical ring slide rail 11 moves linearly on the sliding crossbeam, providing a high degree of freedom. The elliptical ring slide rail 11 is an elliptical ring guide rail. The unique slot structure of the guide rail allows the connecting slider 9 to slide on the guide rail plane, achieving a high degree of freedom in shooting tasks; the openings on both sides facilitate the mounting and dismounting of the connecting slider 9.

[0029] The camera lever 5 is a mechanism for gripping the camera. It is mounted on the connecting slider 9, which mates with the elliptical ring slide rail 11. Displacement allows for omnidirectional camera coverage. The upper part of the camera lever 5 features a screw design for simple height adjustment, which can be adjusted according to the specific needs of the camera. In practical applications, a nut 8 is also included to secure the camera lever 5.

[0030] The detachable surface camera mounted on the surface camera pole 5 and the underwater camera fixed to both sides of the middle wall of the underwater tank 1 can be divided into visible light cameras and infrared cameras according to different needs. The captured data is transmitted to the server 13 for analysis in real time.

[0031] In practical applications, the measurement and imaging components also include a server 13, on which the control module is mounted. The server 13 is used to receive data, the data type of which is acquired by the sensor and transmitted via Bluetooth.

[0032] This application allows for the study of virus transmission in water through mechanical control. While virus transmission may occur through close contact between fish, this fish disease research device can isolate direct contact between fish, ensuring that virus transmission within a fish population occurs solely through water.

[0033] In an exemplary embodiment, the control module uses deep reinforcement learning and PID control algorithms to control the temperature heating module 14 to perform heating based on data collected by the sensor group 7. Specifically, the control action (proportional coefficient, integral coefficient, and control coefficient) corresponding to the PID control algorithm is determined using a reinforcement learning algorithm based on the real-time temperature, the target setpoint, and the current error. Then, the heating power of the temperature heating module is adjusted using the PID control algorithm based on the absolute temperature value, temperature difference, and temperature change rate.

[0034] The traditional PID algorithm, used as a control group, employs a completely independent dual-loop control, following a passive response logic of "correcting problems only after they occur." The controller in the high-temperature zone (Zone A) collects real-time temperature data and calculates errors, outputting positive power to drive the TEC heating system. The controller in the low-temperature zone (Zone B), initially not outputting power because the temperature meets the target, is unaware of the heating behavior in the high-temperature zone and also performs temperature heating and control. When heat is conducted through the medium, causing the actual temperature rise in the low-temperature zone and generating errors, the conventional PID algorithm's lagging data processing directly leads to a steady-state error of approximately +1.5°C in the low-temperature zone.

[0035] In contrast, the core TD3 deep reinforcement learning algorithm of this application is completely different, employing an active prediction mode. In data processing, the system no longer monitors a single point, but instead packages six dimensions of data—real-time temperature of both zones, target setpoint, and current error—into a state vector, which is then input into a centralized neural network. The trained AI agent has mastered the physical laws of thermal coupling, and when processing data, it can predict that heating in the high-temperature zone will interfere with the low-temperature zone. Therefore, at the execution level, when the AI ​​instructs the high-temperature zone to heat up, even if the temperature of the low-temperature zone has not yet fluctuated, it will reduce or even stop the heat power of the low-temperature zone based on the current state. This feedforward compensation mechanism based on global data offsets the impending heat flow, thereby precisely controlling the steady-state error within 0.1°C.

[0036] While receiving temperature data, server 13 sets the power output and temperature settings for heating at different temperatures. PID control uses traditional algorithms to set the temperature; this application employs a deep learning-based intelligent agent to control the temperature—adjusting the power output for zone heating through an algorithm. This intelligent agent refers to the control algorithm on server 13.

[0037] Temperature field control methods and temperature difference comparisons can be achieved by controlling temperature differences in different regions through mechanical structures and temperature field control. This is used to study "behavioral fever in fish"—sick fish "feverish" in different temperature zones according to their own condition, thereby enhancing their immunity. Fish are cold-blooded animals, and observing behavioral fever refers to fish moving to higher-temperature areas to improve their disease resistance when they are sick. High precision in temperature control is required.

[0038] This invention enables research on temperature control in different regions. Existing temperature control methods are unstable, with a range between 3-5℃. This application can reduce the temperature control to within 0.5℃. While 3℃ may not seem like a significant temperature difference for fish farming, it can lead to substantial differences in fish diseases.

[0039] The PID control law in server 13 is: .

[0040] in For output signal, For error signals, This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. For time.

[0041] PID temperature control is implemented in hardware as a high-frequency closed-loop feedback process. First, there's sensing and error calculation: the system periodically collects real-time water temperature data from sensors and compares it to the set target value to determine the current deviation. Next, the algorithm calculates the required energy output value by weighting and summing factors based on the magnitude of the deviation (the proportional term p provides the main thrust), the cumulative history of the deviation (the integral term i eliminates steady-state error), and the rate of change of the deviation (the derivative term d suppresses overshoot). Finally, there's execution and power regulation: the digital output is mapped to the duty cycle of a pulse-width modulation (PWM) signal. By rapidly controlling the on / off time ratio of the heater or TEC module, the average physical power applied to the water is linearly adjusted, ultimately stabilizing the temperature precisely at the set value.

[0042] In a dual-zone temperature difference scenario, heat is transferred to the second zone when the first zone is heated. The PID controller only stops inputting power after detecting an increase in temperature, and this lag means that the second zone can never stabilize at the setpoint, resulting in constant overshoot.

[0043] Decoupling mechanisms of Deep Reinforcement Learning (DRL) agents: Through simulation training, the reinforcement learning agent (TD3) based on the dual-delay deep deterministic policy gradient algorithm learns the following policy: .

[0044] in For the control actions output in region B, The standard feedback control term uses a base control force determined based on the error in region B. For coupling compensation terms, These represent the temperatures of regions A and B, respectively. Describe the influence of region A on region B.

[0045] When the agent observes T A With T B When there is a large temperature difference, neurons inside the network will activate the feedforward compensation path.

[0046] Feasibility verification and simulation conclusions: Simulation settings: Comparison: A parameter-tuned dual-loop PID controller vs. a temperature control agent trained for 60,000 steps.

[0047] Physical environment: includes heat capacity, ambient heat dissipation and strong thermal coupling coefficient (K=1.0).

[0048] Stability comparison: The PID controller exhibits continuous oscillation in Zone B, with a steady-state error of approximately +1.5°C.

[0049] Under DRL agent control, the Zone B temperature curve is smooth and without fluctuations, with a steady-state error of <0.1°C.

[0050] Figure 6 The comparison of the effects of the two temperature control methods in the first zone (zone A) shows that the traditional PID method results in the green line temperature oscillating continuously, while the red line represents the temperature control of the DRL agent combining PID and reinforcement learning, with a very small temperature difference. Figure 7For temperature control in the second zone (zone B), the purple line represents the temperature change of traditional PID control, which also oscillates continuously, while the blue line represents the temperature control of a DRL agent combining PID control and reinforcement learning, resulting in very small temperature changes. The first and second zones are two chambers divided in the cylinder by mesh baffles.

[0051] A continuous and stable temperature field environment allows for observation of changes in the behavioral state of fish in different stable regions, and enables research on "fish behavioral fever" and the transmission of viruses in water.

[0052] Fish diseases are highly sensitive to temperature changes, requiring precise temperature control within 1 degree Celsius. However, currently available fish behavior detection and analysis instruments in research and industry have relatively large temperature control precision, with field tests showing temperature differences of 3-5°C. Fish are cold-blooded animals, and according to this application, a 3-5°C temperature difference significantly affects the incidence of fish diseases, including disease transmission and symptom presentation. Furthermore, research on "behavioral fever" in fish diseases requires interconnected tanks at different temperatures, a currently lacking observation platform. This application, through an integrated structural design, achieves automatic circulation, purification, and oxygenation of the experimental water, providing a stable living environment for hemorrhagic fish. It also possesses multi-faceted, multi-modal observation capabilities, comprehensively analyzing hemorrhagic fish diseases by incorporating various indicator measurement devices. Underwater flow exchange is achieved through power and support components and circulation and flow guiding components; the measurement and imaging components allow for manual adjustment of the surface camera's shooting position and acquisition of various sensor data. Specific advantages are as follows: 1. Temperature field control method - deep reinforcement learning + traditional PID control algorithm, comparing temperature difference results to achieve precise temperature control.

[0053] 2. The water flow velocity between different fish chambers can be controlled by a peristaltic pump, namely pump 4. The peristaltic pump has adjustable power, and both ends of the pump have mesh structures to prevent fish from passing through.

[0054] 3. It can record and analyze fish behavior on and under water, and is equipped with visible light cameras and infrared cameras.

[0055] 4. Temperature differences between different regions can be achieved through mechanical structure control and temperature field control, which can be used to study "behavioral fever in fish"—sick fish "feverish" in different temperature zones according to their own condition, thereby enhancing their immunity. Fish are cold-blooded animals, and observing behavioral fever refers to fish moving to higher temperature areas to improve their disease resistance when they are sick. High precision in temperature control is required.

[0056] 5. The spread of the virus in water can be controlled by mechanical structures. The spread of the virus may be due to close contact between fish, but this application can isolate the direct contact between fish, and the spread of the virus in the fish group is only through water.

[0057] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An intelligent observation device for fish hemorrhagic disease and behavioral fever, characterized in that, The intelligent observation device for fish hemorrhagic disease and behavioral fever includes: a power and support assembly, a circulation and flow diversion assembly, a measurement and imaging assembly, and a control module; The power and support components include: a cylinder and a water pump; the circulation and flow guiding components include: a mesh baffle; the measurement and imaging components include: a surface camera, an underwater camera, a sensor group, and a temperature heating module; the measurement and imaging components are used to observe fish hemorrhagic disease and behavioral fever; The cylinder body includes: multiple chambers and multiple intermediate walls; different chambers are separated by the intermediate walls and the mesh baffles; the mesh baffles are disposed between the intermediate walls; the mesh baffles are used to connect the water in different chambers and isolate fish in different chambers; the water pump is used to realize water circulation between the chambers; the underwater camera and the sensor group are both disposed in the water body of the cylinder body; the sensor group includes various types of sensors; the control module is used to control the temperature heating module to heat according to the data collected by the sensor group.

2. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The circulation and flow guiding component further includes: a separating mesh block; The separator mesh is located at the connection between the water pump and the cylinder.

3. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The number of chambers is two.

4. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The sensor array is fixed to the bottom of the cylinder; the underwater camera is fixed to both sides of the intermediate wall; the temperature heating module is fixed to the intermediate wall and located in the water.

5. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The measurement and imaging components also include: a sliding crossbeam plate, a connecting slider, a camera rod, a slide bar, and an elliptical ring slide rail; The sliding crossbeam is mounted on the intermediate wall; the elliptical ring slide rail is connected to the sliding crossbeam; the elliptical ring slide rail moves linearly on the sliding crossbeam; the camera pole is connected to the elliptical ring slide rail; and the underwater camera is mounted on the camera pole.

6. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 5, characterized in that, The measurement and imaging components also include: a slider and a connecting slider; The elliptical ring slide rail is connected to the sliding crossbeam plate via the slide rod; the slide rod is used to make the elliptical ring slide rail move linearly on the sliding crossbeam plate; the connecting slider is installed in the track of the elliptical ring slide rail; the camera rod is connected to the elliptical ring slide rail via the connecting slider.

7. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 6, characterized in that, A slot is provided in the middle of the sliding crossbeam plate; the slot is connected to the sliding rod.

8. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 6, characterized in that, The elliptical ring slide rail is provided with an opening; the opening is used for loading and unloading the connecting slider.

9. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The sensor group includes a temperature sensor, a pH sensor, and a dissolved oxygen sensor.

10. The intelligent observation device for fish hemorrhagic disease and behavioral fever according to claim 1, characterized in that, The control module uses deep reinforcement learning and PID control algorithms to control the temperature heating module to perform heating based on the data collected by the sensor group.

Images