Aquaculture monitoring and early warning method and system based on artificial intelligence

By combining sensors, cameras and meteorological services in aquaculture, using artificial intelligence to calculate risk indexes, multi-dimensional early warning of water level, water quality, environment and meteorology is achieved, solving the accuracy and reliability of traditional aquaculture management, and improving the The effectiveness of early warning.

CN118329110BActive Publication Date: 2024-11-05CAIMADAME ENTERPRISE MANAGEMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410279355.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-12
Publication Date
2024-11-05
Estimated Expiration
2044-03-12

AI Technical Summary

Technical Problem

The existing aquaculture management mainly relies on artificial experience, lacks accuracy, effectiveness and reliability, and it is difficult to effectively use artificial intelligence for multi-dimensional evaluation and early warning.

Method used

设计了一种基于人工智能的水产养殖监控预警系统,结合水位水质监测传感器、光学监控摄像头、红外监控摄像头和本地气象服务网站,通过4G/5G网络和HTTP协议传输数据,利用风险指数公式计算鱼 Multi-dimensional risk levels of pond water level, water quality, environment and meteorology, and warnings are pushed through WeChat official accounts.

Benefits of technology

It realizes the identification of floating objects on the water surface of fish ponds, fish corpses and illegal fishing behaviors, and eliminates small weather stations. Through the combination of video image recognition and weather forecast data, the accuracy and reliability of early warning are improved, and multi-dimensional Risk assessment and early warning methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118329110B_ABST
    Figure CN118329110B_ABST
Patent Text Reader

Abstract

The present invention discloses an artificial intelligence-based aquaculture monitoring and early warning method and system, comprising: a water level and water quality monitoring sensor, an optical monitoring camera, an infrared monitoring camera, a local meteorological service website, a PC system server, and a remote mobile terminal; the water level and water quality monitoring sensor is used for collecting fish pond water level depth data and water quality data, the optical monitoring camera and the infrared monitoring camera are used for real-time collection of fish pond on-site video, the local meteorological service website is used for providing local meteorological forecast data for the day, 7 days and 14 days, the remote mobile terminal is used for displaying risk levels and specific risk information, and the PC system server is used for receiving video images from the optical monitoring camera and the infrared monitoring camera. The invention combines the fish pond water level, water quality, environment and meteorological data, calculates the risk index using a risk level formula, and pushes it in the form of a WeChat public account, thereby realizing a multi-dimensional early warning method for aquaculture monitoring.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of aquaculture, and in particular to an aquaculture monitoring and early warning method and system based on artificial intelligence. Background Art

[0002] Aquaculture is one of the fastest growing industries in China's agricultural structure. Since 1989, my country's aquatic product output has ranked first in the world and has remained the world's number one for more than 30 years. However, my country's traditional aquaculture management is mainly based on manual experience management, which is time-consuming and labor-intensive, with insufficient risk estimation, and lacks accuracy, effectiveness and reliability.

[0003] With the development of science and technology, the aquaculture industry has also seen the emergence of monitoring and early warning methods based on artificial intelligence technology. For example, Chinese patent CN116228031A discloses an aquaculture intelligent aquaculture monitoring and early warning system based on artificial intelligence, which uses oxygen concentration and nitrogen and phosphorus concentration to evaluate water quality and risk early warning assessment. For example, Chinese patent CN115200647A discloses an aquaculture environmental monitoring and management method and system, which uses water temperature, pH value, dissolved oxygen, conductivity, turbidity, ammonia nitrogen, COD, floating objects and other parameters for monitoring, and the evaluation model and dimensions are limited. Therefore, how to effectively use artificial intelligence, introduce risk index formulas, and evaluate and warn in multiple dimensions such as fish pond water level, water quality, environment, and meteorology is a technical problem that urgently needs to be solved. Summary of the invention

[0004] In view of this, the present invention proposes an artificial intelligence-based aquaculture monitoring and early warning method and system to solve the problems of existing multi-dimensional evaluation and early warning methods in fish pond water level, water quality, environment, weather, etc.

[0005] The technical solution of the present invention is implemented as follows: On the one hand, the present invention provides an artificial intelligence-based aquaculture monitoring and early warning system, which includes a water level and water quality monitoring sensor, an optical monitoring camera, a PC system server, an infrared monitoring camera, a local meteorological service website, and a remote mobile terminal;

[0006] The water level and water quality monitoring sensor is connected to the PC system server 4G / 5G network, and is used to measure the water level and water quality data information of the fish pond, and send the data information to the PC system server in HTTP protocol;

[0007] The optical monitoring camera is connected to the 4G / 5G network of the PC system server, and is used to monitor the environment with sufficient light conditions in the fish pond, and send the video information to the PC system server in HTTP protocol;

[0008] The infrared surveillance camera is connected to the 4G / 5G network of the PC system server to monitor the biometric environment of the fish pond and send the video information to the PC system server in HTTP protocol;

[0009] The local weather service website is connected to the PC system server through the Internet, and is used for the local weather service website to provide local weather information to the PC system server in HTTP protocol;

[0010] The PC system server is connected to the remote mobile terminal 4G / 5G network, and is used to receive video images from the optical surveillance camera and the infrared surveillance camera, identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities through software, and calculate the risk index through the early warning level formula in combination with the information received from the water level and water quality monitoring sensor and the local meteorological service website, divide the risk level and send specific information to the WeChat public account on the remote mobile terminal, prompting the fish pond manager to take corresponding measures.

[0011] Preferably, the water level and water quality monitoring sensor includes a water level probe, a water temperature probe, a CO2 concentration probe, a turbidity probe, a pH probe, a dissolved oxygen probe, a conductivity probe, an ammonia oxygen probe, an ammonia nitrogen probe, a nitrite concentration probe, a COD probe, a blue-green algae concentration probe, an MCU controller and a 4G / 5G module; the water level probe, the water temperature probe, the CO2 concentration probe, the turbidity probe, the pH probe, the dissolved oxygen probe, the conductivity probe, the ammonia oxygen probe, the ammonia nitrogen probe, the nitrite concentration probe, the COD probe, and the blue-green algae concentration probe are electrically connected to the MCU controller, and are used to receive water temperature, CO2 concentration, turbidity, pH value, dissolved oxygen, conductivity, ammonia oxygen, ammonia nitrogen, nitrite concentration, COD, and blue-green algae concentration signals and send the signals to the MCU controller;

[0012] Preferably, the optical monitoring camera and the infrared monitoring camera can collect fish pond video information and send it to the PC system server to identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities;

[0013] Preferably, the local weather service website can provide the PC system server with local temperature, air pressure, humidity and wind forecast data for the current day, 7 days and 14 days.

[0014] The MCU controller is electrically connected to the 4G / 5G module, and is used to receive the electrical signals sent by all probes and convert the electrical signals into digital signals through the MCU controller, and send them to the PC system server through the 4G / 5G module.

[0015] Preferably, the MCU controller includes an embedded system, a TIMER controller, a UART controller, an AD controller and a RAM memory;

[0016] The TIMER controller, UART controller, AD controller and RAM memory transmit information with the embedded system via an internal bus;

[0017] The TIMER controller is used to generate timing according to the requirements of the serial port baud rate;

[0018] The UART controller exchanges information with the embedded system through the internal bus, receives and analyzes the external serial port TTL electrical signal, and converts the data output by the embedded system into a serial port TTL electrical signal according to the serial port baud rate requirements and outputs it to the 4G / 5G module;

[0019] The AD controller exchanges information with the embedded system via an internal bus, receives external analog electrical signals and converts them into digital electrical signals and outputs them to the embedded system;

[0020] The RAM memory is used to provide an address space for program execution and an address space for temporary data storage;

[0021] The functions of the embedded system, the UART controller and the AD controller are all implemented on the MCU hardware by C language code;

[0022] Preferably, the MCU module hardware adopts the STM32F407VET6 model and is compiled using KeiluVision 5 software.

[0023] On the other hand, the present invention provides an aquaculture monitoring and early warning method based on artificial intelligence; comprising the following steps:

[0024] S1. The water level and water quality monitoring sensor sends the fish pond water level and water quality information to the IP address and port number of the PC system server. The PC system server receives the following information: water level height X0, water temperature X1, CO2 concentration X2, turbidity X3, pH value X4, dissolved oxygen content X5, conductivity X6, ammonia oxygen content X7, ammonia nitrogen content X8, nitrite concentration X9, COD content X10, water temperature X11, CO2 concentration X12, water temperature X13, water temperature X14, water temperature X15, water temperature X16, water temperature X17, water temperature X18, water temperature X19, water temperature X20, water temperature X21 10 , blue-green algae concentration X 11 ;

[0025] S2. PC system server sends information X0-X 11 Substitute the water level and water quality risk index formula to calculate the water level and water quality risk index;

[0026] S3. The optical surveillance camera and the infrared surveillance camera send video images to the IP address and port number of the PC system server. The PC system server stores the video information in real time and captures the images. The AI ​​analysis method is used to analyze and identify the number of floating objects on the surface of the fish pond N0, the number of fish carcasses N1, and the number of illegal fishing behaviors N2;

[0027] S4. The PC system server brings the information N0-N2 into the fish pond environmental risk index formula to calculate the fish pond environmental risk index;

[0028] S5. The PC system server accesses the local weather service website to obtain the local temperature forecasts T0, T1 and T2 for the current day, 7 days, and 14 days, the air pressure forecasts A0, A1 and A2, the humidity forecasts H0, H1 and H2, and the wind force forecasts F0, F1 and F2;

[0029] S6. The PC system server brings the meteorological information T0, T1, T2, A0, A1, A2, H0, H1, H2, F0, F1, F2 into the fish pond meteorological risk index formula to calculate the fish pond meteorological risk index;

[0030] S7.PC system server divides water level and water quality risk index, environmental risk index and meteorological risk index into water level and water quality risk level, environmental risk level and meteorological risk level, and sends detailed information to the remote mobile terminal through WeChat public account. If the risk is high, it will be sent every 10 minutes until the user receives the information and replies to confirm.

[0031] The artificial intelligence-based aquaculture monitoring and early warning method and system of the present invention have the following beneficial effects compared with the prior art:

[0032] (1) The combination of optical surveillance cameras and infrared surveillance cameras enables the identification of floating objects on the surface of fish ponds, fish carcasses, and illegal fishing activities;

[0033] (2) By accessing the local weather service website, local weather forecast data for the day, seven days and 14 days are provided, eliminating the need for a physical small weather station;

[0034] (3) Formulas for water level and water quality risk index, environmental risk index, and meteorological risk index were constructed. The water level and water quality risk index was obtained by iterating historical data. The environmental risk index was obtained by introducing video image recognition probability and the duration of illegal fishing behavior. The meteorological risk index was obtained by combining short-term meteorological forecasts with medium- and long-term forecasts. The risk level was divided and received through a remote mobile terminal and displayed on the WeChat public account. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 This is a structural diagram of an aquaculture monitoring and early warning system based on artificial intelligence of the present invention;

[0037] Figure 2 It is the structural diagram of the MCU controller of the present invention;

[0038] Figure 3 The present invention is a flowchart of the artificial intelligence-based aquaculture monitoring and early warning method and system. DETAILED DESCRIPTION

[0039] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Embodiment 1

[0041] like Figure 1-3 As shown, an artificial intelligence-based aquaculture monitoring and early warning system of the present invention includes a water level and water quality monitoring sensor, an optical monitoring camera, a PC system server, an infrared monitoring camera, a local meteorological service website, and a remote mobile terminal;

[0042] The water level and water quality monitoring sensor is connected to the PC system server 4G / 5G network, and is used to measure the water level and water quality data information of the fish pond, and send the data information to the PC system server in HTTP protocol;

[0043] The optical monitoring camera is connected to the 4G / 5G network of the PC system server, and is used to monitor the environment with sufficient light conditions in the fish pond, and send the video information to the PC system server in HTTP protocol;

[0044] The infrared surveillance camera is connected to the 4G / 5G network of the PC system server to monitor the biometric environment of the fish pond and send the video information to the PC system server in HTTP protocol;

[0045] The local weather service website is connected to the PC system server through the Internet, and is used for the local weather service website to provide local weather information to the PC system server in HTTP protocol;

[0046] The PC system server is connected to the remote mobile terminal 4G / 5G network, and is used to receive video images from the optical surveillance camera and the infrared surveillance camera, identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities through software, and calculate the risk index through the early warning level formula in combination with the information received from the water level and water quality monitoring sensor and the local meteorological service website, divide the risk level and send specific information to the WeChat public account on the remote mobile terminal, prompting the fish pond manager to take corresponding measures.

[0047] This system uses water level and water quality monitoring sensors, optical monitoring cameras, PC system servers, infrared monitoring cameras, local meteorological service websites, and remote mobile terminals to implement an artificial intelligence-based aquaculture monitoring and early warning system. The system combines fish pond water level, water quality, environment, and meteorological data, uses the risk level formula to calculate the risk index, and pushes it through the WeChat public account, realizing a multi-dimensional early warning method for aquaculture monitoring; through design, simulation, and verification, a modular product is formed, which can achieve rapid transplantation between different platforms and accelerate the product development process.

[0048] Among them, the water level and water quality monitoring sensor includes a water level probe, a water temperature probe, a CO2 concentration probe, a turbidity probe, a pH probe, a dissolved oxygen probe, a conductivity probe, an ammonia oxygen probe, an ammonia nitrogen probe, a nitrite concentration probe, a COD probe, a blue-green algae concentration probe, an MCU controller and a 4G / 5G module; the water level probe, the water temperature probe, the CO2 concentration probe, the turbidity probe, the pH probe, the dissolved oxygen probe, the conductivity probe, the ammonia oxygen probe, the ammonia nitrogen probe, the nitrite concentration probe, the COD probe, and the blue-green algae concentration probe are electrically connected to the MCU controller, and are used to receive water temperature, CO2 concentration, turbidity, pH value, dissolved oxygen, conductivity, ammonia oxygen, ammonia nitrogen, nitrite concentration, COD, and blue-green algae concentration signals and send the signals to the MCU controller;

[0049] The MCU controller is electrically connected to the 4G / 5G module, and is used to receive the electrical signals sent by all probes and convert the electrical signals into digital signals through the MCU controller, and send them to the PC system server through the 4G / 5G module.

[0050] Among them, the 4G / 5G module is implemented by Fibocom FG132, and the water level probe, water temperature probe, CO2 concentration probe, turbidity probe, pH value probe, dissolved oxygen probe, conductivity probe, ammonia oxygen probe, ammonia nitrogen probe, nitrite concentration probe, COD probe, and blue-green algae concentration probe all use 0-10V or 4-20mA output.

[0051] The optical monitoring camera and the infrared monitoring camera can collect fish pond video information and send it to the PC system server to identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities;

[0052] The local weather service website can provide the PC system server with local temperature, air pressure, humidity and wind forecast data for the current day, 7 days and 14 days.

[0053] The MCU controller includes an embedded system, a TIMER controller, a UART controller, an AD controller and a RAM memory; the TIMER controller, the UART controller, the AD controller and the RAM memory transmit information with the embedded system through an internal bus; the TIMER controller is used to generate timing according to the requirements of the serial port baud rate; the UART controller exchanges information with the embedded system through the internal bus, receives and analyzes the external serial port TTL electrical signal, and at the same time, according to the requirements of the serial port baud rate, converts the data output by the embedded system into a serial port TTL electrical signal and outputs it to the 4G / 5G module; the AD controller exchanges information with the embedded system through the internal bus, receives the external analog electrical signal and converts it into a digital electrical signal and outputs it to the embedded system; the RAM memory is used to provide an address space for program running and an address space for data temporary storage; the functions of the embedded system, the UART controller and the AD controller are all implemented on the MCU hardware by C language code.

[0054] Among them, the MCU hardware uses ST's STM32F407VET6, which is compiled using Keil uVision 5 software.

[0055] The aquaculture monitoring and early warning system in this embodiment adopts water level and water quality monitoring sensors, optical monitoring cameras, PC system servers, infrared monitoring cameras, local meteorological service websites and remote mobile terminals to realize an aquaculture monitoring and early warning system based on artificial intelligence. The system combines fish pond water level, water quality, environment and meteorological data, calculates the risk index using the risk level formula, and pushes it through the WeChat public account, realizing a multi-dimensional early warning method for aquaculture monitoring; through design, simulation and verification, a modular product is formed, which can achieve rapid transplantation between different platforms and accelerate the product development process.

[0056] Embodiment 2

[0057] Provided is an artificial intelligence-based aquaculture monitoring and early warning method, which uses the artificial intelligence-based aquaculture monitoring and early warning system as described in Example 1, and includes the following steps:

[0058] 1. The water level and water quality monitoring sensor sends the fish pond water level and water quality information to the IP address and port number of the PC system server. The PC system server receives the following information: water level height X0, water temperature X1, CO2 concentration X2, turbidity X3, pH value X4, dissolved oxygen content X5, conductivity X6, ammonia oxygen content X7, ammonia nitrogen content X8, nitrite concentration X9, COD content X10, water level X11, water temperature X12, CO2 concentration X13, water level X14, water level X15, water level X16, water level X17, water level X18, water level X19, water level X20, water level X21 10 , blue-green algae concentration X 11 ;

[0059] S2. PC system server sends information X0-X 11 Substitute the water level and water quality risk index formula to calculate the water level and water quality risk index;

[0060] S3. The optical surveillance camera and the infrared surveillance camera send video images to the IP address and port number of the PC system server. The PC system server stores the video information in real time and captures the images. The AI ​​analysis method is used to analyze and identify the number of floating objects on the surface of the fish pond N0, the number of fish carcasses N1, and the number of illegal fishing behaviors N2;

[0061] S4. The PC system server brings the information N0-N2 into the fish pond environmental risk index formula to calculate the fish pond environmental risk index;

[0062] S5. The PC system server accesses the local weather service website to obtain the local temperature forecasts T0, T1 and T2 for the current day, 7 days, and 14 days, the air pressure forecasts A0, A1 and A2, the humidity forecasts H0, H1 and H2, and the wind force forecasts F0, F1 and F2;

[0063] S6. The PC system server brings the meteorological information T0, T1, T2, A0, A1, A2, H0, H1, H2, F0, F1, F2 into the fish pond meteorological risk index formula to calculate the fish pond meteorological risk index;

[0064] S7.PC system server divides water level and water quality risk index, environmental risk index and meteorological risk index into water level and water quality risk level, environmental risk level and meteorological risk level, and sends detailed information to the remote mobile terminal through WeChat public account. If the risk is high, it will be sent every 10 minutes until the user receives the information and replies to confirm.

[0065] Among them, the water level and water quality risk index formula is and Here, X iIndicates that the parameters collected by the water level and water quality monitoring sensors correspond to the 0th to 11th real-time data. represents the ideal data of the 0th to 11th parameters, λ i Indicates the criticality of the 0th to 11th parameters, H w (n) represents the water level and water quality risk index calculated for the Nth time after the fish pond is changed, H w (n-1) represents the water level and water quality risk index calculated for the N-1th time after the fish pond is changed. The initial value H w (0)=0.

[0066] The AI ​​analysis method uses the NI VISION function of the Labview programming software to extract the images transmitted by the optical surveillance camera, extract the appearance contour features of the floating objects on the water surface and compare them with the shape of the fish carcasses. If it is not in the shape of a fish carcass, it is counted as the number of floating objects N0, otherwise it is counted as the number of fish carcasses N1, and then the images transmitted by the infrared surveillance camera are extracted to analyze whether there are people with the appearance contour features of the human body staying around the fish pond for a long time. If so, the images transmitted by the optical surveillance camera are retrieved to analyze whether the person has the contour features of a handheld fishing tool. If so, it is counted as illegal fishing behavior N2.

[0067] The fish pond environmental risk index formula is and Here, N0 represents the number of floating objects, λ0 represents the degree of hazard caused by the number of floating objects, represents the average accuracy of floating object identification, N1 represents the number of fish carcasses, λ1 represents the degree of harm caused by the number of fish carcasses, represents the average accuracy of fish carcass identification, N2 represents the number of illegal fishing activities, λ2 represents the degree of harm caused by the number of illegal fishing activities, represents the average accuracy of illegal fishing behavior identification, T represents the duration of illegal fishing behavior, and T f Indicates the duration threshold of illegal fishing behavior, H e Represents the fish pond environmental risk index.

[0068] The formula for the fish pond meteorological risk index is and λ T +λ A +λ H +λ F =1, here, T i It indicates the local temperature forecast data for the day, 7 days and 14 days. T indicates the ideal temperature for fish survival. Q Ti represents the weight of the deviation ratio of the temperature forecast data for the day, 7 days and 14 days to the total temperature impact, λ T Indicates the degree of damage of temperature to fish ponds, Ai Indicates the local weather pressure forecast data for the day, 7 days and 14 days. The ideal air pressure for fish survival, Q Ai It represents the weight of the deviation ratio of the air pressure forecast data for the day, 7 days and 14 days to the total air pressure impact, λ A Indicates the degree of damage of air pressure to fish ponds, H i Indicates the local humidity forecast data for the day, 7 days and 14 days. Indicates the ideal humidity for fish survival, Q Hi represents the weight of the current day, 7-day and 14-day humidity forecast data deviation ratio to the total humidity impact, λ H Indicates the degree of harm of humidity to fish ponds, F i Indicates the local wind forecast data for the day, 7 days and 14 days. F indicates the ideal wind for fish survival, Q Fi It represents the weight of the deviation ratio of the wind forecast data for the day, 7 days and 14 days to the total wind impact, λ F Indicates the degree of damage caused by wind to the fish pond, H m Represents the meteorological risk index of fish ponds.

[0069] The water level and water quality risk level, environmental risk level and meteorological risk level are defined as high risk when the water quality and water level, environmental and meteorological risk index are in the range [0.7, +∞), medium risk when the risk index is in the range [0.4, 0.7), and low risk in other ranges.

[0070] like Figure 3 As shown, the water level and water quality monitoring sensor sends water quality and water level information to the PC system server, and the PC system server calculates the water quality and water level risk index according to the water quality and water level risk formula. The optical monitoring camera and the infrared monitoring camera send video images to the PC system server. The PC system server identifies floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities. The PC system server calculates the fish pond environmental risk index according to the fish pond environmental risk formula. Then the PC system server accesses the local meteorological service website to obtain the local temperature forecast, air pressure forecast, humidity forecast and wind forecast for the day, 7 days and 14 days. The PC system server calculates the fish pond meteorological risk index according to the fish pond meteorological formula, and finally divides the risk level and pushes the information to the remote mobile terminal WeChat public account.

[0071] The artificial intelligence-based aquaculture monitoring and early warning method in this embodiment adopts a modular design method, uses water level and water quality monitoring sensors, optical monitoring cameras, PC system servers, infrared monitoring cameras, local meteorological service websites and remote mobile terminals, realizes artificial intelligence aquaculture system monitoring and early warning, and combines fish pond water level, water quality, environment, and meteorological data, calculates the risk index using the risk level formula, and pushes it through the WeChat public account, realizing a multi-dimensional early warning method for aquaculture monitoring.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence-based aquaculture monitoring and early warning method, characterized in that: The steps include: S1. The water level and water quality monitoring sensor sends the fish pond water level and water quality information to the IP address and port number of the PC system server, and the PC system server receives the information Including: water level height X0, water temperature X1, CO2 concentration X2, turbidity X3, pH value X4, dissolved oxygen X5, conductivity X6, ammonia oxygen X7, ammonia nitrogen X8, nitrite concentration X9, COD amount X 10 , blue-green algae concentration X 11 ; S2. PC system server sends information X0-X 11 Substitute the water level and water quality risk index formula to calculate the water level and water quality risk index; S3. The optical surveillance camera and the infrared surveillance camera send video images to the IP address and port number of the PC system server. The PC system server stores the video information in real time and captures the images. The AI ​​analysis method is used to analyze and identify the number of floating objects on the surface of the fish pond N0, the number of fish carcasses N1, and the number of illegal fishing behaviors N2; S4. The PC system server brings the information N0-N2 into the fish pond environmental risk index formula to calculate the fish pond environmental risk index; the fish pond environmental risk index formula is and Here, N0 represents the number of floating objects, λ0 represents the degree of hazard caused by the number of floating objects, represents the average accuracy of floating object identification, N1 represents the number of fish carcasses, λ1 represents the degree of harm caused by the number of fish carcasses, represents the average accuracy of fish carcass identification, N2 represents the number of illegal fishing activities, λ2 represents the degree of harm caused by the number of illegal fishing activities, represents the average accuracy of illegal fishing behavior identification, T represents the duration of illegal fishing behavior, and T f Indicates the duration threshold of illegal fishing behavior, H e represents the fish pond environmental risk index; S5. The PC system server accesses the local weather service website to obtain the local temperature forecasts T0, T1 and T2 for the current day, 7 days, and 14 days, the air pressure forecasts A0, A1 and A2, the humidity forecasts H0, H1 and H2, and the wind force forecasts F0, F1 and F2; S6. The PC system server brings the meteorological information T0, T1, T2, A0, A1, A2, H0, H1, H2, F0, F1, F2 into the fish pond meteorological risk index formula to calculate the fish pond meteorological risk index; the fish pond meteorological risk index formula is and λ T +λ A +λ H +λ F =1, here, T i Indicates the local temperature forecast data for the day, 7 days and 14 days. represents the ideal temperature for fish survival, Q Ti represents the weight of the deviation ratio of the temperature forecast data for the day, 7 days and 14 days to the total temperature impact, λ T Indicates the degree of damage of temperature to fish ponds, A i Indicates the local weather pressure forecast data for the day, 7 days and 14 days. The ideal air pressure for fish survival, Q Ai It represents the weight of the deviation ratio of the air pressure forecast data for the day, 7 days and 14 days to the total air pressure impact, λ A Indicates the degree of damage of air pressure to fish ponds, H i Indicates the local humidity forecast data for the day, 7 days and 14 days. Indicates the ideal humidity for fish survival, Q Hi represents the weight of the deviation ratio of humidity forecast data for the day, 7 days and 14 days to the total humidity impact, λ H Indicates the degree of harm of humidity to fish ponds, F i Indicates local wind forecast data for the day, 7 days and 14 days. The ideal wind speed for fish survival, Q Fi It represents the weight of the deviation ratio of the wind forecast data for the day, 7 days and 14 days to the total wind impact, λ F Indicates the degree of damage caused by wind to the fish pond, H m It represents the meteorological risk index of fish pond; S7.PC system server divides water level and water quality risk index, environmental risk index and meteorological risk index into water level and water quality risk level, environmental risk level and meteorological risk level, and sends detailed information to the remote mobile terminal through WeChat public account. If the risk is high, it will be sent every 10 minutes until the user receives the information and replies to confirm.

2. The artificial intelligence-based aquaculture monitoring and early warning method according to claim 1, characterized in that: The step S2 specifically includes: The water level and water quality risk index formula is and Here, X i Indicates that the parameters collected by the water level and water quality monitoring sensors correspond to the 0th to 11th real-time data. represents the ideal data of the 0th to 11th parameters, λ i Indicates the criticality of the 0th to 11th parameters, H w (n) represents the water level and water quality risk index calculated for the Nth time after the fish pond is changed, H w (n-1) represents the water level and water quality risk index calculated for the N-1th time after the fish pond is changed. The initial value H w (0)=0.

3. The artificial intelligence-based aquaculture monitoring and early warning method according to claim 1, characterized in that: The step S3 specifically includes: The AI ​​analysis method uses the NI VISION function of the Labview programming software to extract the images transmitted by the optical surveillance camera, extract the appearance contour features of the floating objects on the water surface and compare them with the shape of the fish carcasses. If it is not in the shape of a fish carcass, it is counted as the number of floating objects N0, otherwise it is counted as the number of fish carcasses N1, and then the images transmitted by the infrared surveillance camera are extracted to analyze whether there are people with the appearance contour features of the human body staying around the fish pond for a long time. If so, the images transmitted by the optical surveillance camera are retrieved to analyze whether the person has the contour features of a handheld fishing tool. If so, it is counted as illegal fishing behavior N2.

4. The artificial intelligence-based aquaculture monitoring and early warning method according to claim 1, characterized in that: The step S7 specifically includes: The water level and water quality risk level, environmental risk level and meteorological risk level are defined as high risk when the water quality and water level, environmental and meteorological risk index are in the range [0.7, +∞), medium risk when the risk index is in the range [0.4, 0.7), and low risk when the rest of the ranges are defined.

5. An artificial intelligence-based aquaculture monitoring and early warning system, which adopts the method according to any one of claims 1 to 4; characterized in that: It mainly includes water level and water quality monitoring sensors, optical monitoring cameras, PC system servers, infrared monitoring cameras, local meteorological service websites, and remote mobile terminals; The water level and water quality monitoring sensor is connected to the PC system server 4G / 5G network, and is used to measure the water level and water quality data information of the fish pond, and send the data information to the PC system server in HTTP protocol; The optical monitoring camera is connected to the 4G / 5G network of the PC system server, and is used to monitor the environment with sufficient light conditions in the fish pond, and send the video information to the PC system server in HTTP protocol; The infrared surveillance camera is connected to the 4G / 5G network of the PC system server to monitor the biometric environment of the fish pond and send the video information to the PC system server in HTTP protocol; The local weather service website is connected to the PC system server through the Internet, and is used for the local weather service website to provide local weather information to the PC system server in HTTP protocol; The PC system server is connected to the remote mobile terminal 4G / 5G network, and is used to receive video images from the optical surveillance camera and the infrared surveillance camera, identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities through software, and calculate the risk index through the early warning level formula in combination with the information received from the water level and water quality monitoring sensor and the local meteorological service website, divide the risk level and send specific information to the WeChat public account on the remote mobile terminal, prompting the fish pond manager to take corresponding measures.

6. The artificial intelligence-based aquaculture monitoring and early warning system as claimed in claim 5, characterized in that: The water level and water quality monitoring sensor includes a water level probe, a water temperature probe, a CO2 concentration probe, a turbidity probe, a pH probe, a dissolved oxygen probe, a conductivity probe, an ammonia oxygen probe, an ammonia nitrogen probe, a nitrite concentration probe, a COD probe, a blue-green algae concentration probe, an MCU controller and a 4G / 5G module; The water level probe, the water temperature probe, the CO2 concentration probe, the turbidity probe, the pH value probe, the dissolved oxygen probe, the conductivity probe, the ammonia oxygen probe, the ammonia nitrogen probe, the nitrite concentration probe, the COD probe, and the blue-green algae concentration probe are electrically connected to the MCU controller, and are used to receive water temperature, CO2 concentration, turbidity, pH value, dissolved oxygen, conductivity, ammonia oxygen, ammonia nitrogen, nitrite concentration, COD, and blue-green algae concentration signals and send the signals to the MCU controller; The MCU controller is electrically connected to the 4G / 5G module, and is used to receive the electrical signals sent by all probes and convert the electrical signals into digital signals through the MCU controller, and send them to the PC system server through the 4G / 5G module.

7. The artificial intelligence-based aquaculture monitoring and early warning system as claimed in claim 5, characterized in that: The optical monitoring camera and the infrared monitoring camera can collect fish pond video information and send it to the PC system server to identify floating objects on the surface of the fish pond, fish carcasses and illegal fishing activities; The local weather service website can provide the PC system server with local temperature, air pressure, humidity and wind forecast data for the current day, 7 days and 14 days.

8. The artificial intelligence-based aquaculture monitoring and early warning system as claimed in claim 6, characterized in that: The MCU controller includes an embedded system, a TIMER controller, a UART controller, an AD controller and a RAM memory; The TIMER controller, UART controller, AD controller and RAM memory transmit information with the embedded system via an internal bus; The TIMER controller is used to generate timing according to the requirements of the serial port baud rate; The UART controller exchanges information with the embedded system through the internal bus, receives and analyzes the external serial port TTL electrical signal, and converts the data output by the embedded system into a serial port TTL electrical signal according to the serial port baud rate requirements and outputs it to the 4G / 5G module; The AD controller exchanges information with the embedded system via an internal bus, receives external analog electrical signals and converts them into digital electrical signals and outputs them to the embedded system; The RAM memory is used to provide an address space for program execution and an address space for temporary data storage; The functions of the embedded system, the UART controller and the AD controller are all implemented on MCU hardware by C language code.