Aquaculture intelligent monitoring method and system based on Internet of Things

Abstract

The invention relates to the technical field of data processing, and discloses an aquaculture intelligent monitoring method and system based on the Internet of Things. The method comprises the following steps: collecting pond environment parameters and acoustic signals through layered arrangement of sensors, and obtaining multi-pond linkage data through wireless transmission; performing Fourier transform processing on the data to generate biology-environment coupling data; the water quality correlation degree between the ponds is calculated, and regional balance control data are formed; biological activity rules are analyzed, and biological rhythm feeding data are obtained; checking the carbon footprint to generate low-carbon operation data; and performing multi-dimensional correlation analysis to form ecological breeding optimization data. According to the method, the layered and partitioned ecological sensing network is constructed, collaborative analysis is performed on multi-pond environment data, and intelligent management of the breeding process is realized in combination with biological rhythm characteristics and carbon emission control requirements of aquatic animals, so that the breeding efficiency is improved, and the carbon emission intensity is reduced.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to an intelligent monitoring method and system for aquaculture based on the Internet of Things. Background Art

[0002] Aquaculture is an important part of modern agriculture. Traditional aquaculture mainly relies on manual experience for management, and adjusts aquaculture measures by regularly detecting water quality parameters and observing the growth status of aquaculture objects. With the development of Internet of Things technology, intelligent aquaculture monitoring systems have gradually been applied to the aquaculture field. Existing intelligent monitoring systems deploy devices such as water quality sensors and dissolved oxygen sensors to achieve real-time monitoring of water environment parameters, and automatically adjust the operation of aeration equipment through a control system. At the same time, by establishing an aquaculture database and analyzing historical data, decision-making support is provided for aquaculture management.

[0003] However, the existing intelligent monitoring methods for aquaculture have the following deficiencies: First, the deployment method of sensors is relatively simple, and it is difficult to reflect the three-dimensional distribution characteristics of the water environment; second, the monitoring system mainly focuses on the parameter changes of a single aquaculture pond, lacking the analysis of the environmental linkage relationship between multiple ponds; third, the equipment control strategy relies too much on preset thresholds and fails to fully consider the biological rhythm characteristics of aquatic animals; finally, the operation optimization of energy-consuming equipment such as aeration equipment mainly considers energy conservation, lacking the accurate accounting and optimized control of carbon emission intensity. These problems lead to low energy utilization efficiency during the aquaculture process and make it difficult to achieve eco-friendly aquaculture. Summary of the Invention

[0004] This application provides an intelligent monitoring method and system for aquaculture based on the Internet of Things. By constructing a hierarchical and zoned ecological perception network, collaborative analysis of multi-pond environmental data is carried out, and combined with the biological rhythm characteristics of aquatic animals and carbon emission control requirements, intelligent management of the aquaculture process is realized, thereby improving aquaculture efficiency and reducing carbon emission intensity.

[0005] In a first aspect, the present application provides an intelligent monitoring method for aquaculture based on the Internet of Things. The intelligent monitoring method for aquaculture based on the Internet of Things includes: collecting temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple aquaculture ponds through hierarchically arranged water quality sensors, and simultaneously collecting acoustic characteristic signals of aquatic animals, and using GPRS, 4G, WiFi, and Zigbee wireless transmission methods to collect and aggregate the parameters to obtain multi-pond ecological environment linkage data; performing Fourier transform processing on the water body stratification parameters and acoustic signals according to the multi-pond ecological environment linkage data, extracting the group behavior characteristics of aquatic animals and the water body flow characteristics, and generating bio-environment coupling data; calculating the water quality correlation degree and water body exchange period between multiple ponds according to the bio-environment coupling data, and coordinately adjusting the operating parameters of the aeration equipment in each pond to form regional balance control data; calculating the feeding amount based on the bio-environment coupling data and the regional balance control data by analyzing the daily and night activity rules and group behavior patterns of aquatic animals and combining the change trend of dissolved oxygen in each pond to obtain biological rhythm feeding data; calculating the carbon footprint of equipment energy consumption and dissolved gases in the water body according to the multi-pond ecological environment linkage data and the regional balance control data, and optimizing carbon emissions by adjusting the equipment operation timing to generate low-carbon operation data; performing multi-dimensional correlation analysis on biological indicators, environmental parameters, and carbon emission data during the aquaculture process according to the biological rhythm feeding data and the low-carbon operation data to form ecological aquaculture optimization data.

[0006] In a second aspect, the present application provides an intelligent monitoring system for aquaculture based on the Internet of Things. The intelligent monitoring system for aquaculture based on the Internet of Things includes:

[0007] A collection module, configured to collect temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple aquaculture ponds through hierarchically arranged water quality sensors, and simultaneously collect acoustic characteristic signals of aquatic animals, and use GPRS, 4G, WiFi, and Zigbee wireless transmission methods to collect and aggregate the parameters to obtain multi-pond ecological environment linkage data;

[0008] A transformation module, configured to perform Fourier transform processing on the water body stratification parameters and acoustic signals according to the multi-pond ecological environment linkage data, extract the group behavior characteristics of aquatic animals and the water body flow characteristics, and generate bio-environment coupling data;

[0009] An adjustment module, configured to calculate the water quality correlation degree and water body exchange period between multiple ponds according to the bio-environment coupling data, and coordinately adjust the operating parameters of the aeration equipment in each pond to form regional balance control data;

[0010] A calculation module, configured to calculate the feeding amount based on the biological-environment coupling data and the regional balance control data, by analyzing the diurnal activity rules and group behavior patterns of aquatic animals, and combining the change trends of dissolved oxygen in each pond, so as to obtain the biological rhythm feeding data;

[0011] A generation module, configured to calculate the carbon footprint of equipment energy consumption and dissolved water gases according to the multi-pond ecological environment linkage data and the regional balance control data, optimize carbon emissions by adjusting the equipment operation timing, and generate low-carbon operation data;

[0012] An association module, configured to perform multi-dimensional association analysis on biological indicators, environmental parameters, and carbon emission data during the breeding process according to the biological rhythm feeding data and the low-carbon operation data, so as to form ecological breeding optimization data.

[0013] In the technical solution provided by this application, through a hierarchically arranged water quality sensor network, the three-dimensional monitoring of water environment parameters is realized, and key indicators such as temperature, dissolved oxygen, and pH value of different water layers are accurately obtained. At the same time, the monitoring of water flow direction, flow velocity, and the collection of underwater acoustic characteristics are introduced, enriching the environmental monitoring dimension and providing a comprehensive data basis for subsequent analysis. The Fourier transform is used to process the water layer parameters and acoustic signals of the water body, extract the group behavior characteristics of aquatic animals and the water flow characteristics, establish a biological-environment coupling data model, and realize the quantitative association analysis of the behavior of breeding objects and environmental factors. Based on the biological-environment coupling data, the water quality correlation degree and water body exchange period between multiple ponds are calculated, realizing the collaborative regulation at the regional level and avoiding the regional imbalance problem that may be caused by the independent control of a single pond. By analyzing the diurnal activity rules and group behavior patterns of aquatic animals, and combining the change trends of dissolved oxygen in each pond to calculate the feeding amount, a precise feeding model based on biological rhythm is established, improving the feed utilization efficiency. During the operation of the equipment, by calculating the carbon footprint of energy consumption and dissolved water gases, the mapping relationship between equipment operation and carbon emissions is established, realizing the operation optimization oriented by low carbon. Finally, through multi-dimensional association analysis, the biological indicators, environmental parameters, and carbon emission data are comprehensively evaluated to form ecological breeding optimization data, providing a scientific basis for the decision-making in the breeding process. The Fourier transform algorithm adopted in this solution has significant advantages in processing periodic signals and is particularly suitable for analyzing the biological rhythm characteristics in aquaculture; the regional cooperation algorithm based on water quality correlation considers the spatial correlation between multiple ponds, improving the accuracy of the control strategy; the carbon footprint accounting model introduces environmental protection requirements into the intelligent control process, reflecting the innovation of the algorithm in specific application scenarios. Through the organic combination of these intelligent algorithms, the precise, collaborative, and low-carbon management goals of the breeding process are realized. Description of the Drawings

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0015] Figure 1 FIG. is a schematic diagram of an embodiment of the intelligent monitoring method for aquaculture based on the Internet of Things in the embodiments of the present application;

[0016] Figure 2 FIG. is a timing diagram for coordinately adjusting the operating parameters of the aeration equipment in each pond in the embodiments of the present application;

[0017] Figure 3 FIG. is a schematic flowchart for carbon footprint accounting of equipment energy consumption and dissolved gases in water bodies in the embodiments of the present application;

[0018] Figure 4 FIG. is a schematic diagram of an embodiment of the intelligent monitoring system for aquaculture based on the Internet of Things in the embodiments of the present application. Detailed implementation manners

[0019] The embodiments of the present application provide an intelligent monitoring method and system for aquaculture based on the Internet of Things. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order different from that shown or described herein. In addition, the term "including" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0020] For ease of understanding, the following describes the specific process of the embodiments of the present application. Please refer to Figure 1 , an embodiment of the intelligent monitoring method for aquaculture based on the Internet of Things in the embodiments of the present application includes:

[0021] Step S1: Collect the temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple aquaculture ponds through hierarchically arranged water quality sensors, and at the same time collect the acoustic characteristic signals of aquatic animals. Use GPRS, 4G, WiFi, and Zigbee wireless transmission methods to collect data of the parameters to obtain multi-pond ecological environment linkage data;

[0022] It can be understood that the execution entity of this application can be an intelligent monitoring system for aquaculture based on the Internet of Things, or it can also be a terminal or a server, and specific details are not limited here. In this embodiment of the application, the server is taken as the execution entity for illustration.

[0023] Specifically, the sensor network adopts a hierarchical and zonal deployment strategy, and sensor arrays are deployed at different water layer positions in the aquaculture pond. Three groups of temperature sensors, dissolved oxygen sensors, and pH value sensors are arranged in the upper layer (0.3 - 0.5 meters below the water surface), middle layer (in the middle of the water depth), and lower layer (0.3 - 0.5 meters from the bottom of the pond) of the water body respectively to achieve the collection of water layer parameters. A matrix layout is adopted at the bottom of the pond, and a set of water flow direction sensors and flow velocity sensors are arranged at each of the four corners and the central area to form a five-point control. The water flow direction sensor uses the magnetic induction principle to determine the direction by detecting the steering signal of the water flow impacting the blade; the flow velocity sensor is based on the Doppler effect and calculates the speed through ultrasonic reflection. Underwater acoustic sensors are evenly arranged around the pond to collect the acoustic characteristics of the swimming and feeding behaviors of aquatic animals. When processing temperature data, the data is grouped according to the water layer position, and the sampling frequency is 5 minutes per time. The data of multiple sensors in the same layer are arithmetically averaged to obtain the average temperature value, and the temperature difference change rate between adjacent sampling points is calculated. The dissolved oxygen data is divided into early morning (0:00 - 6:00), morning (6:00 - 12:00), afternoon (12:00 - 18:00), and night (18:00 - 24:00), and the maximum value, minimum value, and average value of each period are recorded to analyze the change trend. The pH value data is cross-compared with the temperature data to establish a corresponding relationship, and the correlation is analyzed by the Pearson correlation coefficient method. The water flow data calculates the velocity vector (including the magnitude measured by the flow velocity sensor and the direction provided by the direction sensor), and generates a flow field distribution map through two-dimensional interpolation. After the acoustic signal is denoised by band-pass filtering, the amplitude and frequency characteristics are extracted to form a behavioral acoustic fingerprint. The data transmission adopts a multi-level architecture: the sensor nodes inside the pond are networked through low-power Zigbee, and the aggregation nodes are transmitted to the monitoring center through WiFi or 4G, and GPRS is used as a backup channel. The transmitted data needs to be time-stamped and aligned, and the interpolation algorithm is used to compensate for the lost data to ensure the continuity and integrity of the data.

[0024] Step S2: According to the multi-pond ecological environment linkage data, perform Fourier transform processing on the water layer parameters and acoustic signals, extract the group behavior characteristics of aquatic animals and the water flow characteristics, and generate bio-environment coupling data;

[0025] Among them, time series segmentation is performed on the water body stratification parameters. The collected data such as temperature, dissolved oxygen, and pH value are reorganized according to the sampling time points to form a time series dataset of water body parameters. The sampling interval for each parameter is 5 minutes, and continuous recording for 24 hours constitutes a monitoring cycle. Fourier transform processing is performed on the time series data of water body parameters to convert the time domain signal into frequency domain features. Fourier transform extracts the periodic change rules of different frequency components by decomposing complex time series. Fourier transform is performed separately on parameters such as temperature and dissolved oxygen to obtain their respective spectral feature values and identify the main fluctuation periods of water quality parameters.

[0026] For the acoustic feature signal processing, it is first segmented according to the signal intensity. The continuously collected acoustic data are divided into different signal segments according to the amplitude size, and each signal segment represents a relatively independent behavior event. Fourier transform is performed on the segmented signals to obtain spectrogram data, which reflects the acoustic feature distribution of aquatic animals in different behavioral states. The frequency analysis of the spectrogram data focuses on extracting three key parameters: the movement frequency reflects the activity rhythm of aquatic animals, the activity intensity characterizes the intensity of the behavior, and the aggregation degree represents the density state of group behavior. These parameters are obtained through statistical analysis of the spectral energy distribution.

[0027] For the water flow field data analysis, by performing time series processing on the water flow direction and velocity parameters, the spatial distribution characteristics of the water body flow field are calculated. The flow direction and velocity data of 5 monitoring points are used to generate a flow field vector map of the entire pond through an interpolation algorithm to describe the water body movement state. After time alignment of the water quality fluctuation period data and the group behavior data, the correlation between the two is analyzed. By calculating the time delay and correlation coefficient of parameter changes, the influence relationship of water quality changes on the behavior of aquatic animals is determined. The spatial distribution analysis of the water body flow data determines the influence range and degree of flow field changes on biological activities.

[0028] Step S3: According to the biological-environment coupling data, calculate the water quality correlation degree and water body exchange period between multiple ponds, and coordinately adjust the operation parameters of the aeration equipment in each pond to form regional balance control data;

[0029] Among them, time series of water quality parameters are extracted from each pond. These parameters are grouped by type (temperature, dissolved oxygen, pH value) and time points to form pond water quality characteristic data. The sampling frequency for each parameter remains once every 5 minutes, and a complete 24-hour cycle is recorded. Cross-pair analysis is performed on the pond water quality characteristic data to calculate the parameter differences between adjacent ponds. The temperature difference coefficient is calculated by the difference in temperature values at the same time point between adjacent ponds, the dissolved oxygen concentration gradient reflects the unevenness of oxygen distribution, and the pH value deviation represents the spatial change of water quality acidity and alkalinity. These indicators together constitute the water quality correlation degree between ponds.

[0030] In the analysis of water body flow characteristics, spatio-temporal calculations are carried out using water flow direction and velocity data. By monitoring the flow rates of the inlet and outlet of adjacent ponds, the water body exchange volume per unit time is calculated, and the water body exchange period is determined in combination with the pond volume. At the same time, the law of flow direction change is analyzed to judge the main path of water body flow. Ponds with similar water quality parameters are divided into the same regulation area through clustering methods. Zoning is carried out according to the water quality correlation degree, and unified aeration control strategies are adopted for the ponds within the same area. The operation data of the aeration equipment in each pond record the start time, operating power, and the corresponding change in dissolved oxygen, and an equipment operation characteristic model is established.

[0031] Step S4: Based on the biological-environment coupling data and regional balance control data, by analyzing the diurnal activity patterns and group behavior patterns of aquatic animals, and combining the change trends of dissolved oxygen in each pond, calculate the feeding amount to obtain the biological rhythm feeding data;

[0032] Among them, the analysis of diurnal activity patterns is first constructed by processing the acoustic signals of aquatic animals. Extract the data of the change in acoustic signal intensity from the biological-environment coupling data, and segment it according to a 24-hour time window to form a diurnal activity signal sequence. The signal acquisition frequency is set to once per minute to record the changes in acoustic signals throughout the day. Analyze the diurnal activity signal sequence through the peak detection algorithm to identify the occurrence time and duration of the peaks in signal intensity. The peak represents the moment when the activity intensity of aquatic animals reaches the maximum, and the duration reflects the maintenance time of this activity state. Statistically analyze the peak distribution for multiple consecutive days to obtain the activity rhythm period of aquatic animals. Based on the obtained activity rhythm period, conduct segmented statistics on the acoustic characteristic signals. Calculate the total signal energy in each time period, analyze the frequency distribution characteristics, and quantify the group activity intensity. The energy distribution of acoustic signals directly reflects the intensity of activity of aquatic animals, and the frequency characteristics indicate different types of behavior patterns.

[0033] Extract the dissolved oxygen data of each pond from the regional balance control data, and calculate the change rate of dissolved oxygen over time. The fluctuation of dissolved oxygen directly reflects the oxygen consumption of aquatic animals. The change rate is obtained by calculating the difference in dissolved oxygen between adjacent time points, and then the dynamic characteristics of dissolved oxygen are determined.

[0034] Step S5: Based on the multi-pond ecological environment linkage data and regional balance control data, conduct carbon footprint accounting for equipment energy consumption and dissolved gases in the water body, optimize carbon emissions by adjusting the equipment operation timing sequence, and generate low-carbon operation data;

[0035] Among them, in the process of intelligent monitoring of aquaculture, for carbon footprint accounting of equipment energy consumption and dissolved gases in water bodies, first extract the working parameters of each water quality sensor from the multi-pond ecological environment linkage data. Statistically analyze the sampling time points and power data of each sensor, calculate the energy consumption value per unit time according to the operation duration, and obtain the basic equipment energy consumption data. Organize the operation records of the aeration equipment in the regional balance control data in time series, count the start-stop frequency and actual operation duration of the equipment in each time period, and form complete operation time series data. At the same time, record the power level and operation mode of each aeration equipment, and establish an equipment operation file. Conduct time series analysis on the dissolved oxygen data of each pond, calculate the dissolved oxygen increment per unit time by dividing the dissolved oxygen difference between adjacent time points by the operation duration of the aeration equipment, establish the corresponding relationship between the oxygenation amount and the equipment operation time, and obtain the oxygenation efficiency data. By analyzing the corresponding relationship between the operation time series data and the oxygenation efficiency data, calculate the energy consumption required to generate a unit of dissolved oxygen increment, and obtain the carbon emission intensity index of the oxygenation process. The oxygenation efficiency varies in different time periods, mainly affected by environmental factors such as water temperature and air pressure.

[0036] Step S6: According to the biological rhythm feeding data and low-carbon operation data, conduct multi-dimensional correlation analysis on the biological indicators, environmental parameters, and carbon emission data in the aquaculture process to form ecological aquaculture optimization data.

[0037] Among them, in the process of intelligent monitoring of aquaculture, process the acoustic signal characteristics of the feeding period and the dissolved oxygen fluctuation curve in the biological rhythm feeding data, and calculate the time delay relationship between the two through signal cross-correlation analysis. The time difference between the peak of the acoustic signal and the decrease in dissolved oxygen reflects the response process from bait consumption to dissolved oxygen change, and these characteristics constitute the biological response characteristic data. Extract the equipment energy consumption sequence from the low-carbon operation data, and combine it with the change rate of the dissolved gas concentration in the water body to construct a carbon-oxygen conversion dynamic relationship map. By analyzing the corresponding relationship between the energy consumption input and the dissolved oxygen output, establish an energy utilization efficiency evaluation model to obtain the ecological carbon-oxygen flow data. Conduct time-frequency analysis on the biological response characteristic data to identify the biological activity characteristics at different time scales. By comparing the activity frequency with the change law of water quality parameters, determine the influence mechanism of environmental factors on biological behavior, calculate the coupling relationship between parameters, and form ecological rhythm data.

[0038] In the embodiments of the present application, through a water quality sensor network arranged in layers, three-dimensional monitoring of water environment parameters is realized, and key indicators such as temperature, dissolved oxygen, and pH value of different water layers are accurately obtained. At the same time, the monitoring of water flow direction, flow velocity, and underwater acoustic characteristics is introduced, enriching the dimension of environmental monitoring and providing a comprehensive data basis for subsequent analysis. The Fourier transform is used to process the water body stratification parameters and acoustic signals, extract the group behavior characteristics of aquatic animals and the water body flow characteristics, establish a biological-environment coupling data model, and realize the quantitative correlation analysis between the behavior of aquaculture objects and environmental factors. Based on the biological-environment coupling data, the water quality correlation degree and water body exchange period between multiple ponds are calculated, realizing coordinated regulation at the regional level and avoiding the regional imbalance problem that may be caused by independent control of single ponds. By analyzing the daily and night activity rules and group behavior patterns of aquatic animals, and combining the change trend of dissolved oxygen in each pond to calculate the feeding amount, a precise feeding model based on biological rhythm is established, improving the feed utilization efficiency. During the operation of the equipment, by calculating the carbon footprint of energy consumption and dissolved gases in the water body, a mapping relationship between equipment operation and carbon emissions is established, realizing operation optimization oriented to low carbon. Finally, through multi-dimensional correlation analysis, biological indicators, environmental parameters, and carbon emission data are comprehensively evaluated to form ecological aquaculture optimization data, providing a scientific basis for decision-making in the aquaculture process. The Fourier transform algorithm adopted in this solution has significant advantages in processing periodic signals and is particularly suitable for analyzing the biological rhythm characteristics in aquaculture; the regional coordination algorithm based on water quality correlation considers the spatial correlation between multiple ponds and improves the accuracy of control strategies; the carbon footprint accounting model introduces environmental protection requirements into the intelligent control process, reflecting the innovation of the algorithm in specific application scenarios. Through the organic combination of these intelligent algorithms, the precise, coordinated, and low-carbon management goals of the aquaculture process are realized.

[0039] In a specific embodiment, the process of executing step S1 may specifically include the following steps:

[0040] (1) Arrange temperature sensors, dissolved oxygen sensors, and pH value sensors at different water layer positions in the aquaculture pond, arrange a water flow direction sensor and a flow velocity sensor at the bottom of the pond, and arrange underwater acoustic sensors around the pond to collect sensor data;

[0041] (2) Group the temperature data collected by the temperature sensors according to the water layer position, and calculate the average temperature value and temperature difference change rate of each water layer;

[0042] (3) Divide the dissolved oxygen data collected by the dissolved oxygen sensors according to the day and night time periods, and calculate the dissolved oxygen fluctuation range and change trend of each time period;

[0043] (4) Cross-compare the pH value data collected by the pH value sensors with the temperature data, and analyze the correlation between the pH value and the temperature change;

[0044] (5) Calculate the water body flow velocity vector and flow field distribution map in the pond from the data collected by the water flow direction sensor and the flow velocity sensor;

[0045] (6) Extract the signal amplitude and frequency characteristics from the acoustic signals collected by the underwater acoustic sensor to form the acoustic fingerprint of the aquatic animal activities;

[0046] (7) Group and transmit the temperature data, dissolved oxygen data, pH value data, water body flow velocity vector and the acoustic fingerprint of the aquatic animal activities through GPRS, 4G, WiFi, Zigbee wireless networks;

[0047] (8) Align the timestamps and compensate the data for the grouped and transmitted data to generate the multi-pond ecological environment linkage data.

[0048] Specifically, the layout of the sensor network adopts a hierarchical and zonal strategy to achieve a comprehensive perception of the water environment parameters. Temperature sensors, dissolved oxygen sensors and pH value sensors are evenly arranged in the upper layer (0.3 - 0.5 meters below the water surface), middle layer (the middle part of the water depth) and lower layer (0.3 - 0.5 meters from the bottom of the pond) of each aquaculture pond to form a vertical monitoring network. This three-dimensional layout method ensures the accurate monitoring of water quality parameters in different water layers. The "five-point layout control" mode is adopted at the bottom of the pond, that is, a set of water flow direction sensors and flow velocity sensors are arranged at each of the four corners and the center position. The water flow direction sensor works based on the magnetic induction principle and determines the water flow direction by detecting the steering signal generated by the water flow impacting the blade. The flow velocity sensor adopts the Doppler effect principle and calculates the water flow velocity by using the emission and reception of ultrasonic signals. Underwater acoustic sensors are evenly arranged around the pond to collect the acoustic signals generated by the swimming, feeding and other behaviors of aquatic animals.

[0049] For the processing of temperature data, first group the collected data according to the water layer position. Multiple sensors in each water layer synchronously sample once every 5 minutes. The collected temperature data is processed through standardization and then the average temperature value of this layer is calculated. By calculating the ratio of the temperature change amount between adjacent sampling points to the time interval, the temperature difference change rate is obtained, which is used to reflect the dynamic change characteristics of the temperature.

[0050] The processing of dissolved oxygen data uses a time period division method. The 24-hour day is divided into four time periods: early morning (0:00 - 6:00), morning (6:00 - 12:00), afternoon (12:00 - 18:00), and night (18:00 - 24:00). The maximum, minimum, and average values of dissolved oxygen are recorded within each time period, and the fluctuation range is calculated. By analyzing data over consecutive days, a trend model of dissolved oxygen changing with time is established. The cross-comparative analysis of pH value data and temperature data uses a time series pairing method. A corresponding relationship is established between the pH value data and the temperature data at the same moment, and the correlation degree between the two is quantified by calculating the correlation coefficient. The analysis result reflects the mutual influence relationship between water quality parameters.

[0051] The processing of water flow field data converts the flow direction and flow velocity data collected at each monitoring point into two-dimensional velocity vectors. Through a spatial interpolation algorithm, a flow field distribution map of the entire pond is generated based on the data of five monitoring points, visually showing the water body movement state. The processing of underwater acoustic signals includes three steps: signal preprocessing, feature extraction, and pattern recognition. The amplitude and frequency characteristics of the sound signal are extracted through spectrum analysis to form an acoustic fingerprint library for different behavioral states of aquatic animals.

[0052] All the collected data is transmitted through a multi-level wireless network. The sensor nodes inside the pond use Zigbee networking, which has the characteristics of low power consumption and flexible networking. The data aggregation nodes of each pond transmit the data to the monitoring center through 4G or WiFi networks, and GPRS is used as a backup communication method to ensure the reliability of data transmission. The time stamps of the data from different sensors are aligned. Since there are differences in the sampling frequencies and transmission delays of various sensors, a unified time reference needs to be established to ensure the timing of the data. For the data loss that occurs during the transmission process, a linear interpolation algorithm is used for data compensation to maintain the continuity of the data stream. Finally, the processed data is classified and stored according to the pond number, parameter type, and time tag, forming a complete multi-pond ecological environment linkage dataset. For example, the monitoring data of a certain aquaculture base shows that there is an obvious negative correlation between the upper water temperature and the pH value. When the water temperature rises during the day, the pH value decreases accordingly. At the same time, the dissolved oxygen reaches the daily maximum value around 10 am, which coincides with the peak activity period of aquatic animals shown by acoustic monitoring.

[0053] In a specific embodiment, the process of executing step S2 may specifically include the following steps:

[0054] (1) Perform time series segmentation on the water body stratification parameters in the multi-pond ecological environment linkage data, and reorganize the temperature, dissolved oxygen, and pH value data of different water layers according to the sampling time points to obtain water body parameter time series data;

[0055] (2) Process the time-series data of water body parameters through Fourier transform to obtain the spectral characteristic values of each parameter and get the water quality fluctuation period data;

[0056] (3) Segment the acoustic characteristic signals according to the signal intensity and perform Fourier transform processing on each segment of the signals to obtain spectrogram data;

[0057] (4) Conduct frequency analysis on the spectrogram data, extract the movement frequency, activity intensity, and aggregation degree parameters of aquatic animals to form group behavior data;

[0058] (5) Perform time-series analysis on the water flow direction and velocity parameters, calculate the water body flow field distribution, and obtain the water body flow data;

[0059] (6) Align the water quality fluctuation period data and the group behavior data in time, analyze the correlation between water quality changes and the behavior of aquatic animals, and obtain behavior correlation data;

[0060] (7) Conduct spatial distribution analysis on the water body flow data, determine the influence range of water flow on the activities of aquatic animals, and obtain the flow field influence data;

[0061] (8) Perform fusion processing on the behavior correlation data and the flow field influence data to generate bio-environment coupling data, where the bio-environment coupling data includes the group behavior characteristics of aquatic animals, the water body flow characteristics, and the data on the interaction influence relationship between the two.

[0062] Specifically, perform time series segmentation on the multi-pond ecological environment linkage data. For data standardization processing, establish a data matrix mapping formula:

[0063]

[0064] Among them, D ijt represents the standardized value of the jth parameter in the ith water layer at time t, α ijk is the weight coefficient of the parameters in different water layers, P kt is the original monitoring data, and M is the number of sampling points. Recombine the temperature, dissolved oxygen, and pH value data of the three water layers according to the sampling time to form a standard time series structure. The sampling time interval is set to 5 minutes. Each row represents a time point, and each column corresponds to the parameter values of different water layers. For each parameter, use the data of 24 consecutive hours as a processing unit.

[0065] Based on the original data, perform frequency domain analysis through weighted Fourier transform:

[0066]

[0067] Among them, W pq (f) is the spectral function, βn is the window function coefficient, x n is the time series data, γ pq is the coupling coefficient between parameters, and N is the number of points for Fourier transform. For the time series data of water quality parameters, 1024 points are selected as the transform window length, and the spectrum is calculated by the fast Fourier transform algorithm. Several frequency components with the largest amplitudes are extracted from the spectrum, and these components correspond to the main change periods of the water quality parameters.

[0068] The processing of acoustic signals adopts a segmented strategy, and an adaptive energy threshold judgment function is introduced:

[0069]

[0070] where E threshold is the energy threshold, λ is the adjustment factor, η m is the time-varying weight, S m is the amplitude of the acoustic signal, μ m is the environmental factor, and L is the signal length. When the signal energy exceeds the threshold, it is marked as a valid segment. For each valid segment, the short-time Fourier transform is used with a window length of 256 points and an overlap rate of 50%, to obtain the time-frequency distribution diagram.

[0071] The frequency analysis of spectrogram data extracts three key features: the movement frequency is obtained from the center frequency of the main frequency band, the activity intensity is calculated by integrating the frequency band energy, and the aggregation degree is determined based on spatial correlation. An acoustic feature vector is constructed:

[0072]

[0073] where F abc is the feature vector, is the feature weight, f a is the movement frequency, e b is the activity energy, c c is the aggregation coefficient, ψ i is the spatial correction factor, and I is the number of monitoring points.

[0074] The time series analysis of the water flow field is based on the real-time data of a five-point monitoring network. Through the inverse distance weighted spatial interpolation function:

[0075]

[0076] In the formula, V xy is the flow field vector at any point, ω k is the weight of the measurement point, v k is the flow velocity of the measurement point, ρ is the attenuation coefficient, d k is the spatial distance, and K is the number of measurement points. The interpolation adopts the inverse distance weighted method, and the weight coefficient is inversely proportional to the distance from the monitoring point.

[0077] The time alignment of environmental parameters and biological behaviors uses the cross - correlation function:

[0078]

[0079] Among them, R eh (τ) is the cross - correlation function, ξ t is the time weight, E t is the environmental parameter, H t is the behavior characteristic, τ is the time delay, and T is the sequence length.

[0080] Finally, multi - dimensional data fusion is carried out to construct an environment - organism coupling model:

[0081]

[0082] In the formula, C ebs is the output of the coupling model, σ q is the model weight, E q is the environmental characteristic, B q is the behavior characteristic, S q is the spatial characteristic, δ q is the coupling coefficient, Q is the feature dimension, represents the tensor product operation.

[0083] Frequency analysis of spectrogram data extracts three key features: the movement frequency is obtained from the central frequency of the main frequency band, the activity intensity is calculated by integrating the frequency band energy, and the aggregation degree is determined based on spatial correlation. These parameters constitute the feature vector describing the behavior of aquatic animals. In this data - processing process, it is necessary to normalize the acoustic data of all monitoring points to eliminate the influence of spatial position differences.

[0084] The time series analysis of the water flow field is based on the real-time data of the five-point monitoring network. The flow field distribution model is constructed by the spatial interpolation algorithm to calculate the flow velocity and flow direction at any position. The interpolation adopts the inverse distance weighted method, and the weight coefficient is inversely proportional to the distance of the monitoring point. This interpolation method reasonably estimates the flow field characteristics of the unmeasured points while maintaining the accuracy of the original data. After obtaining the water quality fluctuation cycle data and group behavior data, time alignment analysis is required. By calculating the cross-correlation function, the time delay relationship between the two sets of data is determined. This process is actually to find the best matching position of the two time series, so as to quantify the time series association between environmental changes and biological responses. The spatial distribution analysis of water flow data adopts the flow field gradient calculation method. By calculating the spatial change rate of flow velocity and flow direction, the areas with drastic flow field changes are identified. These areas are compared with the activity distribution of aquatic animals to establish a spatial correlation model between flow field characteristics and biological behavior. The behavior association data and flow field influence data are integrated in time and space to construct an environment-biological coupling model. The model includes behavioral characteristics, environmental parameters and the interactive relationship between them, providing a data basis for dynamic monitoring of the aquaculture process.

[0085] For example, during a breeding cycle, Fourier analysis showed that the water temperature had a significant 24-hour periodic feature. At the same time, acoustic monitoring recorded continuous high-intensity signals between 5 and 7 in the morning, and spectrum analysis showed that the main frequency band was concentrated in 800-1200 Hz, indicating that this was the active period of aquatic animals. Water flow monitoring found that a stable circulation structure had formed in the pond, with a mainstream velocity in the range of 0.1-0.2 m / s. Through data association analysis, it was found that there was a time delay of about 2-3 hours between the activity intensity of aquatic animals and the decrease in dissolved oxygen, and the activity area was mainly distributed in the moderate flow velocity area at the edge of the circulation.

[0086] In a specific embodiment, the process of executing step S3 may specifically include the following steps:

[0087] (1) Extract the time series of water quality parameters of each pond from the biological-environmental coupling data, group them according to parameter type and time point, and obtain pond water quality characteristic data;

[0088] (2) Cross-matching the water quality characteristic data of each pond, calculating the temperature difference coefficient, dissolved oxygen concentration gradient, and pH deviation between the paired ponds, and obtaining the water quality correlation between ponds;

[0089] (3) Perform spatiotemporal analysis on the flow direction and velocity data of each pond, calculate the water flow flux and flow direction change patterns between adjacent ponds, and obtain the water exchange cycle;

[0090] (4) The ponds are divided and clustered according to the water quality correlation, and ponds with similar water quality parameters are divided into the same regulation area to obtain a regional coordinated control plan;

[0091] (5) Based on the regional collaborative control scheme, analyze the operation data of the aeration equipment in each pond, extract the corresponding relationship between the equipment start time, operating power, and aeration effect, and obtain the equipment operation characteristics;

[0092] (6) Perform correlation calculations on the water quality correlation degree, water body exchange cycle, and equipment operation characteristics, allocate operation time periods and power parameters for the aeration equipment in each regulation area, and obtain the equipment collaboration scheme;

[0093] (7) Evaluate the operation effect of the equipment collaboration scheme, calculate the uniformity of dissolved oxygen distribution and energy consumption distribution within the area, and generate regional balance control data.

[0094] Specifically, as Figure 2 shown, it is the timing diagram for collaborative adjustment of the operation parameters of the aeration equipment in each pond in the embodiment of the present application, showing the complete process from data collection to regional balance control. The timing diagram includes five modules: a sensor network, a data processor, an analyzer, a controller, and an aeration equipment. Among them, the sensor network collects water quality parameter data every 5 minutes, and the data processor performs median filtering on the collected data; the analyzer calculates the difference index between ponds and the water quality correlation degree based on a 1-hour sliding window, and calculates the water body exchange flux every 30 minutes; the controller reads the equipment operation data every 1 minute, performs performance evaluation every 30 minutes, updates the optimization calculation results every 15 minutes, and updates the evaluation indicators every 1 hour. The entire process realizes the closed-loop management of data collection, analysis and processing, and control execution.

[0095] Extract the water quality parameter time series from the bio-environment coupling data, and set the sampling frequency to once every 5 minutes. Align the original data of temperature, dissolved oxygen, and pH value according to the time stamp, eliminate the sampling time error, and form a time series data set. The original data recorded by the acquisition equipment enters the database after digital filtering to eliminate noise, and at the same time, meta-information such as the data acquisition time and sensor number is recorded.

[0096] For the calculation of the water quality correlation degree between ponds, use the sliding window method to process the time series data. The window length is set to 1 hour, that is, 12 sampling points, and calculate the parameter differences between adjacent ponds within the window. The temperature difference coefficient is obtained by dividing the difference between the temperature values of two ponds at the same moment by the average temperature value, the dissolved oxygen concentration gradient is calculated by the concentration difference standardized by the spatial distance, and the pH value deviation is directly taken as the difference between the two values. In actual operations, there is a valid value range for the difference calculation of each parameter, and data outside the range will be marked as outliers for special processing.

[0097] The analysis of water body flow characteristics adopts the following exchange flux model:

[0098]

[0099] Among them, Q ij is the water body exchange flux between adjacent ponds i and j, and r n is the flow field weight coefficient (taking values from 0 to 1), and u n (x, y, t) is the time-varying velocity field (m / s), and g n (x, y) is the spatial position function (dimensionless), and A ij is the junction area (m 2 ), and N is the number of monitoring points. The model calculation uses a 30-minute cycle to accumulate 24-hour data to determine the water body exchange law. The calculation of the exchange flux requires real-time velocity and flow direction data support, and these data are provided by the velocity sensors deployed at the pond boundaries.

[0100] The analysis of the operating characteristics of the aeration equipment uses the following performance evaluation function:

[0101]

[0102] In the formula, P aer is the aeration efficiency index, and m k is the equipment type coefficient (dimensionless), and w k (t) is the time-varying power parameter (kW), and h k is the equipment health (0 - 1), and y k (DO) is the dissolved oxygen response function (mg / L), and K is the number of equipment. The data acquisition interval is 1 minute, and the evaluation period is 30 minutes. The operating data of each aeration equipment includes information such as the start time, operating power, and equipment status, and these data are collected through the equipment controller and uploaded in real time.

[0103] The equipment collaborative optimization adopts the following objective function:

[0104]

[0105] Here, Z opt is the collaborative optimization objective (dimensionless), and θ l is the regional weight (0 - 1), and q l (t) is the time period allocation coefficient (0 - 1), and j l (x, y) is the spatial distribution function (dimensionless), and z l (DO) is the dissolved oxygen control target (mg / L), and L is the number of control regions. The optimization calculation is carried out at 15-minute intervals to roll and update the control parameters. The calculation result of the objective function is directly used to guide the operation scheduling of the aeration equipment, including the start and stop times and power setting.

[0106] The effect evaluation of the device collaboration solution calculates the uniformity of dissolved oxygen distribution using the standard deviation method. The dissolved oxygen data of all monitoring points in the area are statistically analyzed to calculate the mean value and the standard deviation. The energy consumption distribution is calculated by recording the cumulative operation time and average power of each device, and then calculating the energy consumption per unit time. These evaluation indicators are updated every hour and used as the basis for adjusting the operation plan.

[0107] For example, in the actual monitoring of a group of adjacent aquaculture ponds, the water quality data of two ponds for 24 hours were recorded through continuous sampling at 5-minute intervals. After removing the mutation points from the original data of the temperature sensor through median filtering, the temperature difference coefficient was calculated. Similarly, the dissolved oxygen and pH value data were processed to obtain the difference indicators of the three parameters. The water flow monitoring data showed that there was a stable water body exchange phenomenon between the two ponds, and the exchange period was calculated to be about 4 hours according to the exchange flux model. Based on the water quality correlation index, these two ponds were divided into the same regulation area. In terms of device collaborative control, the operation characteristics of the aeration device were analyzed through the efficiency evaluation function, and the optimal operation timing was calculated in combination with the optimization objective function. The evaluation results showed that after adopting the collaborative control solution, the standard deviation of dissolved oxygen in the area decreased, indicating that the dissolved oxygen distribution was more uniform, and at the same time, the energy consumption distribution of the devices also tended to be reasonable.

[0108] In a specific embodiment, the process of executing step S4 may specifically include the following steps:

[0109] (1) Extract the data of the change in the acoustic signal intensity of aquatic animals from the bio-environment coupling data, and segment it according to a 24-hour time window to obtain the day-night activity signal sequence;

[0110] (2) Perform peak detection on the day-night activity signal sequence, calculate the time interval and duration when the signal intensity peak appears, and obtain the activity rhythm period;

[0111] (3) Based on the activity rhythm period, perform segmented statistics on the acoustic feature signals, calculate the total signal energy and frequency distribution in each time period, and obtain the population activity intensity;

[0112] (4) Extract the dissolved oxygen data of each pond from the regional balance control data, calculate the change rate and fluctuation range of the dissolved oxygen concentration over time, and obtain the dynamic characteristics of dissolved oxygen;

[0113] (5) Make a time correspondence between the population activity intensity and the dynamic characteristics of dissolved oxygen, calculate the correlation coefficient and time lag relationship between the two, and obtain the activity oxygen consumption law;

[0114] (6) Conduct a comprehensive analysis of the activity oxygen consumption law and the population activity intensity, calculate the oxygen consumption per unit biomass in each time period, and obtain the feed coefficient;

[0115] (7) Combine the feed conversion ratio with the activity rhythm period, calculate the feeding amount according to different time periods, and generate the biological rhythm feeding data.

[0116] Specifically, extract the data on the change in the acoustic signal intensity of aquatic animals. Using a sampling frequency of once per minute, record the original signal collected by the underwater acoustic sensor. Take the 24-hour continuous data as a processing window. First, perform band-pass filtering on the original signal to filter out the noise components below 50 Hz and above 2000 Hz, and retain the effective acoustic signal generated by the activities of aquatic animals.

[0117] Perform peak detection on the processed day-night activity signal sequence, and use the moving average threshold method to identify signal peaks. Set a 60-minute moving window, calculate 1.5 times the mean value of the signal within the window as the dynamic threshold, and mark it as a peak point when the signal intensity exceeds the threshold. Record the timestamp and duration of each peak point, and obtain the activity rhythm period by statistically analyzing the time interval between adjacent peak points. Based on the obtained activity rhythm period, segment the acoustic feature signal. Calculate the total energy for the signal data in each time period, and at the same time perform spectral analysis to obtain the frequency distribution characteristics. The total energy is calculated by the sum of the squares of the signal amplitudes, and the frequency distribution uses the short-time Fourier transform method. Set a 256-point analysis window to obtain the group activity intensity index.

[0118] The correlation analysis between the group activity intensity and the dynamic characteristics of dissolved oxygen uses the following model:

[0119]

[0120] where R acs is the activity-oxygen consumption correlation index, χ i is the time period weight, b i (t) is the behavior intensity, d i (t) is the dissolved oxygen change rate, λ is the time delay, ψ i is the environmental factor, represents the cross-correlation operation, and I is the number of analysis periods.

[0121] The calculation model for the oxygen consumption per unit biomass is as follows:

[0122]

[0123] where C_{oxy} is the oxygen consumption coefficient, v p is the biomass coefficient, f p (a) is the activity intensity function, g p (o) is the oxygen consumption function, ∈ p is the efficiency factor, and P is the parameter dimension.

[0124] The biological rhythm feeding model is defined as:

[0125]

[0126] In the formula, F feed is the feeding amount, l n is the cycle coefficient, c n (t) is the time distribution function, m n (s) is the specification coefficient, κ n is the environmental correction factor, and N is the number of control parameters.

[0127] For example, in the actual aquaculture process, by analyzing the monitoring data of a certain aquaculture pond for 24 hours, the acoustic signal shows an obvious bimodal distribution characteristic. The first activity peak appears at 5 - 7 in the early morning, and the second peak appears at 17 - 19 in the evening. The signal intensity in these two time periods is significantly higher than that in other time periods. The corresponding dissolved oxygen data shows that the dissolved oxygen concentration begins to decrease significantly 2 - 3 hours after the activity peak, and the time lag relationship is obtained through calculation. Based on this correspondence, combined with the calculation result of the oxygen consumption per unit biomass, it is determined that the feeding times in the morning and evening are 6 o'clock and 18 o'clock respectively, and the distribution ratio of the feeding amount is determined according to the activity intensity.

[0128] In a specific embodiment, the process of executing step S5 may specifically include the following steps:

[0129] (1) Extract the sampling time points and power data of each water quality sensor from the multi - pond ecological environment linkage data, calculate the energy consumption value per unit time of each device, and obtain the device energy consumption data;

[0130] (2) Extract the operation time series of the aeration equipment from the regional balance control data, calculate the start - stop frequency and operation duration of the equipment in each time period, and obtain the operation time series data;

[0131] (3) Perform time series segmentation on the dissolved oxygen concentration data of each pond, calculate the corresponding relationship between the change in dissolved oxygen and the operation duration of the aeration equipment, and obtain the aeration efficiency data;

[0132] (4) Conduct cross - analysis on the operation time series data and the aeration efficiency data, calculate the energy consumption corresponding to the unit oxygen increase, and obtain the carbon emission intensity;

[0133] (5) Group and statistically analyze the device energy consumption data according to time periods, calculate the total energy consumption and carbon emissions in each time period, and obtain the carbon footprint data;

[0134] (6) Rearrange the device operation time series based on the carbon footprint data, and transfer the operation tasks in the high - energy - consumption time period to the low - energy - consumption time period to obtain the time series optimization scheme;

[0135] (7) Combine the timing optimization plan with the carbon footprint data, calculate the optimized energy consumption distribution and carbon emissions, and generate low-carbon operation data.

[0136] Specifically, as Figure 3 shown, it is a schematic flowchart of carbon footprint accounting for equipment energy consumption and dissolved gases in water bodies in the embodiments of the present application, mainly including three parallel processing branches: The first branch is the process from sensor data collection to the generation of equipment energy consumption data; the second branch is the process from dissolved oxygen data collection to the calculation of aeration efficiency data; the third branch is the process from data combination analysis to the generation of low-carbon operation data. Different colors are used in the figure to mark data nodes and processing nodes. The data nodes are filled with light red, and the processing nodes are filled with light green. The arrows indicate the data flow direction. The flowchart shows the complete process of data collection, processing, analysis, and optimization, and finally outputs low-carbon operation data.

[0137] Extract the sampling time points and power data of each water quality sensor from the multi-pond ecological environment linkage data. These data include basic parameters such as the model, location information, sampling frequency, working voltage, and current of each sensor. By analyzing the sensor sampling time series, a sampling time point matrix is established. Each matrix element contains three dimensions: timestamp, sensor number, and power value. The energy consumption value per unit time of each device is calculated by multiplying the power value by the sampling time interval. For example, for a dissolved oxygen sensor, the sampling time interval is 1 minute, the working voltage is 12V, and the working current is 100mA, then the energy consumption per unit time is 1.2W·h.

[0138] When extracting the operation time series of the aeration equipment from the regional balance control data, first preprocess the original operation data, including removing outliers, filling in missing values, etc. Then segment the data according to a preset time window (such as 1 hour), and count the number of start times, stop times, and cumulative operation duration of the equipment in each time period. The operation state of the equipment is collected in real time through the current detection module. When the detected current value exceeds the start threshold, it is recorded as a start event, and when it is lower than the stop threshold, it is recorded as a stop event. When performing time series segmentation on the pond dissolved oxygen concentration data, the sliding window method is used for data segmentation, and the window size is determined according to the typical operation cycle of the aeration equipment. The corresponding relationship between the change in dissolved oxygen and the operation duration of the aeration equipment is calculated by the following formula:

[0139]

[0140] Among them, E oxy represents the aeration efficiency coefficient, D i represents the dissolved oxygen concentration (mg / L) at the i-th time point, W i represents the water body diffusion coefficient, V p represents the pond volume (m3 ),T i represents the operating duration (h) of the aeration equipment, P i represents the equipment power (kW), and n represents the number of sampling points.

[0141] When calculating the energy consumption corresponding to the unit oxygen increment, the cross-analysis method is adopted, and the carbon emission intensity is determined through the following formula:

[0142]

[0143] Among them, C int represents the carbon emission intensity coefficient, F j represents the energy consumption (kWh) in the j-th time period, K j represents the energy conversion coefficient, Q j represents the meteorological correction coefficient, O j represents the dissolved oxygen increment (mg / L), R_j represents the water body utilization coefficient, and m represents the number of statistical time periods.

[0144] When conducting time period grouping statistics on the equipment energy consumption data, the cumulative quantity calculation method is adopted, and the carbon footprint is calculated through the following formula:

[0145]

[0146] Among them, H crb represents the total carbon footprint, M k represents the operating duration (h) of the equipment in the k-th time period, U k represents the energy consumption per unit time (kWh), Y k represents the carbon emission factor, Z k represents the equipment efficiency coefficient, and l represents the total number of statistical time periods.

[0147] When rearranging the equipment operation timing based on the carbon footprint data, high-energy-consuming time periods and low-energy-consuming time periods are first identified. High-energy-consuming time periods usually occur in time periods with high dissolved oxygen demand and high water temperature, such as noon. Through the timing analysis method, some operation tasks in these time periods are transferred to low-energy-consuming time periods such as early morning, while ensuring that the transferred operation plan can still meet the dissolved oxygen demand of aquaculture. The biological rhythm characteristics of aquatic animals need to be considered during the timing optimization process to avoid causing stress responses to the cultured objects due to the adjustment of the operation timing.

[0148] When generating low-carbon operation data, the optimized time series plan is compared and analyzed with the original carbon footprint data. By calculating parameters such as the operating power, duration, start-stop frequency, etc. of the equipment before and after optimization, the energy consumption distribution curve for each time period is obtained. At the same time, combined with water quality monitoring data, the impact of the optimization plan on the aquaculture environment is verified to ensure the stability of water quality while reducing carbon emissions. During the whole process, all data processing and calculations are executed by the intelligent controller, and the data is transmitted and stored in real time through the Internet of Things platform to ensure the accuracy and timeliness of the data.

[0149] For example, taking the optimization of the aeration equipment in a certain farm as an example, in the original operation plan, 4 aerators were simultaneously turned on during the high-temperature period from 2 pm to 4 pm, and the energy consumption per unit time reached 8 kWh. Through data analysis, it was found that the change in dissolved oxygen concentration during this period was not proportional to the equipment operation duration, and the aeration efficiency was low. After time series optimization, the operation time of 2 aerators was adjusted to 3 am to 5 am, when the water temperature was lower and the aeration efficiency was improved. Recalculating the energy consumption data showed that, while maintaining the same dissolved oxygen level, the daily average energy consumption was reduced from the original 96 kWh to 78 kWh. The carbon footprint calculation results showed that the carbon emissions after optimization were correspondingly reduced. By continuously monitoring water quality parameters and the activity status of aquatic animals, it was confirmed that the time series optimization plan had no adverse impact on the aquaculture environment.

[0150] In a specific embodiment, the process of executing step S6 may specifically include the following steps:

[0151] (1) Extract the acoustic signal characteristics of the feeding period and the dissolved oxygen fluctuation curve from the biological rhythm feeding data, calculate the time delay relationship between the two through signal cross-correlation analysis, and obtain the biological response characteristic data;

[0152] (2) Extract the equipment energy consumption sequence from the low-carbon operation data, combine it with the change rate of the dissolved gas concentration in the water body, and construct a dynamic carbon-oxygen conversion relationship map to obtain the ecological carbon-oxygen flow data;

[0153] (3) Perform time-frequency analysis on the biological response characteristic data, extract the biological activity cycle and intensity change rules at different time scales, calculate the coupling relationship with water quality parameters, and obtain the ecological rhythm data;

[0154] (4) Classify the ecological carbon-oxygen flow data according to the carbon-oxygen conversion efficiency, identify the optimal operation interval within the key time window, and obtain the energy efficiency balance data;

[0155] (5) Based on the ecological rhythm data and the energy efficiency balance data, establish a carbon efficiency growth evaluation system by calculating the synergy index of biological activity and energy utilization, and obtain the ecological efficiency data;

[0156] (6) Analyze the spatio-temporal distribution of ecological efficiency data, identify the periods and regions of efficient aquaculture, calculate the optimal aquaculture density in combination with the environmental carrying capacity, and generate optimized ecological aquaculture data.

[0157] Specifically, data processing is carried out for the acoustic signal characteristics during the feeding period. The acoustic signal characteristics refer to the acoustic wave signals generated by the activities of aquatic animals collected by underwater acoustic sensors, and these signals are obtained through the acquisition of an underwater acoustic sensor array. The acoustic signals are preprocessed by a band-pass filter to filter out environmental noise and equipment vibration signals, obtaining pure biological activity acoustic signals. The processing of the dissolved oxygen fluctuation curve uses time series analysis methods. The continuously collected dissolved oxygen data are segmented according to the feeding period, and each period contains data for three stages: before feeding, during feeding, and after feeding. Signal cross-correlation analysis is a mathematical method for calculating the similarity and time delay relationship between two signals. By aligning the time series of acoustic signals and dissolved oxygen data, the correlation coefficients at different time offsets are calculated, and the time offset corresponding to the maximum correlation coefficient is found, which is the time delay relationship between the two. When extracting the equipment energy consumption sequence from the low-carbon operation data, the operation data of various energy-consuming equipment such as aeration equipment and feeding equipment need to be sorted in chronological order. The calculation of the change rate of the concentration of dissolved gases in the water body is based on the real-time monitoring data of the dissolved oxygen sensor, and the change amount of the dissolved oxygen concentration per unit time is obtained through differential calculation. The carbon-oxygen conversion dynamic relationship map is a data model describing the relationship between energy consumption and dissolved oxygen changes. This model maps time series data onto a two-dimensional plane, with the horizontal axis representing the energy consumption value and the vertical axis representing the change rate of dissolved oxygen. The distribution of data points reflects the energy utilization efficiency.

[0158] When performing time-frequency analysis on the biological response characteristic data, the wavelet transform method is used to perform multi-scale decomposition on the acoustic signals to extract the energy distribution characteristics of different frequency bands. The biological activity cycle and intensity change law are calculated by the following formula:

[0159]

[0160] where B rhy represents the biological rhythm index, α x represents the activity intensity coefficient at the x-th time scale, S x represents the acoustic signal energy value, β x represents the time weight factor, γ x represents the environmental correction coefficient, θ x represents the behavior cycle coefficient, λ x represents the water quality coupling factor, and g represents the number of time scales.

[0161] When grading ecological carbon-oxygen flow data according to carbon-oxygen conversion efficiency, it is necessary to first establish an evaluation index system. Carbon-oxygen conversion efficiency refers to the dissolved oxygen increment generated by unit energy consumption, which is obtained by calculating the ratio of energy consumption to dissolved oxygen change in different time windows. The identification of the optimal operating range is based on the statistical analysis of historical data, dividing the carbon-oxygen conversion efficiency into multiple levels according to the numerical value, and combining the changing trend of water quality parameters to determine the operating efficiency level of each time window.

[0162] Based on the ecological rhythm data and energy efficiency balance data, the synergy index between biological activity and energy utilization was calculated by the following formula:

[0163]

[0164] Among them, A eco represents the ecological synergy index, represents the bioactivity index of the vth period, ψ v represents the energy efficiency score, ω v represents the growth contribution rate, ξ v represents energy consumption, η v represents the basal metabolic coefficient, μ v represents environmental adaptability, and h represents the number of evaluation cycles.

[0165] The spatiotemporal distribution analysis of ecological efficiency data uses a gridding method to divide the aquaculture area into several evaluation units, each of which contains water quality parameters, biological activity indicators and energy efficiency data. The identification of efficient aquaculture periods and areas is based on cluster analysis of multidimensional data, and spatiotemporal units with similar characteristics are grouped together. Environmental carrying capacity refers to the maximum aquaculture density that aquaculture water can withstand without destroying the ecological balance. It is calculated comprehensively based on factors such as water quality parameters, dissolved oxygen supply capacity and biological oxygen consumption.

[0166] Taking a certain aquaculture pond as an example, the activity signals of aquatic animals are collected by acoustic sensors arranged at different water layers, and the monitoring data of dissolved oxygen sensors are recorded simultaneously. After band-pass filtering the collected acoustic signals, the effective signals with a frequency range of 20 Hz - 2 kHz are extracted. The cross-correlation function of the acoustic signals and the dissolved oxygen data is calculated, and it is found that the correlation peak appears about 15 minutes after feeding, indicating that there is an obvious time-delay characteristic in the response of aquatic animals to feeding. Through wavelet analysis of the acoustic signals, multiple characteristic frequency bands are identified, and the energy change in the 100 - 200 Hz frequency band is highly correlated with feeding activities. Combining the change trend of dissolved oxygen, a quantitative relationship model between the feeding amount and biological activity is established. On this basis, by adjusting the operation timing of the aeration equipment to match the activity pattern of aquatic animals, the energy utilization efficiency is improved. The spatio-temporal distribution analysis shows that in the water layer with stable water temperature and sufficient dissolved oxygen, the activity intensity of aquatic animals is higher, and the corresponding allowable value of aquaculture density is also larger. This precise management method based on biological behavior and environmental parameters effectively realizes the energy conservation and emission reduction goals in the aquaculture process.

[0167] In a specific embodiment, the process of performing the step of grading the ecological carbon-oxygen flow data according to the carbon-oxygen conversion efficiency may specifically include the following steps:

[0168] (1) Perform phase decomposition on the ecological carbon-oxygen flow data to separate the instantaneous frequency characteristics of carbon emission fluctuations and oxygen conversion, and obtain a carbon-oxygen coupling sequence;

[0169] (2) Perform non-linear dynamic analysis on the carbon-oxygen coupling sequence, calculate the mutation points and steady-state intervals in the carbon-oxygen conversion process, and obtain conversion characteristic data;

[0170] (3) Perform balance calculation on the conversion characteristic data according to the principle of energy conservation, establish a quantitative relationship between carbon emissions and dissolved oxygen production, and obtain energy balance data;

[0171] (4) Construct a time window matrix based on the energy balance data, extract the peak and valley values of the energy conversion efficiency within the window, and obtain critical threshold data;

[0172] (5) Perform adaptive segmentation on the critical threshold data to identify the optimal working point and saturation point of the carbon-oxygen conversion efficiency, and obtain operation boundary data;

[0173] (6) Perform dynamic compensation calculation on the operation boundary data to balance the carbon emission intensity and the dissolved oxygen supply rate, and generate energy efficiency balance data.

[0174] Specifically, in the intelligent monitoring method for aquaculture based on the Internet of Things, the phase decomposition of ecological carbon-oxygen flow data is achieved through the Hilbert-Huang Transform (HHT) method. Phase decomposition is a signal processing technique that decomposes complex time-series signals into different frequency components. First, the original carbon emission data and dissolved oxygen data are preprocessed, including denoising and normalization. Then, using the Empirical Mode Decomposition (EMD) algorithm, the signal is decomposed into several Intrinsic Mode Functions (IMFs), and each IMF represents the fluctuation characteristics at different time scales. By performing the Hilbert transform on each IMF, the instantaneous frequency and instantaneous amplitude are obtained, thereby extracting the frequency characteristics of carbon emission fluctuations and oxygen conversion. When performing nonlinear dynamic analysis on the obtained carbon-oxygen coupling sequence, the phase space reconstruction technique is adopted. First, the time delay and embedding dimension are determined to construct the phase space trajectory. The time delay is determined by the mutual information method, and the embedding dimension is calculated by the false nearest neighbor method. In the reconstructed phase space, the local linear fitting method is used to identify the mutation points, that is, the positions where the trajectory changes significantly. The steady-state interval is the time period when the trajectory is relatively stable, and it is judged by calculating the local Lyapunov exponent of the trajectory.

[0175] The energy balance calculation of the conversion characteristic data adopts the following formula:

[0176]

[0177] Among them, G bal represents the energy balance factor, σ p represents the carbon emission intensity in the p-th period, ρ p represents the dissolved oxygen conversion rate, τ p represents the energy utilization coefficient, ε p represents the water body diffusion coefficient, χ p represents the temperature correction factor, and q represents the number of calculation cycles.

[0178] The construction of the time window matrix adopts the sliding window method, and the window size is determined according to the actual aquaculture cycle. Calculate the energy conversion efficiency for the data within each window, including the ratio of the input energy (carbon emission) to the output energy (dissolved oxygen increment). Identify the peak and valley values of the efficiency through statistical analysis and establish a critical threshold system.

[0179] The adaptive segmentation of the critical threshold data adopts the dynamic programming algorithm. The time series is divided into multiple intervals according to the change trend of the efficiency value, and the boundary points of each interval are potential working points. By calculating the average efficiency and variance within each interval and combining the constraint conditions of the environmental parameters, the positions of the optimal working point and the saturation point are determined.

[0180] The dynamic compensation calculation of the operating boundary data adopts the following formula:

[0181]

[0182] Among them, L opt represents the operation optimization coefficient, ν r represents the carbon emission compensation factor in the r-th period, π r represents the dissolved oxygen supply efficiency, Δ r represents the environmental correction coefficient, ζ r represents the energy consumption benchmark value, φ r represents the operation duration coefficient, ι r represents the season adjustment factor, and s represents the number of optimization cycles.

[0183] Taking the oxygenation process of a certain aquaculture pond as an example to illustrate the data processing flow: First, collect 24-hour operation data, including the energy consumption data of the oxygenation equipment and the monitoring data of the dissolved oxygen sensor. Through phase decomposition, it is found that the energy consumption data shows obvious periodic fluctuations, and the main frequency components are concentrated in two intervals of 2 - 4 hours and 8 - 12 hours, corresponding to the operation cycle of the equipment and the diurnal variation respectively. The main frequency component of the dissolved oxygen data shows a cycle of about 6 hours, reflecting the physiological rhythm of aquatic animals. In the non-linear dynamic analysis, a 6-dimensional phase space is selected for trajectory reconstruction, and the time delay is taken as 20 minutes. By analyzing the reconstructed trajectory, 4 main mutation points are identified, corresponding to the operation state transitions in the early morning, morning, afternoon, and evening periods respectively. Between these mutation points, there are 3 relatively stable operation intervals, and the interval from 2 am to 6 am shows the highest energy conversion efficiency.

[0184] Based on the energy balance calculation results, a 3-hour sliding window is used to analyze the data, and it is found that the peak of the energy conversion efficiency appears in the period with lower water temperature and relatively stable dissolved oxygen demand. Through the dynamic programming algorithm, the 24-hour period is divided into 6 operation intervals, and 2 optimal working points and 1 saturation point are identified. The periods corresponding to the optimal working points become the key regulation objects, and through dynamic compensation calculation, the equipment operation parameters for different periods are determined, realizing the balanced optimization of carbon emission intensity and dissolved oxygen supply.

[0185] In a specific embodiment, the process of performing the step of spatio-temporal distribution analysis on the ecological efficiency data may specifically include the following steps:

[0186] (1) Perform wavelet decomposition on the ecological efficiency data, separate the time-varying characteristics of different frequency bands, extract the fluctuation periods of biological activity and energy utilization, and obtain multi-scale efficiency data;

[0187] (2) Perform interpolation calculation on the multi-scale efficiency data according to the spatial coordinates, construct the three-dimensional efficiency field distribution of the aquaculture area, and obtain the efficiency field data;

[0188] (3) Perform gradient analysis on the efficiency field data, identify the change rate and propagation direction of the efficiency field, calculate the efficiency flow relationship between regions, and obtain the efficiency flow data;

[0189] (4) Quantify the water body diffusion characteristics based on the efficiency flow data, and combine with the migration laws of dissolved oxygen and nutrients to obtain the material cycle data;

[0190] (5) Perform correlation calculation between the material cycle data and the dissolved oxygen consumption rate, construct the dynamic balance relationship between the water body self-purification ability and the aquaculture load, and obtain the bearing threshold data;

[0191] (6) Perform spatio-temporal optimization on the bearing threshold data, calculate the maximum aquaculture capacity in different regions and time periods, and generate the ecological aquaculture optimization data.

[0192] It should be noted that wavelet decomposition is a time-frequency analysis method. By selecting an appropriate wavelet basis function, the original signal is decomposed into sub-signals of different frequency bands. The ecological efficiency data includes two dimensions: biological activity data and energy utilization data. The biological activity data comes from the biological behavior signals collected by underwater acoustic sensors, and the energy utilization data comes from the operating parameters of various devices. During the wavelet decomposition process, the original data is first preprocessed, including denoising and normalization, and then the db4 wavelet basis function is selected for multi-scale decomposition to obtain wavelet coefficients reflecting the characteristics of different time scales. Energy statistics are performed on the wavelet coefficients of each scale, and significant periodic fluctuation characteristics are extracted. The spatial interpolation calculation of multi-scale efficiency data uses the Kriging interpolation method, which is an optimal linear unbiased estimation method based on the variogram theory. First, multiple monitoring points are arranged in the aquaculture area, and each point contains three-dimensional coordinate information and efficiency index values. By calculating the spatial correlation between sampling points, a variogram model is established, and then the efficiency index value at any position is calculated. The construction of the three-dimensional efficiency field distribution requires data interpolation in the vertical and horizontal directions to form a continuous efficiency distribution field.

[0193] When performing gradient analysis on the efficiency field data, the central difference method is used to calculate the spatial gradient. In three-dimensional space, the efficiency change rates in the x, y, and z directions are calculated respectively to obtain the gradient vector field. The modulus of the gradient vector represents the severity of the efficiency change, and the direction indicates the dominant direction of efficiency propagation. By calculating the efficiency difference and flow direction between adjacent regions, a network of efficiency flow relationships between regions is established. During the quantification process of water body diffusion characteristics, the tracer experiment method is used. Tracer substances are released into the water body, and by monitoring their diffusion process, the diffusion coefficient and migration rate are obtained. Combining the real-time monitoring data of dissolved oxygen sensors and water quality sensors, the migration laws of dissolved oxygen and nutrients in the water body are analyzed. The material cycle data includes information such as the concentration distribution, migration rate, and transformation relationship of various substances.

[0194] The associated calculation of material cycle data and the dissolved oxygen consumption rate adopts the following formula:

[0195]

[0196] Among them, J cap represents the water body carrying capacity index, κ u represents the dissolved oxygen consumption rate of the u-th area, υ u represents the water body renewal coefficient, ψ u represents the nutrient conversion rate, represents the biomass density, ω u represents the self-purification coefficient, represents the environmental stress factor, and w represents the number of area divisions.

[0197] The spatio-temporal optimization of the carrying threshold data adopts the multi-objective programming method. First, an optimization objective function is established, including two dimensions: maximizing aquaculture production and minimizing environmental impact. The constraint conditions include water quality parameter limits, energy consumption upper limits, equipment operation capabilities, etc. By solving the optimization model, the maximum aquaculture capacity of different regions and time periods can be obtained.

[0198] Taking a multi-pond aquaculture area as an example, ecological efficiency data is collected through the deployed sensor network. The wavelet decomposition results show that there are significant periodic fluctuations in the bioactivity data at the 6-hour and 24-hour scales, corresponding to the feeding activity cycle and the circadian rhythm respectively. The energy utilization data shows the main periods of 4 hours and 12 hours, reflecting the equipment operation rules. Through Kriging interpolation calculation, a three-dimensional efficiency field distribution map covering the entire aquaculture area is established, clearly showing the spatial distribution characteristics of the efficiency. The gradient analysis results show that the efficiency field shows obvious stratification characteristics in the vertical direction, and local circulation is formed in the horizontal direction under the influence of water flow and the layout of aeration equipment. The tracer experiment data shows that the diffusion rate of dissolved oxygen increases with the increase of water temperature, while the migration of nutrients is mainly affected by the water flow field. Through correlation analysis, a balance model between the water body self-purification ability and the aquaculture load is established, and the maximum aquaculture capacity of different water layers and regions is calculated. The final ecological aquaculture optimization data guides farmers to reasonably adjust the aquaculture density and layout, achieving the dual goals of aquaculture benefits and environmental protection.

[0199] The above describes the Internet of Things-based intelligent monitoring method for aquaculture in the embodiments of the present application. Next, the Internet of Things-based intelligent monitoring system for aquaculture in the embodiments of the present application will be described. Please refer to Figure 4 One embodiment of the Internet of Things-based intelligent monitoring system for aquaculture in the embodiments of the present application includes:

[0200] The acquisition module is used to collect the temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple breeding ponds through layered water quality sensors, and collect acoustic characteristic signals of aquatic animals. The parameters are collected using GPRS, 4G, WiFi, and Zigbee wireless transmission methods to obtain multi-pond ecological environment linkage data;

[0201] A transformation module is used to perform Fourier transformation processing on water body stratification parameters and acoustic signals based on the multi-pond ecological environment linkage data, extract the behavior characteristics of aquatic animal groups and water body flow characteristics, and generate biological-environment coupling data;

[0202] A regulation module, for calculating the water quality correlation and water exchange period between multiple ponds according to the biological-environmental coupling data, and coordinating the operating parameters of the aeration equipment of each pond to form regional balance control data;

[0203] A calculation module is used to calculate the feeding amount based on the biological-environment coupling data and the regional balance control data, by analyzing the diurnal activity patterns and group behavior patterns of aquatic animals and combining the dissolved oxygen change trend of each pond to obtain the biorhythm feeding data;

[0204] A generation module is used to calculate the carbon footprint of equipment energy consumption and dissolved gas in water according to the multi-pond ecological environment linkage data and regional balance control data, optimize carbon emissions by adjusting the equipment operation timing, and generate low-carbon operation data;

[0205] The correlation module is used to perform multi-dimensional correlation analysis on biological indicators, environmental parameters and carbon emission data in the breeding process according to the biorhythm feeding data and low-carbon operation data to form ecological breeding optimization data.

[0206] Through the collaborative cooperation of the above-mentioned various components, through the hierarchically arranged water quality sensor network, the three-dimensional monitoring of water environment parameters is realized, and key indicators such as temperature, dissolved oxygen, pH value, etc. of different water layers are accurately obtained. At the same time, the monitoring of water flow direction, flow velocity and underwater acoustic feature collection are introduced, enriching the dimension of environmental monitoring. The Fourier transform is used to process the water body stratification parameters and acoustic signals, extract the group behavior characteristics of aquatic animals and water body flow characteristics, establish a biological-environment coupling data model, and realize the quantitative correlation analysis of the behavior of aquaculture objects and environmental factors. Based on the biological-environment coupling data, the water quality correlation degree and water body exchange period between multiple ponds are calculated, realizing the collaborative regulation at the regional level and avoiding the regional imbalance problem that may be caused by the independent control of single ponds. By analyzing the daily and night activity rules and group behavior patterns of aquatic animals, and combining with the change trend of dissolved oxygen in each pond to calculate the feeding amount, a precise feeding model based on biological rhythm is established, improving the feed utilization efficiency. During the operation of the equipment, by calculating the carbon footprint of energy consumption and dissolved gases in the water body, the mapping relationship between equipment operation and carbon emissions is established, realizing the operation optimization oriented by low carbon. Finally, through multi-dimensional correlation analysis, biological indicators, environmental parameters and carbon emission data are comprehensively evaluated to form ecological aquaculture optimization data, providing a scientific basis for the decision-making in the aquaculture process. The Fourier transform algorithm adopted in this solution has significant advantages in processing periodic signals and is particularly suitable for analyzing the biological rhythm characteristics in aquaculture; the regional collaborative algorithm based on water quality correlation considers the spatial correlation between multiple ponds, improving the accuracy of control strategies; the carbon footprint accounting model introduces environmental protection requirements into the intelligent control process, reflecting the innovation of the algorithm in specific application scenarios. Through the organic combination of these intelligent algorithms, the precise, collaborative and low-carbon management goals of the aquaculture process are realized.

[0207] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, system and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0208] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or equivalently replace some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent monitoring method for aquaculture based on the Internet of Things, characterized in that: The aquaculture intelligent monitoring method based on the Internet of Things includes: The temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple breeding ponds are collected through layered water quality sensors. The acoustic characteristic signals of aquatic animals are also collected. The parameters are collected using GPRS, 4G, WiFi, and Zigbee wireless transmission methods to obtain multi-pond ecological environment linkage data. Based on the multi-pond ecological environment linkage data, Fourier transform processing is performed on water body stratification parameters and acoustic signals to extract aquatic animal group behavior characteristics and water body flow characteristics to generate biological-environment coupling data; According to the biological-environmental coupling data, the water quality correlation and water exchange cycle between multiple ponds are calculated, and the operating parameters of the aeration equipment of each pond are coordinated to form regional balance control data; Based on the biological-environment coupling data and regional balance control data, the feeding amount is calculated by analyzing the diurnal activity patterns and group behavior patterns of aquatic animals and combining the dissolved oxygen change trend of each pond to obtain the biorhythm feeding data; Based on the multi-pond ecological environment linkage data and regional balance control data, the carbon footprint of equipment energy consumption and dissolved gas in water is calculated, and carbon emissions are optimized by adjusting the equipment operation timing to generate low-carbon operation data; Based on the biorhythm feeding data and low-carbon operation data, a multi-dimensional correlation analysis is performed on the biological indicators, environmental parameters and carbon emission data in the breeding process to form ecological breeding optimization data.

2. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: The water quality sensors arranged in layers collect the temperature, dissolved oxygen, pH value, water flow direction, and flow velocity parameters of multiple breeding ponds, and collect acoustic characteristic signals of aquatic animals at the same time, and use GPRS, 4G, WiFi, and Zigbee wireless transmission methods to collect data on the parameters to obtain multi-pond ecological environment linkage data, including: Temperature sensors, dissolved oxygen sensors, and pH sensors are placed at different water layers in the aquaculture pond, water flow direction sensors and flow velocity sensors are placed at the bottom of the pond, and underwater acoustic sensors are placed around the pond to collect sensor data; The temperature data collected by the temperature sensor are grouped according to the water layer position, and the average temperature value and temperature difference change rate of each water layer are calculated; The dissolved oxygen data collected by the dissolved oxygen sensor is divided into day and night time periods, and the fluctuation range and change trend of dissolved oxygen in each time period are calculated; Cross-comparing the pH data collected by the pH sensor with the temperature data to analyze the correlation between pH value and temperature change; Calculate the flow velocity vector and flow field distribution diagram of the water body in the pond from the data collected by the water flow direction sensor and the flow velocity sensor; Extracting signal amplitude and frequency characteristics from the acoustic signal collected by the underwater acoustic sensor to form an acoustic fingerprint of aquatic animal activities; The temperature data, the dissolved oxygen data, the pH value data, the water flow velocity vector and the acoustic fingerprint of the aquatic animal activity are transmitted in groups via GPRS, 4G, WiFi, or Zigbee wireless networks; Perform timestamp alignment and data compensation on the data transmitted in packets to generate multi-pond ecological environment linkage data.

3. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: According to the multi-pond ecological environment linkage data, the water body stratification parameters and acoustic signals are processed by Fourier transform, the behavior characteristics of aquatic animal groups and water body flow characteristics are extracted, and the biological-environment coupling data are generated, including: The water body layer parameters in the multi-pond ecological environment linkage data were segmented into time series, and the temperature, dissolved oxygen, and pH value data of different water layers were reorganized according to the sampling time points to obtain the water body parameter time series data; The water body parameter time series data is processed by Fourier transform to obtain the frequency spectrum characteristic value of each parameter and obtain water quality fluctuation period data; The acoustic characteristic signal is segmented according to the signal strength, and each segment of the signal is processed by Fourier transform to obtain the spectrogram data; Performing frequency analysis on the spectrogram data to extract the movement frequency, activity intensity, and aggregation degree parameters of aquatic animals to form group behavior data; Conduct time series analysis on water flow direction and velocity parameters, calculate water flow field distribution, and obtain water flow data; Time-aligning the water quality fluctuation cycle data with the group behavior data, analyzing the correlation between water quality changes and aquatic animal behaviors, and obtaining behavior correlation data; Performing spatial distribution analysis on the water flow data to determine the impact range of water flow on aquatic animal activities and obtaining flow field impact data; The behavior association data is fused with the flow field impact data to generate biological-environment coupling data, wherein the biological-environment coupling data includes the behavior characteristics of aquatic animal groups, the flow characteristics of water bodies, and the interactive impact relationship data between the two.

4. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: The method calculates the water quality correlation and water exchange cycle between multiple ponds based on the biological-environmental coupling data, coordinates and adjusts the operating parameters of the aeration equipment in each pond to form regional balance control data, including: Extract the time series of water quality parameters of each pond from the biological-environmental coupling data, group them according to parameter type and time point, and obtain pond water quality characteristic data; The water quality characteristic data of each pond were cross-matched, and the temperature difference coefficient, dissolved oxygen concentration gradient and pH deviation between the paired ponds were calculated to obtain the water quality correlation between the ponds; The water flow direction and velocity data of each pond were analyzed in time and space, and the water flow flux and flow direction change law between adjacent ponds were calculated to obtain the water exchange cycle; The ponds are divided and clustered according to the water quality correlation, and ponds with similar water quality parameters are divided into the same control area to obtain a regional coordinated control plan; Based on the regional collaborative control scheme, the operation data of the aeration equipment of each pond is analyzed, the corresponding relationship between the equipment start time, the operating power and the aeration effect is extracted, and the equipment operation characteristics are obtained; The water quality correlation, water exchange cycle and equipment operation characteristics are correlated and calculated, and the operation time period and power parameters are allocated to each oxygenation equipment in the control area to obtain an equipment coordination plan; The operation effect of the equipment coordination scheme is evaluated, the uniformity of dissolved oxygen distribution and energy consumption distribution in the area are calculated, and regional balance control data are generated.

5. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: Based on the biological-environment coupling data and regional balance control data, the feeding amount is calculated by analyzing the diurnal activity patterns and group behavior patterns of aquatic animals and combining the dissolved oxygen change trend of each pond to obtain the biological rhythm feeding data, including: Extract the acoustic signal intensity variation data of aquatic animals from the biological-environmental coupling data, divide it into 24-hour time windows, and obtain the diurnal activity signal sequence; Performing peak detection on the circadian activity signal sequence, calculating the time interval and duration of the signal intensity peak, and obtaining the activity rhythm cycle; Based on the activity rhythm cycle, the acoustic characteristic signal is segmented and counted, and the total signal energy and frequency distribution in each time period are calculated to obtain the group activity intensity; Extract the dissolved oxygen data of each pond from the regional balance control data, calculate the change rate and fluctuation range of dissolved oxygen concentration over time, and obtain the dynamic characteristics of dissolved oxygen; The group activity intensity is time-correlated with the dissolved oxygen dynamic characteristics, and the correlation coefficient and time lag relationship between the two are calculated to obtain the activity oxygen consumption law; Comprehensively analyzing the oxygen consumption pattern of the activity and the intensity of group activity, calculating the oxygen consumption per unit biomass in each time period, and obtaining the feed coefficient; The feed coefficient is combined with the activity rhythm cycle, the feeding amount is calculated according to different time periods, and the biorhythm feeding data is generated.

6. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: According to the multi-pond ecological environment linkage data and regional balance control data, the carbon footprint of equipment energy consumption and dissolved gas in water is calculated, and carbon emissions are optimized by adjusting the equipment operation timing to generate low-carbon operation data, including: Extract the sampling time point and power data of each water quality sensor from the multi-pond ecological environment linkage data, calculate the energy consumption value per unit time of each device, and obtain the device energy consumption data; Extract the operation time series of the oxygenation equipment from the regional balance control data, calculate the start and stop frequency and operation time of the equipment in each time period, and obtain the operation sequence data; The dissolved oxygen concentration data of each pond is divided into time series, and the corresponding relationship between the change in dissolved oxygen and the operating time of the aeration equipment is calculated to obtain the aeration efficiency data; Cross-analyze the operation sequence data and the oxygenation efficiency data, calculate the energy consumption corresponding to the unit oxygenation amount, and obtain the carbon emission intensity; The energy consumption data of the equipment is grouped and counted according to time periods, and the total energy consumption and carbon emissions of each time period are calculated to obtain carbon footprint data; Rearrange the equipment operation schedule based on the carbon footprint data, transfer the operation tasks in the high energy consumption period to the low energy consumption period, and obtain a schedule optimization solution; The timing optimization scheme is combined with the carbon footprint data, and the optimized energy consumption distribution and carbon emissions are calculated to generate low-carbon operation data.

7. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 1 is characterized in that: According to the biorhythm feeding data and low-carbon operation data, a multi-dimensional correlation analysis is performed on the biological indicators, environmental parameters and carbon emission data in the breeding process to form ecological breeding optimization data, including: Extract the acoustic signal characteristics and dissolved oxygen fluctuation curve during the feeding period from the biological rhythm feeding data, calculate the time delay relationship between the two through signal cross-correlation analysis, and obtain the biological response characteristic data; Extract equipment energy consumption series from low-carbon operation data, combine with the concentration change rate of dissolved gas in water, construct dynamic relationship map of carbon-oxygen conversion, and obtain ecological carbon-oxygen flow data; Performing time-frequency analysis on the biological response characteristic data, extracting the biological activity cycle and intensity variation law at different time scales, calculating the coupling relationship with water quality parameters, and obtaining ecological rhythm data; The ecological carbon-oxygen flow data is classified according to the carbon-oxygen conversion efficiency, the optimal operation range within the key time window is identified, and the energy efficiency balance data is obtained; Based on the ecological rhythm data and the energy efficiency balance data, a carbon efficiency growth evaluation system is established by calculating the synergy index of biological activity and energy utilization to obtain ecological efficiency data; The spatiotemporal distribution of the ecological efficiency data is analyzed to identify efficient breeding periods and areas, and the optimal breeding density is calculated in combination with the environmental carrying capacity to generate ecological breeding optimization data.

8. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 7 is characterized in that: The ecological carbon-oxygen flow data is classified according to the carbon-oxygen conversion efficiency, the optimal operation range in the key time window is identified, and the energy efficiency balance data is obtained, including: Phase decomposition of ecological carbon and oxygen flow data is performed to separate the instantaneous frequency characteristics of carbon emission fluctuations and oxygen conversion, and obtain the carbon-oxygen coupling sequence; Performing nonlinear dynamic analysis on the carbon-oxygen coupling sequence, calculating the mutation point and steady-state interval in the carbon-oxygen conversion process, and obtaining conversion characteristic data; Performing a balance calculation on the conversion characteristic data according to the energy conservation principle, establishing a quantitative relationship between carbon emissions and dissolved oxygen production, and obtaining energy balance data; Constructing a time window matrix based on the energy balance data, extracting the peak and valley values ​​of energy conversion efficiency within the window, and obtaining critical threshold data; Adaptively segmenting the critical threshold data, identifying the optimal operating point and saturation point of carbon-oxygen conversion efficiency, and obtaining operating boundary data; Dynamic compensation calculation is performed on the operation boundary data to balance the carbon emission intensity and the dissolved oxygen supply rate to generate energy efficiency balance data.

9. The aquaculture intelligent monitoring method based on the Internet of Things according to claim 7 is characterized in that: The temporal and spatial distribution analysis of the ecological efficiency data is performed to identify efficient breeding time periods and areas, and the optimal breeding density is calculated in combination with the environmental carrying capacity to generate ecological breeding optimization data, including: Performing wavelet decomposition on the ecological efficiency data to separate the time-varying characteristics of different frequency bands, extracting the fluctuation period of biological activity and energy utilization, and obtaining multi-scale efficiency data; Interpolating the multi-scale efficiency data according to spatial coordinates to construct a three-dimensional efficiency field distribution of the breeding area to obtain efficiency field data; Performing gradient analysis on the efficiency field data, identifying the rate of change and propagation direction of the efficiency field, calculating the efficiency flow relationship between the regions, and obtaining efficiency flow data; Based on the said efficient flow data, the diffusion characteristics of the water body are quantified, and the material circulation data are obtained by combining the migration law of dissolved oxygen and nutrients; The material circulation data is correlated with the dissolved oxygen consumption rate to construct a dynamic equilibrium relationship between the water body's self-purification capacity and the aquaculture load, and the carrying threshold data is obtained; The carrying threshold data is optimized in time and space, the maximum breeding capacity in different areas and time periods is calculated, and ecological breeding optimization data is generated.

10. An intelligent monitoring system for aquaculture based on the Internet of Things, used to implement the intelligent monitoring method for aquaculture based on the Internet of Things as claimed in any one of claims 1 to 9, characterized in that: The aquaculture intelligent monitoring system based on the Internet of Things includes: The acquisition module is used to collect the temperature, dissolved oxygen, pH value, water flow direction, and flow rate parameters of multiple breeding ponds through layered water quality sensors, and collect acoustic characteristic signals of aquatic animals. The parameters are collected using GPRS, 4G, WiFi, and Zigbee wireless transmission methods to obtain multi-pond ecological environment linkage data; A transformation module is used to perform Fourier transformation processing on water body stratification parameters and acoustic signals based on the multi-pond ecological environment linkage data, extract the behavior characteristics of aquatic animal groups and water body flow characteristics, and generate biological-environment coupling data; A regulation module, for calculating the water quality correlation and water exchange period between multiple ponds according to the biological-environmental coupling data, and coordinating the operating parameters of the aeration equipment of each pond to form regional balance control data; A calculation module is used to calculate the feeding amount based on the biological-environment coupling data and the regional balance control data, by analyzing the diurnal activity patterns and group behavior patterns of aquatic animals and combining the dissolved oxygen change trend of each pond to obtain the biorhythm feeding data; A generation module is used to calculate the carbon footprint of equipment energy consumption and dissolved gas in water according to the multi-pond ecological environment linkage data and regional balance control data, optimize carbon emissions by adjusting the equipment operation timing, and generate low-carbon operation data; The correlation module is used to perform multi-dimensional correlation analysis on biological indicators, environmental parameters and carbon emission data in the breeding process according to the biorhythm feeding data and low-carbon operation data to form ecological breeding optimization data.

Images