Aquaculture tail water flow real-time estimation method

Abstract

The invention discloses a real-time estimation method for tail water flow of aquaculture. The method comprises the following steps: firstly, comprehensively collecting basic environment data of geographic position, landform, water quality and geometric dimension of an aquaculture pond, and arranging and storing the data; secondly, monitoring the initial flow velocity, flow direction and cross sectional area change of the water flow at the tail water discharge port in real time by using professional equipment, and transmitting data according to preset intervals; then analyzing interference factors such as wind power, tide and cultivated organism activity, constructing a correlation model to correct water flow state parameters, and preliminarily estimating instantaneous flow based on a water flow dynamics algorithm; and predicting a flow change trend through time sequence analysis, and constructing an incidence matrix optimization estimation value in combination with production activity data. And finally, the accurate estimated value is transmitted to a culture management system database in a JSON format through a Modbus protocol, and key data support is provided for scientific management. According to the method, a more scientific, efficient and accurate solution is provided for estimating the tail water flow in the aquaculture industry.

Description

Technical Field

[0001] This invention relates to the field of crayfish farming counting, and in particular to a method for real-time estimation of aquaculture tailwater flow. Background Technology

[0002] Aquaculture is growing rapidly worldwide, playing a vital role in meeting market demand for aquatic products. However, with the continuous expansion of aquaculture scale, the problem of wastewater discharge has become increasingly prominent. Wastewater carries large amounts of uneaten feed, feces, chemicals, and microbial pollutants. If discharged directly without proper treatment, it will severely damage the surrounding aquatic ecosystem, causing a series of environmental problems such as eutrophication and water quality deterioration. Accurately determining wastewater flow rate is crucial for the rational planning of wastewater treatment facilities, precise assessment of pollutant discharge, and effective formulation of environmental protection strategies. This is the important background for the development of "a real-time estimation method for aquaculture wastewater flow rate".

[0003] Existing technologies for estimating aquaculture wastewater flow rates have several shortcomings. Firstly, traditional flow monitoring equipment is often complex and costly to install, requiring specialized installation and commissioning in wastewater discharge pipes or ditches. For aquaculture ponds that are geographically dispersed, large-scale deployment of such equipment not only incurs significant financial costs but also high maintenance costs, limiting its widespread application. Secondly, early estimation methods were mostly simplistic and crude, considering only one or a few influencing factors, such as relying solely on water flow velocity or the cross-sectional area of ​​the discharge outlet to estimate wastewater flow rates, ignoring the combined effects of complex factors such as wind, tides, and aquaculture biological activity. This one-sided estimation approach leads to significant discrepancies between the estimated results and actual wastewater flow rates, failing to provide reliable data support for wastewater treatment and environmental management.

[0004] Furthermore, existing tailwater flow estimation technologies are inadequate in terms of real-time performance and dynamic adaptability. Aquaculture production activities are characterized by dynamic changes; for example, operations such as feeding, water exchange, and aeration can significantly impact tailwater flow at different times. However, existing methods struggle to track these changes in production activities in real time and adjust tailwater flow estimates accordingly. Simultaneously, existing technologies lack effective mechanisms to cope with changes in the aquaculture environment under different seasons and weather conditions, failing to accurately reflect the dynamic changes in tailwater flow and thus failing to meet the practical needs of modern aquaculture for refined management and environmental protection. Summary of the Invention

[0005] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a method for real-time estimation of aquaculture tailwater flow.

[0006] A method for real-time estimation of aquaculture tailwater flow includes the following steps:

[0007] Collect basic environmental data related to wastewater discharge within aquaculture areas;

[0008] The initial state parameters of the water flow at the aquaculture wastewater discharge outlet are monitored in real time using preset monitoring equipment. The initial flow velocity, initial flow direction and real-time change data of the cross-sectional area of ​​the water flow at the discharge outlet are obtained, and the monitored initial state parameters are transmitted to the data at set time intervals.

[0009] Based on the collected basic environmental data and the monitored initial state parameters of the water flow, a systematic analysis was conducted on the interference factors affecting the flow rate during the discharge of aquaculture wastewater, including the assessment of the impact of wind, tides, and aquaculture biological activity on the water flow, and a preliminary correlation model between interference factors and wastewater flow rate was established.

[0010] Based on the preliminary correlation model, the monitored initial state parameters of the water flow are corrected for interference factors. By using the set correction algorithm, the influence caused by interference factors is removed from the initial state parameters, and the water flow state parameters after interference correction are obtained.

[0011] Using a pre-defined flow estimation algorithm, the flow state parameters after interference correction are input, and combined with the basic information of the geometric shape and size of the aquaculture pond, the instantaneous flow rate of the aquaculture tailwater is initially estimated. This flow estimation algorithm is based on the principle of hydrodynamics and is constructed by comprehensively considering factors such as flow velocity, cross-sectional area and flow characteristics.

[0012] Time series analysis was performed on the instantaneous flow data obtained from the preliminary estimation. The time series analysis model was used to predict and analyze the trend of tailwater flow over time. The predicted value of tailwater flow over a future period was obtained through model calculation. The time series analysis model was trained and optimized based on historical flow data and current flow state parameters.

[0013] The predicted tailwater flow rate trend of the time series analysis model is correlated and compared with the actual production activity data of the aquaculture area. The actual production activity data includes the feeding time of the aquaculture organisms, the water exchange operation time, and the start time of the aeration equipment. Based on the comparative analysis results, the estimated tailwater flow rate is further corrected and optimized to obtain the real-time estimated tailwater flow rate.

[0014] The real-time estimated tailwater flow rate is transmitted to the designated database of the aquaculture management system according to the preset data transmission protocol and format. Aquaculture managers can obtain tailwater flow rate information in real time and perform data analysis and management decisions.

[0015] Furthermore, when establishing a preliminary correlation model between the interfering factors and the tailrace flow rate, a multiple linear regression model was used. The model expression is as follows:

[0016]

[0017] in, Represents tailwater flow rate; These represent various interfering factors, including wind intensity, tidal force, and aquaculture activity. These are the coefficients corresponding to each interference factor; For constant terms, the coefficients and constant terms The parameters were determined through statistical analysis and parameter fitting of a large amount of basic environmental data covering different seasons and weather conditions, initial water flow parameters, and historical monitoring data of corresponding tailwater flow measured values.

[0018] Furthermore, when fitting the parameters of the aforementioned multiple linear regression model, the method of minimizing the sum of the absolute values ​​of the residuals is used to solve the problem, that is:

[0019] For a set containing Historical monitoring data of sample number , let the first sample be... The actual measured tailwater flow rate in each sample was [value missing]. The tailwater flow rate predicted by the model is Construct the objective function By analyzing the coefficients and constant terms Find the partial derivatives, set them to zero, construct a system of equations to solve, and determine the solution. To achieve the minimum model parameters as well as The value of .

[0020] Furthermore, the flow estimation algorithm employed is based on an improved extended form of the Darcy-Weisbach equation. Taking into account energy losses during the tailrace flow and the complex turbulent characteristics of the flow, the equation is modified. The modified equation is as follows:

[0021]

[0022] in, This refers to the tailwater flow rate; The comprehensive flow correction coefficient is related to various factors such as the degree of turbulence in the water flow, the roughness and shape of the discharge outlet; It is the cross-sectional area of ​​the water flow at the tailwater discharge outlet; It is the acceleration due to gravity; The head difference between the upstream and downstream sides of the tailwater discharge outlet; This is the friction factor, which is related to the Reynolds number of the water flow and the roughness of the pipe. The length of the tailrace flow path; The hydraulic diameter; This is the sum of local drag coefficients, related to the boundary conditions of various local obstruction structures at the discharge outlet, and is a comprehensive flow correction coefficient. The friction coefficient was determined through a combination of experimental testing and theoretical derivation. and the sum of local drag coefficients Based on the detailed structure of the discharge outlet and the real-time water flow status, the data is obtained through theoretical calculations or numerical simulations.

[0023] Furthermore, in determining the comprehensive flow correction coefficient An experimental testing platform was built to simulate the actual discharge scenario of aquaculture wastewater. Under different water flow velocities, turbulence levels, outlet roughness, and shapes, the wastewater flow rate was measured and compared with the flow rate calculated based on the modified extended form of the Darcy-Weisbach equation. A comprehensive flow correction coefficient was obtained through data fitting. With water flow velocity Parameters of water flow turbulence Roughness parameters of the discharge port and emission outlet shape parameters The functional relationship between them is:

[0024]

[0025] Based on the real-time monitored initial flow parameters and the geometry and roughness information of the discharge outlet, the comprehensive flow correction coefficient is determined through this functional relationship. The value of .

[0026] Furthermore, the time series analysis model used is an improved model based on the threshold autoregressive TAR model. This improved model is superior to the traditional TAR (… Based on the model, an external dynamic variable closely related to aquaculture production activities is introduced. Constructing an extended model TAR ( The model expression is: -Z

[0027]

[0028] in, express The tailwater flow rate at any given moment; For different threshold intervals, the autoregression order is denoted as ; and These are the autoregressive coefficients for the corresponding intervals; This is the threshold value; The delay order; external variables Coefficients in different intervals; For white noise sequences in different intervals, the autoregressive coefficients Threshold Delay order and external variable coefficients The aquaculture production activity data was determined by performing maximum likelihood estimation on historical tailwater flow data and corresponding aquaculture production activity data. It was obtained by extracting and organizing production record data from the aquaculture management system.

[0029] Furthermore, in the context of the aforementioned When estimating the parameters of the model, the Akaike Information Criterion (AICc) is used as the key criterion for model selection. Adjustments are made based on the finite number of samples, and different autoregressive orders are iterated through. Threshold and delay order The combinations are used to calculate the AIC of the model under each combination. The optimal model parameters are selected based on the combination of model parameters that minimizes the AICC value. The formula for calculating the AICC value is as follows:

[0030]

[0031] in, The number of samples; It is the sum of squared residuals; This represents the number of parameters to be estimated in the model.

[0032] Furthermore, when comparing and analyzing the predicted tailwater flow rate trend by the time series analysis model with the actual production activity data of the aquaculture area, a dynamic correlation tensor between production activities and flow rate is constructed. The first dimension of the tensor represents different types of production activities, including feeding, water exchange, and aeration; the second dimension represents different time scale divisions, from short-term periods to long-term cycles; and the third dimension represents the degree of influence of different production activities on tailwater flow rate at different spatial locations at the corresponding time scale. This degree of influence is obtained through three-dimensional correlation analysis of the occurrence time and spatial location of production activities with the changes in tailwater flow rate at corresponding spatiotemporal points in historical data. The specific calculation method is as follows: for each type of production activity, at different time scales, the three-dimensional cross-correlation coefficient between the occurrence time at different spatial locations and the change time of tailwater flow rate at the corresponding spatial locations is calculated. The strength of the influence is determined based on the magnitude of the correlation coefficient. The dynamic correlation tensor between production activities and flow rate is used as a reference to correct and optimize the estimated value of tailwater flow rate.

[0033] Furthermore, when correcting the estimated tailwater flow rate based on the aforementioned production activity-flow dynamic correlation tensor, a spatiotemporal weighted correction method is adopted, assuming that the time series analysis model predicts... Time, spatial location The estimated tailwater flow rate at the location is ,exist The time scale interval corresponding to the time and Within the corresponding spatial location range, there exists a production activity type of The corresponding degrees of influence are as follows: Corrected tailwater flow rate estimate for:

[0034]

[0035] in, For production activity type exist Time, spatial location The impact of production activities on tailwater flow rate is obtained by statistical analysis of the changes in tailwater flow rate before and after the corresponding time and space points in historical data.

[0036] Beneficial Effects: This invention proposes a real-time estimation method for aquaculture wastewater flow. In the basic data acquisition stage, this method comprehensively collects fundamental environmental data such as the geographical location, topography, initial water quality, and geometric dimensions of the aquaculture pond, providing solid support for subsequent accurate analysis and helping to grasp the background conditions of wastewater discharge from a holistic environmental perspective. In terms of real-time monitoring, the method dynamically tracks changes in the initial flow velocity, direction, and cross-sectional area of ​​the wastewater discharge outlet, transmitting data at set intervals to ensure timely and accurate initial state information. Regarding interference factor handling, the method systematically analyzes interference factors such as wind, tides, and aquaculture biological activity, constructs a correlation model, and corrects the initial state parameters of the flow accordingly, significantly improving the reliability of the basic data for flow estimation. The flow estimation algorithm, based on the principles of hydrodynamics, comprehensively considers multiple characteristics of water flow and can initially and accurately estimate instantaneous flow. The introduction of a time series analysis model, combined with historical flow and current water flow parameters, predicts the flow change trend, providing a powerful tool for dynamic flow assessment. By correlating and comparing the predicted trend with aquaculture production activity data, the estimated values ​​are further corrected and optimized, ensuring that the results closely reflect the actual impact of aquaculture operations on wastewater flow and significantly improving estimation accuracy. Finally, the real-time estimated values ​​are transmitted to the aquaculture management system database according to the preset protocol, which makes it convenient for managers to obtain information in real time and help them make timely and scientific and reasonable decisions. Whether it is adjusting aquaculture strategies to optimize wastewater discharge or conducting environmental assessments, the accurate flow data provided by this method can make the work more targeted and effective. Attached Figure Description

[0037] Figure 1 This is a flowchart of the method steps of the present invention;

[0038] Figure 2 This is a diagram showing the composition of the method operation unit of the present invention. Detailed Implementation

[0039] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] like Figure 1 As shown, a method for real-time estimation of aquaculture tailwater flow includes the following steps:

[0041] Step 1: Collect comprehensive basic environmental data related to wastewater discharge in the aquaculture area, including the geographical location of the aquaculture ponds, surrounding topographic data, initial water quality parameters of the aquaculture water body, and basic information such as the geometric shape and size of the ponds. Organize and store the collected basic environmental data according to the set data format.

[0042] Specifically, this step focuses on comprehensively collecting basic environmental data closely related to aquaculture wastewater discharge. Using the Global Positioning System (GPS), the geographical location of the aquaculture pond can be precisely pinpointed, with latitude and longitude accurate to six decimal places. This is akin to marking the pond with extremely precise coordinates on a map, laying a solid foundation for accurately locating the starting point of wastewater discharge. The topographic mapping instrument records detailed information about the surrounding terrain, such as the presence of slopes and the direction of ditches. For example, if there is a slope of approximately 5° around the pond, the wastewater is highly likely to flow naturally along this slope due to gravity during discharge. This data is crucial for in-depth analysis of the natural flow path of the wastewater.

[0043] In terms of water quality testing, professional water quality testing instruments can accurately measure initial parameters such as pH, dissolved oxygen content, and ammonia nitrogen concentration in aquaculture water. pH is usually expressed as a value; the normal pH of aquaculture water is generally maintained between 7.5 and 8.5. A pH deviation from this range may indicate an abnormality in the water, affecting the characteristics of the effluent. Dissolved oxygen content generally needs to be maintained above 5 mg / L to ensure the normal survival of aquatic organisms and also to avoid impacting the environment after effluent discharge. Ammonia nitrogen concentration should be controlled at a low level, such as below 0.2 mg / L, as excessively high ammonia nitrogen concentrations can be toxic to aquatic organisms.

[0044] For the geometric dimensions of the pond, a laser rangefinder and measuring tape are used to accurately obtain data such as the pond's length, width, and depth. For example, a rectangular pond might be 50 meters long, 20 meters wide, and have an average depth of 1.5 meters. These data are the core parameters for subsequently building a tailrace flow estimation model. After data collection, it is organized and stored in a specific format similar to that required by an SQL database, so that the data can be easily and quickly retrieved at any time during subsequent analysis.

[0045] Step 2: Use the preset monitoring equipment to monitor the initial state parameters of the water flow at the aquaculture wastewater discharge outlet in real time, obtain the real-time change data of the initial flow velocity, initial flow direction and cross-sectional area of ​​the water flow at the discharge outlet, and transmit the monitored initial state parameters at set time intervals.

[0046] Specifically, this step relies on specialized monitoring equipment to continuously and in real-time track the initial state parameters of the water flow at the aquaculture wastewater discharge outlet. Taking a Doppler current meter as an example, it cleverly utilizes the Doppler effect by transmitting and receiving ultrasonic signals to accurately measure water flow velocity down to 0.01 meters per second. In practical applications, if the monitored water flow velocity at the discharge outlet is 0.35 meters per second, this precise data is crucial for subsequent analysis of the wastewater flow rate.

[0047] The initial flow direction can be accurately obtained using an electronic compass, with an accuracy of 1 degree. Assuming the electronic compass measures the initial flow direction of the tailwater to 30 degrees northeast, this information helps determine the initial flow direction and whether it conforms to the planned discharge schedule. For the cross-sectional area of ​​the water flow at the discharge outlet, real-time monitoring of its dynamic changes is possible by combining advanced image recognition technology and professional measurement equipment. For example, when there is debris clogging the discharge outlet, image recognition technology can promptly detect a decrease in the cross-sectional area.

[0048] The data collected from these monitoring sessions is rapidly transmitted to the data processing terminal via a stable and efficient 4G or 5G wireless communication module at pre-set 5-minute intervals. This setup ensures that data can be collected in a timely, continuous, and stable manner, providing sufficient and real-time data support for subsequent data analysis and processing, and guaranteeing the timeliness and accuracy of tailwater flow estimation.

[0049] Step 3: Based on the collected basic environmental data and the monitored initial state parameters of the water flow, conduct a systematic analysis of the interference factors that may affect the flow rate during the discharge of aquaculture wastewater, including but not limited to the assessment of the degree of influence of factors such as wind, tides, and aquaculture biological activities on the water flow, and establish a preliminary correlation model between interference factors and wastewater flow rate.

[0050] Specifically, this step, based on the basic environmental data collected in the first step and the initial water flow parameters monitored in the second step, comprehensively and deeply analyzes various interference factors affecting the tailrace flow rate. Taking wind interference as an example, an anemometer can accurately measure wind speed. In a real-world scenario, if the anemometer measures a wind force of level 4, according to Bernoulli's principle, such a strong wind will exert pressure on the water surface, thus significantly affecting the water flow velocity and direction. For example, it might increase the water flow velocity from 0.3 meters per second to 0.35 meters per second, and the flow direction might shift by about 5 degrees.

[0051] Regarding tides, by referring to professional tidal forecast data, the impact on the water level difference at the tailrace discharge outlet can be analyzed in detail. In some aquaculture areas near the sea, tidal fluctuations are significant, with the water level at high tide potentially being 2 meters higher than at low tide. This greatly alters the water level difference at the tailrace discharge outlet, thus affecting the tailrace flow. As for disturbances caused by aquaculture activities, their impact on water flow can be assessed by carefully observing indicators such as the density and activity intensity of the aquaculture organisms. For example, when the fish density in aquaculture ponds is high and they are actively feeding, their swimming will cause minor but measurable disturbances to the water flow.

[0052] Using scientific methods such as multiple linear regression, a preliminary correlation model was established between various interfering factors, including wind speed, tidal intensity, and the intensity of aquaculture activity, and tailwater flow. Through in-depth statistical analysis of a large amount of historical monitoring data, the specific impact of each interfering factor on tailwater flow and its corresponding constants were determined. For example, the analysis revealed that under specific aquaculture conditions, a 1-level increase in wind speed may lead to a 5% increase in tailwater flow, and a 1-unit change in tidal intensity may result in a 3% change in tailwater flow. These data provide important basis for subsequent revisions to tailwater flow estimates.

[0053] Step 4: Based on the preliminary correlation model between interference factors and tailwater flow established in Step 3, the monitored initial state parameters of the water flow are corrected for interference factors. Through the set correction algorithm, the influence caused by interference factors is removed from the initial state parameters, and the water flow state parameters after interference correction are obtained.

[0054] Specifically, based on the preliminary correlation model between the interfering factors and the tailrace flow rate successfully established in step three, a specially designed correction algorithm is used to accurately correct the initial state parameters of the water flow monitored in step two. For example, during actual monitoring, it was found that due to the current wind force being level 5, according to the previously established correlation model, this wind force would cause an additional increase in water flow velocity of 0.05 meters per second. Through rigorous calculations using the influence degree and constants corresponding to this wind force determined in the correlation model, it was concluded that this additional increase caused by the wind force should be subtracted from the initially monitored water flow velocity.

[0055] Assuming the initial monitored water flow velocity is 0.4 m / s, subtracting the 0.05 m / s increase due to wind force yields a corrected flow velocity of 0.35 m / s. Regarding the flow direction, if the correlation model indicates that a force 5 wind will shift the flow direction by 8 degrees, and the initial monitored flow direction is 40 degrees southeast, then the corrected flow direction is adjusted to 32 degrees southeast. Similarly, for the cross-sectional area of ​​the water flow at the outlet, if the correlation model indicates that factors such as wind force and tides will cause some changes, corresponding corrections are also necessary based on the model calculations. This interference correction process effectively removes the influence of interfering factors on the initial monitoring data, resulting in more accurate and reliable flow state parameters, laying a solid foundation for more precise estimation of the tailrace flow rate.

[0056] Step 5: Using the set flow estimation algorithm, input the water flow state parameters after interference correction, and combine the basic information such as the geometric shape and size of the aquaculture pond to make a preliminary estimate of the instantaneous flow rate of the aquaculture tailwater. This flow estimation algorithm is based on the principle of hydrodynamics and is constructed by comprehensively considering factors such as water flow velocity, cross-sectional area and water flow characteristics.

[0057] Specifically, this step employs a carefully designed flow estimation algorithm based on hydrodynamic principles, using the corrected flow state parameters from the fourth step, along with key fundamental information such as the geometric dimensions of the aquaculture pond obtained in the first step, as input data. In practice, several important factors need to be considered. For example, the degree of turbulence is judged by observing the fluctuations and splash patterns of the flow; if the flow exhibits significant fluctuations and splashing, it indicates a high degree of turbulence. The shape of the outlet is also crucial; circular and square outlets have different effects on obstructing and guiding the flow.

[0058] The cross-sectional area of ​​the water flow at the discharge outlet, after correction in the previous steps, is a key parameter for calculating the flow rate. The water level difference between the upstream and downstream of the discharge outlet can be measured using water level gauges installed at the upstream and downstream locations. Simultaneously, local obstructions must be considered, such as the presence of debris accumulation at the discharge outlet or the existence of simple man-made filtration devices, as these will affect the smoothness of water flow. Based on the values ​​of each factor determined through actual measurements and numerous experiments, a specific flow rate estimation algorithm is used for calculation. For example, in a certain aquaculture pond, after measurement and analysis to determine the values ​​of each factor, the algorithm calculates the instantaneous flow rate of the aquaculture wastewater to be 5 cubic meters per minute. This preliminary estimate provides important starting data for further analysis and optimization of the wastewater flow rate estimation.

[0059] Step 6: Perform time series analysis on the instantaneous flow data obtained from the preliminary estimation. Use the time series analysis model to predict and analyze the trend of tailwater flow over time. Obtain the predicted value of tailwater flow trend in the future through model calculation. This time series analysis model is trained and optimized based on historical flow data and current water flow state parameters.

[0060] Specifically, this step mainly focuses on time series analysis of the instantaneous flow data obtained from the preliminary estimation in step five. An improved model based on an autoregressive integral moving average model is used, fully integrating historical flow data and current real-time water flow parameters for comprehensive training and optimization. In practical applications, in-depth analysis of long-term accumulated historical flow data reveals certain patterns. For example, statistical analysis of flow data from the past year shows a regular change in tailwater flow after the feeding time at 10:00 AM each day. This is because after feeding, the activity of aquatic organisms intensifies, consuming more oxygen and increasing the demand for water exchange. The tailwater flow typically begins to rise within half an hour after feeding, reaching its peak in about one hour, increasing by about 15% compared to before feeding.

[0061] Based on these findings, feeding, a factor closely related to aquaculture production activities, was incorporated into the improved time series analysis model. By continuously adjusting the model parameters, it was made to better fit historical data and current water flow conditions. The trained and optimized model can predict the trend of tailwater flow changes over a future period with relatively high accuracy. For example, based on current water flow parameters and historical data patterns, the model predicts that after feeding at 10:00 AM tomorrow, the tailwater flow will gradually increase from the current 6 cubic meters per minute, reaching approximately 7 cubic meters per minute around 11:00 AM. This prediction provides an important basis for subsequent dynamic adjustments to tailwater flow estimation, helping to prepare for tailwater treatment and discharge in advance.

[0062] Step 7: Compare and analyze the predicted tailwater flow rate trend by the time series analysis model with the actual production activity data of the aquaculture area. The actual production activity data includes the feeding time of the aquaculture organisms, the water exchange operation time, and the start-up time of the aeration equipment. Based on the comparative analysis results, the estimated tailwater flow rate is further corrected and optimized to obtain a more accurate real-time estimated tailwater flow rate.

[0063] Specifically, this step involves a thorough comparative analysis of the predicted tailwater flow rate trend from the time series analysis model in step six, comparing it with actual production activity data from the aquaculture area. Taking feeding time as an example, a detailed analysis of feeding time and tailwater flow rate changes in a large amount of historical data reveals a significant pattern: within one hour after feeding, the rapid increase in aquaculture activity, including increased respiration and excretion, leads to a decrease in dissolved oxygen and an increase in metabolic waste in the water. To maintain a good aquaculture environment, farmers typically turn on aeration equipment or perform water changes. These combined factors increase tailwater flow rate by approximately 10%.

[0064] Based on this, a comprehensive correlation matrix between production activities and flow rate is constructed. This matrix details the specific impact of various production activities, such as feeding, water exchange, and aeration, on the tailwater flow rate at different time intervals. For example, within half an hour after the start of a water exchange operation, the tailwater flow rate may rapidly increase by 30% due to the injection of new water and the discharge of old water; after the aeration equipment is turned on, the tailwater flow rate will slowly increase by about 5% within one hour as the water flow increases. Based on this correlation matrix, the estimated tailwater flow rate is further corrected and optimized. Assuming the time series analysis model predicts a tailwater flow rate of 7 cubic meters per minute at a certain moment, and this occurs half an hour after feeding, according to the correlation matrix, the feeding factor will increase the flow rate by 10%. Therefore, the corrected and optimized estimated tailwater flow rate is adjusted to 7.7 cubic meters per minute, thus significantly improving the accuracy of the tailwater flow rate estimation.

[0065] Step 8: Transmit the real-time estimated tailwater flow rate value to the designated database of the aquaculture management system according to the preset data transmission protocol and format, so that aquaculture managers can obtain tailwater flow rate information in real time and conduct subsequent data analysis and management decisions.

[0066] Specifically, the more accurate real-time estimate of the effluent flow rate obtained after correction and optimization in step seven is transmitted to the designated database of the aquaculture management system in a simple and universal JSON format, according to a pre-set data transmission protocol, such as the widely used Modbus protocol. During actual transmission, the Modbus protocol ensures the stability and accuracy of data transmission, effectively preventing data loss or errors. The JSON format makes the data easy to parse and process across different systems and platforms.

[0067] Aquaculture managers can obtain this wastewater flow information in real time and efficiently through a convenient aquaculture management system. This accurate wastewater flow data has significant application value, allowing managers to analyze wastewater discharge patterns in depth. For example, analysis of wastewater flow data over a period of time reveals a significant peak in flow every Monday morning due to concentrated water changes and equipment cleaning. Simultaneously, this data helps evaluate the effectiveness of wastewater treatment. If the trend of wastewater flow and pollutant content improves effectively after implementing new wastewater treatment measures, it indicates that the treatment measures have achieved good results. The scientific analysis and utilization of wastewater flow information provides strong data support for the scientific and rational management of aquaculture, helping to improve aquaculture efficiency and protect the surrounding aquatic environment.

[0068] Preferably, in step three, when establishing the preliminary correlation model between the interfering factors and the tailwater flow rate, a multiple linear regression model is used, and the model expression is:

[0069]

[0070] in, Represents tailwater flow rate; These represent various interfering factors, such as wind intensity, tidal force, and activity level of aquaculture organisms. These are the coefficients corresponding to each interference factor; For constant terms, the coefficients and constant terms It was determined through statistical analysis and parameter fitting of a large amount of basic environmental data, initial water flow parameters, and corresponding measured tailwater flow values ​​under different seasons and weather conditions.

[0071] Preferably, when fitting the parameters of the multiple linear regression model, the method of minimizing the sum of the absolute values ​​of the residuals is used, that is:

[0072] For a set containing Historical monitoring data of sample number , let the first sample be... The actual measured tailwater flow rate in each sample was [value missing]. The tailwater flow rate predicted by the model is Construct the objective function By analyzing the coefficients and constant terms Find the partial derivatives, set them to zero, construct a system of equations to solve, and then determine the solution. To achieve the minimum model parameters as well as The value of .

[0073] Preferably, in step five, the flow estimation algorithm is based on an improved extended form of the Darcy-Weisbach equation. Considering the energy loss during the tailrace flow and the complex turbulent characteristics of the flow, the traditional equation is modified. The modified equation is as follows:

[0074]

[0075] in, This refers to the tailwater flow rate; The comprehensive flow correction coefficient is related to various factors such as the degree of turbulence in the water flow, the roughness and shape of the discharge outlet; It is the cross-sectional area of ​​the water flow at the tailwater discharge outlet; It is the acceleration due to gravity; The head difference between the upstream and downstream sides of the tailwater discharge outlet; This is the friction factor, which is related to the Reynolds number of the water flow and the roughness of the pipe. The length of the tailrace flow path; The hydraulic diameter; This is the sum of local drag coefficients, related to the boundary conditions of various local obstruction structures at the discharge outlet, and is a comprehensive flow correction coefficient. The friction coefficient was determined through a combination of experimental testing and theoretical derivation. and the sum of local drag coefficients Based on the detailed structure of the discharge outlet and the real-time water flow status, the data is obtained through theoretical calculations or numerical simulations.

[0076] Preferably, in determining the comprehensive flow correction coefficient An experimental testing platform was built to simulate the actual discharge scenario of aquaculture wastewater. Under different conditions such as water flow velocity, water turbulence, and outlet roughness and shape, the wastewater flow rate was measured and compared with the flow rate calculated based on the modified extended form of the Darcy-Weisbach equation. A comprehensive flow correction coefficient was obtained through data fitting. With water flow velocity Parameters of water flow turbulence Roughness parameters of the discharge port and emission outlet shape parameters The functional relationship between them is:

[0077]

[0078] In practical applications, the comprehensive flow correction coefficient is determined based on the real-time monitored initial flow parameters and the geometry and roughness information of the discharge outlet, using this functional relationship. The value of .

[0079] Preferably, in step six, the time series analysis model used is an improved model based on the threshold autoregressive (TAR) model. This improved model is based on the traditional TAR (… Based on the model, an external dynamic variable closely related to aquaculture production activities is introduced. Constructing an extended model TAR ( The model expression is: -Z

[0080]

[0081] in, express The tailwater flow rate at any given moment; For different threshold intervals, the autoregression order is denoted as ; and These are the autoregressive coefficients for the corresponding intervals; This is the threshold value; The delay order; external variables Coefficients in different intervals; For white noise sequences in different intervals, the autoregressive coefficients Threshold Delay order and external variable coefficients The aquaculture production activity data was determined by performing maximum likelihood estimation on historical tailwater flow data and corresponding aquaculture production activity data. It was obtained by extracting and organizing production record data from the aquaculture management system.

[0082] Preferably, in the case of the When estimating the parameters of the model, the Akaike Information Criterion (AICc) is used as the key criterion for model selection. However, adjustments are made due to the potential finiteness of the sample size, by iterating through different autoregressive orders. Threshold and delay order The combinations are used to calculate the AIC of the model under each combination. The optimal model parameters are selected based on the combination of model parameters that minimizes the AICC value. The formula for calculating the AICC value is as follows:

[0083]

[0084] in, The number of samples; It is the sum of squared residuals; The number of parameters to be estimated in the model is given. Model parameters determined in this way can effectively avoid overfitting while ensuring the model's fitting accuracy.

[0085] Preferably, in step seven, when comparing the predicted tailwater flow rate trend by the time series analysis model with the actual production activity data of the aquaculture area, a dynamic correlation tensor between production activities and flow rate is constructed. The first dimension of the tensor represents different types of production activities, such as feeding, water exchange, and aeration; the second dimension represents different time scale divisions, from short-term periods to long-term cycles; and the third dimension represents the degree of influence of different production activities on tailwater flow rate at different spatial locations at the corresponding time scale. This degree of influence is obtained through three-dimensional correlation analysis of the occurrence time and spatial location of production activities with the changes in tailwater flow rate at corresponding spatiotemporal points in historical data. Specifically, for each type of production activity, at different time scales, the three-dimensional cross-correlation coefficient between the occurrence time at different spatial locations and the change time of tailwater flow rate at the corresponding spatial location is calculated. The strength of the influence is determined based on the magnitude of the correlation coefficient. The dynamic correlation tensor between production activities and flow rate is used as an important reference to correct and optimize the estimated value of tailwater flow rate.

[0086] Preferably, when correcting the estimated tailwater flow rate based on the production activity-flow dynamic correlation tensor, a spatiotemporal weighted correction method is used, assuming the time series analysis model predicts... Time, spatial location The estimated tailwater flow rate at the location is ,exist The time scale interval corresponding to the time and Within the corresponding spatial location range, there exists a production activity type of The corresponding degrees of influence are as follows: Corrected tailwater flow rate estimate for:

[0087]

[0088] in, For production activity type exist Time, spatial location The impact of production activities on tailwater flow is obtained by statistically analyzing the changes in tailwater flow before and after the corresponding time and space points in historical data. This spatiotemporal weighted correction method can more accurately reflect the impact of actual production activities on tailwater flow in the spatiotemporal dimension, and significantly improve the accuracy of tailwater flow estimation.

[0089] like Figure 2As shown, a real-time estimation method for aquaculture wastewater flow is implemented through the following eight interconnected units: A basic environmental data acquisition and processing unit, responsible for comprehensively collecting basic environmental data related to wastewater discharge within the aquaculture area, processing and storing it according to a set data format, with its output connected to the input of a water flow initial state parameter monitoring unit; a water flow initial state parameter monitoring unit, used to perform real-time dynamic monitoring of the water flow initial state parameters at the aquaculture wastewater discharge outlet using preset monitoring equipment, and transmitting the monitored data to an interference factor analysis unit at set time intervals; an interference factor analysis unit, based on the basic environmental data and water flow initial state parameters, analyzes the interference factors in the aquaculture wastewater discharge process, establishes a preliminary correlation model between interference factors and wastewater flow, with its output connected to the input of a water flow state parameter correction unit; and a water flow state parameter correction unit, based on the preliminary correlation model between interference factors and wastewater flow, corrects the monitored water flow initial state parameters for interference factors, obtaining... The water flow state parameters, after interference correction, are transmitted to the preliminary flow estimation unit. The preliminary flow estimation unit, using a pre-defined flow estimation algorithm, takes the interference-corrected water flow state parameters and basic information such as the geometric dimensions of the aquaculture pond as input to perform a preliminary estimate of the instantaneous flow rate of the aquaculture wastewater. Its output is connected to the input of the time series analysis unit. The time series analysis unit performs time series analysis on the instantaneous flow data obtained from the preliminary estimate, using a time series analysis model to predict the trend of wastewater flow rate over time. Its output is connected to the input of the flow estimation optimization unit. The flow estimation optimization unit correlates and compares the wastewater flow rate trend predicted by the time series analysis model with the actual production activity data of the aquaculture area, further correcting and optimizing the estimated wastewater flow rate to obtain a more accurate real-time wastewater flow rate estimate, which is then transmitted to the data transmission unit. The data transmission unit transmits the real-time estimated wastewater flow rate value to the designated database of the aquaculture management system according to a preset data transmission protocol and format.

[0090] A real-time estimation method for aquaculture tailwater flow has many significant advantages and effectively overcomes many shortcomings of existing technologies.

[0091] The advantages of this method lie in its comprehensive and accurate data collection. Utilizing a global positioning system, topographic mapping instruments, and professional water quality testing instruments, it comprehensively collects basic environmental data such as the geographical location of the aquaculture pond, surrounding topography, initial water quality, and geometric dimensions. Latitude and longitude are accurate to six decimal places, and water quality parameters are precisely measured, providing a solid data foundation for estimating tailwater flow. In terms of real-time monitoring, Doppler current meters, electronic compasses, and image recognition technology are used to track changes in initial flow velocity, direction, and cross-sectional area in real time. Data transmission intervals are as short as 5 minutes, ensuring the timeliness and accuracy of the data.

[0092] This method effectively handles interference factors. By analyzing various complex interference factors such as wind, tides, and aquaculture biological activities, it establishes a correlation model and corrects water flow state parameters, significantly improving the reliability of flow estimation. It employs an advanced estimation algorithm based on hydrodynamic principles, comprehensively considering factors such as turbulence, outlet shape, water level difference, and local obstructions, resulting in a more accurate preliminary estimate of the instantaneous flow rate of the tailwater. The time series analysis and trend prediction functions, combining historical flow and real-time water flow parameters, can predict flow change trends, providing a basis for dynamically adjusting the estimated values. Furthermore, by correlating and comparing the predicted trends with actual production activity data, a correlation matrix is ​​constructed to optimize the estimated values, further improving estimation accuracy. Finally, the method ensures stable data transmission, using the Modbus protocol and JSON format to transmit the estimated values ​​to the aquaculture management system database, facilitating real-time access for managers and supporting scientific management.

[0093] Existing technologies for estimating tailwater flow have significant drawbacks. Traditional monitoring equipment is complex to install and expensive, making large-scale deployment in dispersed aquaculture ponds difficult and costly to maintain. Early estimation methods were simplistic and crude, considering only single or a few factors and ignoring the combined impact of multiple complex interference factors on tailwater flow, leading to large estimation errors. Moreover, existing technologies lack real-time performance and dynamic adaptability, making it difficult to track changes in aquaculture production activities and adjust estimation results in real time. They also fail to accurately reflect dynamic changes in tailwater flow under different seasons and weather conditions. The proposed "Real-time Estimation Method for Aquaculture Tailwater Flow" successfully overcomes these shortcomings, providing a more scientific, efficient, and accurate solution for tailwater flow estimation in aquaculture.

[0094] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time estimation of aquaculture tailwater flow, characterized in that, Includes the following steps: Collect basic environmental data related to wastewater discharge within aquaculture areas; The initial state parameters of the water flow at the aquaculture wastewater discharge outlet are monitored in real time using preset monitoring equipment. The initial flow velocity, initial flow direction and real-time change data of the cross-sectional area of ​​the water flow at the discharge outlet are obtained, and the monitored initial state parameters are transmitted to the data at set time intervals. Based on the collected basic environmental data and the monitored initial state parameters of the water flow, a systematic analysis was conducted on the interference factors affecting the flow rate during the discharge of aquaculture wastewater, including the assessment of the impact of wind, tides, and aquaculture biological activity on the water flow, and a preliminary correlation model between interference factors and wastewater flow rate was established. Based on the preliminary correlation model, the monitored initial state parameters of the water flow are corrected for interference factors. By using the set correction algorithm, the influence caused by interference factors is removed from the initial state parameters, and the water flow state parameters after interference correction are obtained. Using a pre-defined flow estimation algorithm, the flow state parameters after interference correction are input, and combined with the basic information of the geometric shape and size of the aquaculture pond, the instantaneous flow rate of the aquaculture tailwater is initially estimated. This flow estimation algorithm is based on the principle of hydrodynamics and is constructed by comprehensively considering factors such as flow velocity, cross-sectional area and flow characteristics. Time series analysis was performed on the instantaneous flow data obtained from the preliminary estimation. The time series analysis model was used to predict and analyze the trend of tailwater flow over time. The predicted value of tailwater flow over a future period was obtained through model calculation. The time series analysis model was trained and optimized based on historical flow data and current flow state parameters. The predicted tailwater flow rate trend of the time series analysis model is correlated and compared with the actual production activity data of the aquaculture area. The actual production activity data includes the feeding time of the aquaculture organisms, the water exchange operation time, and the start time of the aeration equipment. Based on the comparative analysis results, the estimated tailwater flow rate is further corrected and optimized to obtain the real-time estimated tailwater flow rate. The real-time estimated tailwater flow rate is transmitted to the designated database of the aquaculture management system according to the preset data transmission protocol and format. Aquaculture managers can obtain tailwater flow rate information in real time and perform data analysis and management decisions.

2. The method for real-time estimation of aquaculture tailwater flow rate according to claim 1, characterized in that, When establishing a preliminary correlation model between interference factors and tailwater flow rate, a multiple linear regression model was used. The model expression is as follows: ； in, Represents tailwater flow rate; These represent various interfering factors, including wind intensity, tidal force, and aquaculture activity. These are the coefficients corresponding to each interference factor; For constant terms, the coefficients and constant terms The parameters were determined through statistical analysis and parameter fitting of a large amount of basic environmental data covering different seasons and weather conditions, initial water flow parameters, and historical monitoring data of corresponding tailwater flow measured values.

3. The method for real-time estimation of aquaculture tailwater flow rate according to claim 2, characterized in that, When fitting the parameters of the aforementioned multiple linear regression model, the method of minimizing the sum of the absolute values ​​of the residuals is used to solve the problem, namely: For a set containing Historical monitoring data of sample number , let the first sample be... The actual measured tailwater flow rate in each sample was [value missing]. The tailwater flow rate predicted by the model is Construct the objective function By analyzing the coefficients and constant terms Find the partial derivatives, set them to zero, construct a system of equations to solve, and determine the solution. To achieve the minimum model parameters as well as The value of .

4. The method for real-time estimation of aquaculture tailwater flow rate according to claim 1, characterized in that, The flow estimation algorithm used is based on an improved extended form of the Darcy-Weisbach equation. Taking into account energy losses during the tailrace flow and the complex turbulent characteristics of the flow, the equation is modified. The modified equation is as follows: ； in, This refers to the tailwater flow rate; The comprehensive flow correction coefficient is related to various factors such as the degree of turbulence in the water flow, the roughness and shape of the discharge outlet; It is the cross-sectional area of ​​the water flow at the tailwater discharge outlet; It is the acceleration due to gravity; The head difference between the upstream and downstream sides of the tailwater discharge outlet; This is the friction factor, which is related to the Reynolds number of the water flow and the roughness of the pipe. The length of the tailrace flow path; The hydraulic diameter; This is the sum of local drag coefficients, related to the boundary conditions of various local obstruction structures at the discharge outlet, and is a comprehensive flow correction coefficient. The friction coefficient was determined through a combination of experimental testing and theoretical derivation. and the sum of local drag coefficients Based on the detailed structure of the discharge outlet and the real-time water flow status, the data is obtained through theoretical calculations or numerical simulations.

5. The method for real-time estimation of aquaculture tailwater flow rate according to claim 4, characterized in that, In determining the comprehensive flow correction coefficient An experimental testing platform was built to simulate the actual discharge scenario of aquaculture wastewater. Under different water flow velocities, turbulence levels, outlet roughness, and shapes, the wastewater flow rate was measured and compared with the flow rate calculated based on the modified extended form of the Darcy-Weisbach equation. A comprehensive flow correction coefficient was obtained through data fitting. With water flow velocity Parameters of water flow turbulence Roughness parameters of the discharge port and emission outlet shape parameters The functional relationship between them is: ； Based on the real-time monitored initial flow parameters and the geometry and roughness information of the discharge outlet, the comprehensive flow correction coefficient is determined through this functional relationship. The value of .

6. The method for real-time estimation of aquaculture tailwater flow rate according to claim 1, characterized in that, The time series analysis model used is an improved model based on the threshold autoregressive (TAR) model. This improved model is superior to the traditional TAR model. Based on the model, an external dynamic variable closely related to aquaculture production activities is introduced. Constructing an extended model TAR ( The model expression is: -Z ； in, express The tailwater flow rate at any given moment; For different threshold intervals, the autoregression order is denoted as ; and These are the autoregressive coefficients for the corresponding intervals; This is the threshold value; The delay order; external variables Coefficients in different intervals; For white noise sequences in different intervals, the autoregressive coefficients Threshold Delay order and external variable coefficients The aquaculture production activity data was determined by performing maximum likelihood estimation on historical tailwater flow data and corresponding aquaculture production activity data. It was obtained by extracting and organizing production record data from the aquaculture management system.

7. The method for real-time estimation of aquaculture tailwater flow rate according to claim 6, characterized in that, In the above When estimating the parameters of the model, the Akaike Information Criterion (AICc) is used as the key criterion for model selection. Adjustments are made based on the finite number of samples, and different autoregressive orders are iterated through. Threshold and delay order The combinations are used to calculate the AIC of the model under each combination. The optimal model parameters are selected based on the combination of model parameters that minimizes the AICC value. The formula for calculating the AICC value is as follows: ； in, The number of samples; It is the sum of squared residuals; This represents the number of parameters to be estimated in the model.

8. The method for real-time estimation of aquaculture tailwater flow rate according to claim 1, characterized in that, When comparing and contrasting the predicted tailwater flow rate trend of the time series analysis model with the actual production activity data of the aquaculture area, a dynamic correlation tensor between production activity and flow rate is constructed. The first dimension of the tensor represents different types of production activities, including feeding, water exchange, and oxygenation. The second dimension represents different time scale divisions, from short-term periods to long-term cycles; the third dimension represents the degree of influence of different production activities on tailwater flow at different spatial locations under the corresponding time scale. This degree of influence is obtained through three-dimensional correlation analysis of the occurrence time and spatial location of production activities and the changes in tailwater flow at corresponding spatiotemporal points in historical data. The specific calculation method is as follows: for each type of production activity, at different time scales, calculate the three-dimensional cross-correlation coefficient between the occurrence time of the activity at different spatial locations and the change time of tailwater flow at the corresponding spatial locations. The strength of the influence is determined based on the magnitude of the correlation coefficient. The production activity-flow dynamic correlation tensor is used as a reference to correct and optimize the estimated value of tailwater flow.

9. The method for real-time estimation of aquaculture tailwater flow rate according to claim 8, characterized in that, When correcting the tailwater flow estimate based on the aforementioned production activity-flow dynamic correlation tensor, a spatiotemporal weighted correction method is used, assuming the time series analysis model predicts... Time, spatial location The estimated tailwater flow rate at the location is ,exist The time scale interval corresponding to the time and Within the corresponding spatial location range, there exists a production activity type of The corresponding degrees of influence are as follows: Corrected tailwater flow rate estimate for: ； in, For production activity type exist Time, spatial location The impact of production activities on tailwater flow rate is obtained by statistical analysis of the changes in tailwater flow rate before and after the corresponding time and space points in historical data.

Images