Water quality monitoring method for aquaculture

Abstract

The invention discloses a water quality monitoring method for aquaculture, which comprises the following steps of: selecting aquatic organisms which are sensitive to water quality change, short in life cycle and high in breeding speed, collecting the aquatic organisms, pretreating the aquatic organisms, temporarily culturing the aquatic organisms in a similar environment, constructing a small monitoring water body unit, arranging the small monitoring water body unit in an aquaculture water area, and arranging an independent water inlet and outlet device. Meanwhile, physical and chemical parameter real-time monitoring sensors and an underwater camera are installed, then physical and chemical parameters and biological behavior data are collected in real time and subjected to correlation analysis, a quantitative relation between the physical and chemical parameters and the biological behavior data is established through a machine learning algorithm, an early warning mechanism is set accordingly, and finally a biological sample is updated regularly, and a water quality sensor is calibrated. Biological monitoring equipment and related devices of the monitoring unit are maintained, and the monitoring accuracy and effectiveness are ensured. According to the invention, by selecting sensitive organisms, constructing a monitoring unit, collecting and analyzing data in real time and setting an early warning mechanism, the problem that the organisms have response hysteresis to the water quality change can be effectively solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of aquaculture supervision, and specifically relates to a method for monitoring the water quality of aquaculture. Background Art

[0002] Aquaculture, as a production activity of artificially breeding, cultivating and harvesting aquatic animals and plants, the growth of aquatic organisms is deeply affected by the aquaculture environment. Different environmental values have different effects on the growth of different fish species. Therefore, effectively monitoring and managing the aquaculture environment is extremely crucial for aquaculture, directly related to the aquaculture results and the healthy growth of aquatic organisms.

[0003] Most of the existing methods for monitoring the water quality of aquaculture rely on regular water sample collection and laboratory analysis to monitor the physical and chemical parameters of water quality. The water samples are taken once a week or even once a month, making it difficult to detect sudden water quality changes in a timely manner. Moreover, the existing methods for monitoring the water quality of aquaculture often ignore the real-time correlation between biological behaviors and water quality changes, and cannot judge the water quality status from the immediate biological reactions, nor can they capture the impact of water quality fluctuations on organisms in the first time. As a result, when water quality problems are found, the organisms may have already shown serious adverse reactions, delaying the regulation time and causing losses to aquaculture. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for monitoring the water quality of aquaculture to solve the technical problems that the existing methods for monitoring the water quality of aquaculture cannot capture the instantaneous changes of water quality in a timely manner, and it is difficult to immediately judge the immediate impact of water quality on organisms through biological reactions, showing obvious hysteresis.

[0005] The technical scheme adopted by the present invention is as follows:[[]]

[0006] A method for monitoring the water quality of aquaculture, comprising the following steps:[[]]

[0007] First, it is necessary to select aquatic organisms that are sensitive to water quality changes, have a short life cycle and a fast reproduction rate. After collection, they are pretreated and temporarily raised in a similar environment.

[0008] Then, a small monitoring water body unit is constructed, set in the aquaculture water area, equipped with an independent water inlet and outlet device, and at the same time, physical and chemical parameter real-time monitoring sensors and underwater cameras are installed.

[0009] Subsequently, physical and chemical parameter and biological behavior data are collected in real time, and correlation analysis is carried out. A quantitative relationship between the two is established using machine learning algorithms, and an early warning threshold is set accordingly. Once the data exceeds the threshold, an early warning mechanism for sending out an early warning is triggered.

[0010] Finally, the biological samples are updated regularly, the water quality sensors are calibrated, and the relevant devices of the biological monitoring equipment and the monitoring unit are maintained to ensure the accuracy and effectiveness of the monitoring.

[0011] Among them, the pre-treatment is to temporarily raise the organisms in a pre-culture environment similar to the aquaculture environment for a period of time, generally 1-2 days, to allow the organisms to adapt to the new environment, and at the same time record their physiological and behavioral characteristics under normal conditions.

[0012] Among them, the physical and chemical parameter real-time monitoring sensors include a dissolved oxygen sensor, a pH sensor, a temperature sensor, and a conductivity sensor. The dissolved oxygen sensor is based on the electrochemical principle and can real-time monitor the dissolved oxygen content in water, helping aquaculturists take oxygen-increasing measures in a timely manner to ensure the respiration of aquatic organisms. The pH sensor measures the acidity and alkalinity by using the sensitive characteristic of the glass electrode to hydrogen ions. Since the pH value affects the activity of biological enzymes and the acid-base balance, the water body acidity and alkalinity can be adjusted accordingly. The temperature sensor relies on the principle of thermistor or thermocouple and, according to the influence of water temperature on the metabolism of organisms, helps aquaculturists control the water temperature to promote the growth of organisms. The conductivity sensor calculates the conductivity by measuring the ionic current in water, can indirectly reflect the salinity and detect water body pollution, and real-time monitors the change of water quality.

[0013] Among them, the physical and chemical parameters cover dissolved oxygen, pH value, temperature, conductivity, nutrient salts, and heavy metal ions. These data can timely reflect the basic state of water quality. The biological behavior data includes swimming behavior, aggregation behavior, feeding behavior, and stress response behavior. These behavior changes directly reflect the immediate response of organisms to water quality changes.

[0014] Among them, in correlation analysis and machine learning, by establishing a mathematical model, the quantitative relationship between the changes in water quality parameters and the changes in biological behaviors is analyzed, and then machine learning algorithms are used to train a large amount of data.

[0015] Among them, the mathematical models include multiple linear regression models, logistic regression models, and time series models. The machine learning algorithms include decision tree algorithms, random forest algorithms, and artificial neural networks. The mathematical models and machine learning algorithms work together to determine the changes in water quality and the biological response patterns.

[0016] Among them, the early warning mechanism includes setting an early warning threshold and issuing an early warning when the threshold is exceeded. The setting of the early warning threshold includes associating and analyzing data, combining the biological tolerance limit and dynamically considering environmental factors. By associating and analyzing data, the internal relationship between water quality parameters and biological behaviors is accurately explored. Combining the biological tolerance limit, the bottom line of water quality safety is clarified. Then, considering environmental factors dynamically, the adjusted threshold is determined to fit the complex and changeable actual aquaculture scenario. The issuing of an early warning when the threshold is exceeded includes real-time data comparison, triggering early warning condition judgment, and sending early warning signals. Through real-time data comparison, the newly collected water quality physicochemical parameters and biological behavior data are continuously compared with the pre-set early warning threshold hour by hour to ensure timely discovery of data changes. Then, the triggering of early warning condition judgment is when the water quality parameters reach or exceed the threshold, and the biological behaviors show corresponding abnormal change patterns, it is determined that the early warning conditions are met, and at the same time, the early warning signal will be triggered and sent.

[0017] Among them, the early warning signal sending methods include sounding an alarm in the farm monitoring room, then using mobile phone text messages and APP push messages to inform the farmers, and at the same time highlighting abnormal data and early warning information on the monitoring software interface to comprehensively remind relevant personnel to take measures in time to deal with the water quality deterioration situation.

[0018] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are:

[0019] In the present invention, due to the coordinated cooperation of selecting sensitive organisms, constructing monitoring units, real-time collecting and analyzing data, and setting up an early warning mechanism, by selecting sensitive aquatic organisms and using their characteristics of quickly responding to water quality changes, combined with real-time monitoring of physicochemical parameters and biological behavior data, a quantitative relationship is established by means of association analysis and machine learning algorithms to accurately predict the impact of water quality changes on organisms. Thus, it can be detected in time when the water quality just shows abnormalities. At the same time, biological samples are updated regularly and equipment is calibrated to ensure that the monitoring system always maintains high sensitivity and accuracy, effectively solving the problem of the lag in the reaction of organisms to water quality changes, enabling farmers to take measures in the first time to deal with water quality deterioration and ensuring the healthy growth of aquaculture organisms. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the method steps of the present invention;

[0021] Figure 2 It is a flowchart of the implementation of the early warning mechanism in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0023] Reference Figure 1 - Figure 2 , a water quality monitoring method for aquaculture, comprising the following steps:

[0024] First, it is necessary to select aquatic organisms that are sensitive to water quality changes, have a short life cycle and a fast reproduction rate. After collection, they are pretreated and temporarily raised in a similar environment;

[0025] Next, a small monitoring water body unit is constructed and set in the aquaculture water area, equipped with an independent inlet and outlet device, and at the same time, real-time monitoring sensors for physical and chemical parameters and an underwater camera are installed;

[0026] Subsequently, physical and chemical parameters and biological behavior data are collected in real time, and correlation analysis is carried out. Using machine learning algorithms, a quantitative relationship between the two is established, and an early warning threshold is set accordingly. Once the data exceeds the threshold, an early warning mechanism is triggered;

[0027] Finally, the biological samples are updated regularly, the water quality sensors are calibrated, and the relevant devices of the biological monitoring equipment and the monitoring unit are maintained to ensure the accuracy and effectiveness of the monitoring.

[0028] Specifically, by selecting aquatic organisms that are sensitive to water quality changes, these organisms can respond more quickly to small changes in water quality. When the water quality begins to deteriorate, the changes in biological behavior can be captured in a timely manner, so as to issue an early warning in advance, avoiding the discovery of water quality anomalies only when obvious health problems or deaths occur in the cultured organisms. Constructing a small monitoring water body unit and installing real-time monitoring sensors for physical and chemical parameters and an underwater camera can simultaneously obtain water quality physical and chemical parameters and biological behavior data. The sensors can provide real-time feedback on the changes in key parameters such as temperature, pH value, and dissolved oxygen, while the underwater camera can synchronously observe the immediate behavior of the organisms. This real-time multi-dimensional monitoring can quickly discover the correlation between water quality changes and biological reactions. Using correlation analysis and machine learning algorithms to establish a quantitative relationship between physical and chemical parameters and biological behavior can more accurately predict the impact of water quality changes on organisms. Based on this relationship, when the water quality parameters change, the reaction of the organisms can be quickly predicted according to the model. Regularly updating biological samples and calibrating equipment ensure that the monitoring system can adapt to the dynamic changes of the aquaculture environment. Over time, the sensitivity of organisms to water quality may change, and the water quality sensors may also deviate. Through regular updates and calibrations, the accuracy of the monitoring can be guaranteed, and the sensitive monitoring of biological reactions can be maintained at all times, avoiding missing the early reaction of organisms to water quality changes due to systematic errors.

[0029] Reference Figure 1As shown, the pretreatment is to temporarily raise the organisms in a pre-culture environment similar to the aquaculture environment for a period of time, generally 1-2 days, to allow the organisms to adapt to the new environment. At the same time, record their physiological and behavioral characteristics under normal conditions. The pretreatment provides a transition stage for the organisms to transition from the collection environment to the pre-culture environment similar to the aquaculture environment. This process can reduce the stress response of the organisms caused by sudden environmental changes. Through 1-2 days of temporary raising, the organisms can gradually adapt to the new water quality parameters, water flow conditions, etc., so that when they enter the aquaculture environment or the monitored water body unit subsequently, they can integrate faster, reduce behavioral and physiological abnormalities caused by environmental discomfort, and thus ensure that the subsequent monitoring data can more truly reflect the impact of water quality changes on organisms.

[0030] Refer to Figure 1 As shown, the physical and chemical parameter real-time monitoring sensors include a dissolved oxygen sensor, a pH sensor, a temperature sensor, and a conductivity sensor. The dissolved oxygen sensor is based on the electrochemical principle and can real-time monitor the dissolved oxygen content in water, helping aquaculturists take oxygen-increasing measures in a timely manner to ensure the respiration of aquatic organisms and avoid health problems or even death of organisms due to hypoxia. The pH sensor measures the acidity and alkalinity using the sensitive characteristic of the glass electrode to hydrogen ions. Since the pH value affects the activity of biological enzymes and the acid-base balance, the water body acidity and alkalinity can be adjusted accordingly to maintain a suitable chemical environment for the survival and growth of organisms. The temperature sensor relies on the principle of thermistor or thermocouple. According to the influence of water temperature on the metabolism of organisms, it helps aquaculturists control the water temperature to promote the growth of organisms. The conductivity sensor calculates the conductivity by measuring the ionic current in water, can indirectly reflect the salinity and detect water body pollution, real-time monitor the water quality changes, and make response strategies in advance to ensure the stability of the aquaculture environment and the health of organisms.

[0031] Refer to Figure 1 As shown, the physical and chemical parameters cover dissolved oxygen, pH value, temperature, conductivity, nutrient salts, and heavy metal ions. These data can timely reflect the basic state of water quality. The biological behavior data includes swimming behavior, aggregation behavior, feeding behavior, and stress response behavior. These behavior changes directly reflect the immediate response of organisms to water quality changes. The physical and chemical parameters can timely show the basic state of water quality from the chemical and physical aspects, helping to judge whether the water quality meets the survival needs of aquaculture organisms. The biological behavior data can directly reflect the immediate response of organisms to water quality changes, concretize the abstract changes in water quality. The two cooperate with each other to provide comprehensive and accurate water quality information for aquaculturists, helping them quickly detect water quality anomalies and take effective measures in a timely manner to ensure the stability of the aquaculture environment and the healthy growth of organisms.

[0032] Refer to Figure 1 As shown, in correlation analysis and machine learning, by establishing a mathematical model, analyze the quantitative relationship between the changes in water quality parameters and the changes in biological behaviors, and then use machine learning algorithms to train a large amount of data.

[0033] Specifically, in aquaculture water quality monitoring, by establishing a mathematical model, the complex quantitative relationship between water quality parameters and biological behavior changes can be clearly sorted out, providing a quantitative basis for the correlation between water quality and biological reactions. By training with machine learning algorithms on massive data, the system can automatically identify the hidden complex patterns and laws between the two. This not only greatly improves the accuracy of water quality monitoring and prediction but also helps to anticipate in advance the impact of water quality changes on organisms, assisting farmers in adjusting aquaculture strategies in a timely manner, effectively preventing losses caused by water quality deterioration, and ensuring the stable development of the aquaculture industry.

[0034] Refer to Figure 1 As shown, the mathematical models include multiple linear regression models, logistic regression models, and time series models, and the machine learning algorithms include decision tree algorithms, random forest algorithms, and artificial neural networks. The mathematical models and machine learning algorithms work together to determine water quality changes and biological reaction patterns. The multiple linear regression model quantifies the impact of each water quality parameter on biological behavior by estimating regression coefficients and further predicts changes in biological behavior based on changes in water quality parameters. The logistic regression model can take water quality parameters as inputs and output the probability that biological behavior belongs to a certain category (normal or abnormal), predicting the probability of fish showing stress behavior under specific water quality conditions. The time series model is used to analyze time series data, capture the changing patterns of water quality parameters and biological behavior over time, including trends, seasonality, etc., and further helps to analyze the quantitative relationship between the time series changes of dissolved oxygen and the time series changes of fish swimming behavior. The decision tree algorithm can visually display the decision rules between water quality parameters and biological behavior and can help farmers quickly understand the relationship between water quality changes and biological behavior in practical applications. The random forest algorithm is an ensemble learning algorithm based on decision trees. When predicting changes in biological feeding behavior caused by water quality changes, the random forest can comprehensively consider the combined effects of multiple water quality parameters and provide more reliable prediction results. The artificial neural network can automatically learn complex non-linear relationships. In aquaculture, the relationship between water quality parameters and biological behavior is often non-linear. The artificial neural network can fit this complex relationship through the connection of multiple neurons and activation functions. They work together to accurately analyze water quality changes and deeply understand biological reaction patterns, providing key support for timely warning of water quality anomalies and scientifically regulating the aquaculture environment, and strongly ensuring the healthy and stable development of aquaculture.

[0035] Refer to Figure 2As shown, the early warning mechanism includes setting an early warning threshold and issuing an early warning when the threshold is exceeded. Setting the early warning threshold includes associating and analyzing data, combining the biological tolerance limit and considering environmental factors dynamically. By associating and analyzing data, the internal relationship between water quality parameters and biological behavior is accurately explored. By combining the biological tolerance limit, the bottom line of water quality safety is clarified. Then, considering environmental factors dynamically, the adjusted threshold is determined to fit the complex and changeable actual aquaculture scenario. Issuing an early warning when the threshold is exceeded includes real-time data comparison, triggering early warning condition judgment, and sending early warning signals. Through real-time data comparison, the newly collected physical and chemical water quality parameters and biological behavior data are continuously compared with the pre-set early warning threshold hour by hour to ensure timely detection of data changes. Then, triggering early warning condition judgment is to determine that the early warning condition is met when the water quality parameters reach or exceed the threshold and the biological behavior shows a corresponding abnormal change pattern. At the same time, the early warning signal will be triggered and sent. By associating and analyzing data, the internal relationship between water quality and biological behavior is sorted out, providing a scientific basis for early warning. By combining the biological tolerance limit, the water quality red line for ensuring the survival of organisms is established. Considering dynamic environmental factors, the threshold is adapted to the changeable aquaculture environment. In the stage of exceeding the threshold, real-time data comparison ensures timely capture of abnormalities in water quality and biological behavior, accurately judges the early warning trigger conditions, and once satisfied, quickly sends a signal, enabling farmers to detect water quality hidden dangers in advance, intervene in a timely manner, prevent losses to aquaculture caused by water quality deterioration, and maintain the stability of the aquaculture environment.

[0036] Refer to Figure 2 As shown, the early warning signal sending methods include sounding an alarm in the farm monitoring room, then using mobile phone text messages and APP push messages to inform farmers, and at the same time highlighting abnormal data and early warning information on the monitoring software interface, comprehensively reminding relevant personnel to take measures in a timely manner to deal with the water quality deterioration situation. Through the multi-channel early warning signal sending method, relevant personnel can be covered comprehensively and without dead ends. Sounding an alarm in the farm monitoring room allows on-site personnel to detect water quality abnormalities immediately and take preliminary response measures. Mobile phone text messages can break through spatial limitations and transmit information to farmers immediately, enabling them to know in time even if they are not on site. APP push messages provide a convenient interaction channel for farmers to further view details. Highlighting abnormal data and early warning information on the monitoring software interface helps staff quickly understand the water quality deterioration situation and provides an intuitive basis for subsequent decision-making.

[0037] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring water quality for aquaculture, characterized in that: The following steps are involved: First, we need to select aquatic organisms that are sensitive to water quality changes, have a short life cycle and a fast reproduction rate, collect them, pre-treat them, and temporarily raise them in a similar environment; Then, a small water monitoring unit is constructed and set up in the aquaculture waters, equipped with independent water inlet and outlet devices, and equipped with real-time monitoring sensors for physical and chemical parameters and underwater cameras; Then, the physical and chemical parameters and biological behavior data are collected in real time, and correlation analysis is performed. The quantitative relationship between the two is established using machine learning algorithms, and the warning threshold is set accordingly. Once the data exceeds the threshold, an early warning mechanism will issue an early warning; Finally, regularly update biological samples, calibrate water quality sensors, and maintain biological monitoring equipment and related devices of the monitoring unit to ensure the accuracy and effectiveness of monitoring.

2. A method for monitoring water quality for aquaculture according to claim 1, characterized in that: Pretreatment is to temporarily culture the organisms in a pre-culture environment similar to the breeding environment for 1-2 days to allow the organisms to adapt to the new environment, while recording their physiological and behavioral characteristics under normal conditions.

3. A method for monitoring water quality for aquaculture according to claim 1, characterized in that: Real-time monitoring sensors for physical and chemical parameters include dissolved oxygen sensors, pH sensors, temperature sensors, and conductivity sensors. The dissolved oxygen sensor is based on electrochemical principles and can monitor the dissolved oxygen content in water in real time, helping farmers to take oxygenation measures in a timely manner to ensure the breathing of aquatic organisms. The pH sensor uses the glass electrode's sensitivity to hydrogen ions to measure pH. Since the pH value affects the activity of biological enzymes and the acid-base balance, the pH of the water can be adjusted accordingly. The temperature sensor uses the principle of thermistors or thermocouples to help farmers control water temperature to promote biological growth based on the effect of water temperature on biological metabolism. The conductivity sensor calculates conductivity by measuring the ion current in water, which can indirectly reflect salinity and detect water pollution, and monitor water quality changes in real time.

4. A method for monitoring water quality for aquaculture as claimed in claim 1, characterized in that: The physical and chemical parameters include dissolved oxygen, pH value, temperature, conductivity, nutrients and heavy metal ions. These data can timely reflect the basic state of water quality. The biological behavior data include swimming behavior, aggregation behavior, feeding behavior and stress response behavior. These behavioral changes intuitively reflect the immediate response of organisms to changes in water quality.

5. A method for monitoring water quality for aquaculture as claimed in claim 1, characterized in that: In correlation analysis and machine learning, mathematical models are established to analyze the quantitative relationship between changes in water quality parameters and changes in biological behavior, and then machine learning algorithms are used to train large amounts of data.

6. A method for monitoring water quality for aquaculture as claimed in claim 5, characterized in that: The mathematical model includes a multiple linear regression model, a logistic regression model and a time series model, and the machine learning algorithm includes a decision tree algorithm, a random forest algorithm and an artificial neural network. The mathematical model and the machine learning algorithm work together to determine changes in water quality and biological response patterns.

7. A method for monitoring water quality for aquaculture as claimed in claim 1, characterized in that: The early warning mechanism includes setting early warning thresholds and issuing early warnings when thresholds are exceeded. Setting early warning thresholds includes correlating and analyzing data, combining biological tolerance limits and considering the dynamics of environmental factors. The intrinsic connection between water quality parameters and biological behavior is accurately explored through correlating and analyzing data. The bottom line of water quality safety is clarified in combination with biological tolerance limits. Then, the dynamics of environmental factors are considered to determine the threshold for adjustment to suit the complex and changeable actual breeding scenarios. The issuing of early warnings when thresholds are exceeded includes real-time data comparison, judgment of triggering early warning conditions and sending of early warning signals. Real-time data comparison is performed by continuously comparing newly collected water quality physicochemical parameters and biological behavior data with pre-set early warning thresholds hour by hour to ensure timely detection of data changes. Then, when the water quality parameters reach or exceed the thresholds and the biological behavior presents a corresponding abnormal change pattern, it is determined that the early warning conditions are met and the early warning signal is triggered and sent.

8. A method for monitoring water quality for aquaculture as claimed in claim 7, characterized in that: The warning signal is sent by sounding an alarm in the farm's monitoring room, then notifying farmers via text messages and APP push messages. At the same time, abnormal data and warning information are highlighted on the monitoring software interface, reminding relevant personnel in all aspects to take timely measures to deal with the deterioration of water quality.

Images