Mandarin fish hydrogenation breeding method and system based on historical data analysis

Abstract

The present invention belongs to the field of data processing technology and proposes a hydrogen-filled aquaculture method and system for mandarin fish based on historical data analysis. Specifically, the method comprises: first, installing a data collector in the aquaculture pond, the data collector including a camera and a hydrogen concentration monitor; calculating activity using video data obtained by the camera; then obtaining hydrogen concentration using the hydrogen concentration monitor; and finally, using a binary tuple consisting of activity and hydrogen concentration as a monitoring array; calculating the utilization loss risk based on the real-time monitoring array; and adjusting the output rate of the hydrogen pump based on the utilization loss risk. This method effectively quantifies and extracts the risk of hydrogen waste, avoids hydrogen waste and utilization instability caused by traditional fixed hydrogen release rates, ensures the continuity and stability of the hydrogen effect, and effectively addresses activity fluctuations caused by changes in environmental variables, thereby achieving the goal of improving hydrogen efficiency and mandarin fish aquaculture quality.

Description

Technical Field

[0001] The present invention belongs to the technical field of aquaculture, and in particular relates to a method and system for hydrogen-filled culture of mandarin fish based on historical data analysis. Background Art

[0002] Mandarin fish is a highly valuable aquatic animal among freshwater fish. However, mandarin fish is highly sensitive to the water environment, including indicators such as dissolved oxygen, nitrite and ammonia nitrogen in the water environment. Especially under high-density breeding conditions, it is prone to health problems caused by oxidative stress, including decreased immunity and frequent diseases. In addition, mandarin fish seedlings are more sensitive to the water environment and have weaker resistance, resulting in a higher mortality rate.

[0003] Nowadays, hydrogen is often used to assist in environmental regulation in the field of aquaculture because hydrogen has selective antioxidant capacity. When used in aquaculture, it can neutralize harmful free radicals in aquatic animals, reduce the toxicity of nitrites, nitrates or certain heavy metals in the water, and reduce the production of ammonia nitrogen, thereby effectively improving the health of aquatic animals and enhancing their immunity. It can significantly improve the health and survival rate of aquatic products. This hydrogen-assisted aquaculture method is highly consistent with the characteristics of mandarin fish, which have weak resistance and high sensitivity to the water environment, and significantly improves the output rate and overall breeding quality.

[0004] However, in aquaculture, mandarin fish have a high oxygen demand, so farms are typically equipped with oxygen aeration or stirring machinery to maintain dissolved oxygen. These aeration or stirring machinery accelerates the escape of hydrogen, resulting in a loss of hydrogen utilization. Furthermore, mandarin fish exhibit significant variations in activity under irregular environmental variables such as water temperature, light intensity, and feeding patterns, leading to even more unstable hydrogen utilization in the water. This is because mandarin fish feed significantly more under higher water temperatures and sunlight, leading to a rapid increase in their metabolic rate, which in turn increases free radical production in the body and dramatically increases hydrogen consumption. This change in activity also accelerates the escape of hydrogen from the water.

[0005] The existing technology usually adopts a fixed hydrogen release rate, which obviously cannot take into account the loss of hydrogen utilization rate under the high fluctuation of mandarin fish activity during mandarin fish farming. Therefore, there is an urgent need for a mandarin fish hydrogenation farming method and system based on historical data analysis. Summary of the Invention

[0006] The purpose of the present invention is to propose a method and system for hydrogen-filled culture of mandarin fish based on historical data analysis, so as to solve one or more technical problems existing in the prior art and at least provide a beneficial option or create conditions.

[0007] In order to achieve the above object, according to one aspect of the present invention, a method for hydrogen-filled culture of mandarin fish based on historical data analysis is provided, the method comprising the following steps:

[0008] A data collector is installed in the aquaculture pond, which includes a camera and a hydrogen concentration monitor. The activity is calculated using video data obtained from the camera. The hydrogen concentration is obtained through the hydrogen concentration monitor. The binary group consisting of the activity and hydrogen concentration is used as a monitoring array. The utilization loss risk is calculated based on the monitoring array obtained in real time. The output rate of the hydrogen pump is adjusted according to the utilization loss risk.

[0009] Furthermore, a method for installing a data collector in the aquaculture pond, wherein the data collector includes a camera and a hydrogen concentration monitor, is as follows: a water inlet of a circulating water treatment device is provided in the center of the aquaculture pond, and a hydrogen concentration monitor is arranged in the water inlet; the hydrogen concentration monitor is any one of an infrared gas analyzer, a semiconductor hydrogen sensor, or a gas chromatograph; and cameras are installed in the center and at the four corners of the aquaculture pond, respectively, at a height of 0.5 to 1 meter below the water surface.

[0010] When installing the camera, avoid the reflection of the water surface and adjust the angle so that the camera can capture the water layer where the mandarin fish are active.

[0011] The reason for installing a camera in the center and at each of the four corners of the aquaculture pond is that the water flow in the four corners and the center of the aquaculture pond is slower, making it easier for mandarin fish to stay and move around.

[0012] Preferably, the camera is a high-definition underwater camera with infrared night vision function so that it can be used normally in low-light or nighttime environments.

[0013] Furthermore, the method for calculating activity using video data obtained by a camera is as follows: monitoring the water body of the breeding pond through a camera and obtaining video data in real time; identifying individual mandarin fish from each frame of the video data through a target detection method, and obtaining the movement path of the individual mandarin fish using a tracking algorithm, wherein the constraint time of the movement path is 2 seconds to 20 seconds; the ratio of the total length of the movement path to the constraint time is the corresponding activity of the individual mandarin fish; wherein the target detection method includes any one of YOLO, SSD or Faster R-CNN; and the tracking algorithm includes any one of SORT, DeepSORT or KLT tracking algorithm.

[0014] Furthermore, the method of using the binary group consisting of activity and hydrogen concentration as a monitoring array is: setting a time interval as the feedback interval TG, the feedback interval value range is 2 minutes to 10 minutes, and obtaining a monitoring array every TG, the monitoring array is a binary group consisting of hydrogen concentration and activity.

[0015] The monitoring array data stored therein has a timestamp, that is, the monitoring array obtained at any moment should be marked with the time point when the tuple is obtained, so as to be used for retrospective analysis of subsequent historical data.

[0016] Furthermore, the method for calculating the utilization loss risk based on the monitoring array obtained in real time is: set a time period as the monitoring period KETW, KETW∈[0.25,2] hours; within the current monitoring period, the moment when the activity has an extreme value is recorded as a first-class extreme point, and the moment when the hydrogen concentration has an extreme value is recorded as a second-class extreme point. The number of moments between any first-class extreme point and the first second-class extreme point with the same polarity obtained by reverse time search is recorded as the concentration time lag interval selected value, and the median value of each concentration time lag interval selected value is recorded as the concentration time lag interval GOMT; the same polarity in the second-class extreme point with the same polarity of the first-class extreme point means that when the corresponding extreme value of the first-class extreme point is a maximum value, the corresponding extreme value of the second-class extreme point with the same polarity is also a maximum value, and when the corresponding extreme value of the first-class extreme point is a minimum value, the corresponding extreme value of the second-class extreme point with the same polarity is also a minimum value.

[0017] The calculation principle of the concentration lag interval is that when the activity of mandarin fish changes significantly, the hydrogen consumption will also change. However, the speed of gas escaping in the water varies in different areas due to the density of mandarin fish. Therefore, there is a certain delay in the change of hydrogen concentration in some local areas. By studying the time difference between the moments of rapid change in activity and hydrogen concentration over a period of time, it is possible to effectively capture the loss of hydrogen utilization due to fluctuations in activity.

[0018] Calculate the average value of the hydrogen concentration at any moment and its counterclockwise GOMT moments as the hydrogen dynamic mean Medrc at that moment. If the number of hydrogen concentrations in the counterclockwise direction at any moment is less than the concentration time lag interval, then record the hydrogen concentration corresponding to that moment and the average value of all hydrogen concentrations in the counterclockwise direction as the hydrogen dynamic mean at that moment. Standardize the activity and hydrogen concentration in the monitoring array sequence to form a normalized feature sequence. Record the activity of the normalized feature sequence as the activity normalization value Nrmac. Record the average and range of all hydrogen concentrations in the normalized feature sequence as the first concentration mean Fstra and the first concentration range Fsonr, respectively. For any moment, calculate the utilization loss risk Utzas based on the activity normalization value and the hydrogen dynamic mean: Utzas=Nrmac -1 / 3 ·Fsonr -1 ·(Medrc-Fstra).

[0019] The default standardization process is minmax standardization. The normalized feature sequence is a sequence constructed with two-tuples consisting of the activity and hydrogen concentration after each standardization process as elements.

[0020] Since the above-mentioned hydrogen dynamic mean is obtained by calculating the time difference between the activity and the extreme point of hydrogen concentration, the concentration time lag interval is obtained by the hydrogen concentration obtained at any moment and the average of the hydrogen concentrations at the previous moments, and the concentration time lag interval accurately shows the lag effect of the hydrogen concentration as the activity of the mandarin fish changes. Therefore, the loss of hydrogen utilization under the fluctuation of activity can be accurately characterized according to the hydrogen dynamic mean at any moment, which helps to quickly identify the real-time utilization of hydrogen so as to quickly adjust the release rate of hydrogen. However, in the data scenario with a limited number of farmed mandarin fish individuals, the activity is highly unstable. It is somewhat rough to rely solely on the attribute of the hydrogen dynamic mean to measure the utilization loss risk. It lacks the phenomenon of considering the interaction between the activity noise data and the measured hydrogen concentration, so that the loss risk obtained is often easily magnified or reduced, and the result is inaccurate and unstable. In order to obtain a more accurate utilization loss risk, the present invention proposes a more preferred solution;

[0021] Furthermore, the method for calculating the utilization loss risk based on the monitoring array obtained in real time is as follows: set a time period as the monitoring period KETW, KETW∈[0.5,3] hours; record the average value of the concentration equilibrium risk in the monitoring period as the concentration risk mean Cerik, record the monitoring arrays obtained at each moment in the current monitoring period to form a monitoring array sequence, record the activity as Vilve, and record the hydrogen concentration as Hygen; draw a time series curve for all hydrogen concentrations in the monitoring array sequence; wherein the time series curve is drawn using a linear regression algorithm or an exponential smoothing algorithm; starting from any moment, search the time series curve in the reverse time direction for the first extreme point, which includes the points as the maximum and minimum points on the time series curve; the number of moments between the obtained extreme point and any moment is the fitting delay Cceva, wherein the first monitoring moment is defaulted to be the extreme point, and the absolute value of the difference between the hydrogen concentration value at any moment and its previous moment is defined as the concentration change value Fseca, and the concentration equilibrium risk Tequm is calculated based on the fitting delay and the concentration change value:

[0022] ;

[0023] Among them, i1, i2 are cumulative variables, Fseca i1 is the concentration change value at the i1th moment in the counterclockwise direction at the current moment, Fseca i2 , Hygen i2 are the concentration change value and hydrogen concentration at the i2th moment counterclockwise from the current moment, respectively. Fseca0 and Hygen0 are the concentration change value and hydrogen concentration at the current moment, respectively.

[0024] The calculation principle of concentration equilibrium risk is to calculate the weighted moving average of the current hydrogen concentration value by using the concentration change delay value and concentration change value over a period of time. This method can quickly and effectively respond to sudden changes in hydrogen concentration while maintaining a certain degree of stability. It can also extract useful trend information from time series data containing a lot of noise, and accurately characterize the real-time hydrogen utilization rate as its activity changes.

[0025] Let the mean and standard deviation of all activities in the monitoring array sequence be the first activity mean Fstvm and the first activity fluctuation value Fsfva, respectively. Calculate the activity risk threshold Acths = max(0, Fstvm - Fsfva - Vilve) at any moment; here Vilve is the activity corresponding to any moment; calculate the concentration equilibrium risk threshold Azres = max(0, Tequm - Cerik) at any moment, where max() is the maximum value function; here Tequm is the concentration equilibrium risk corresponding to any moment;

[0026] The activity and concentration equilibrium risk at any moment are formed into a binary pair and recorded as a derivation array. The sequence formed by each derivation array is recorded as a risk derivation sequence. The eigenvalue is calculated using the correlation coefficient matrix constructed by the risk derivation sequence. The ratio of the maximum eigenvalue to the sum of all eigenvalues ​​is recorded as the concentration risk load Cntla. The correlation coefficient between the activity and concentration equilibrium risk is the loss interaction coefficient Lsraf. Here, the correlation coefficient is read from the correlation coefficient matrix.

[0027] The calculation principle of the concentration risk load is to calculate the mutual influence of activity and concentration equilibrium risk over time, and capture the weight of the influence effect of the two influencing factors when the two show a negative correlation. This provides a weight reference for the subsequent construction of an indicator system for evaluating utilization loss risk. Finally, the indicator system is used to quantify the loss of utilization caused by changes in mandarin fish activity and hydrogen concentration.

[0028] The maximum values ​​of the activity risk threshold and the concentration equilibrium risk threshold at all times during the monitoring period are respectively the activity risk peak value Hcths and the concentration equilibrium risk peak value Hzres; the utilization loss risk Utzas is calculated based on the concentration risk load, the activity risk threshold and the concentration equilibrium risk threshold:

[0029] .

[0030] Furthermore, the method for adjusting the output rate of the hydrogen pump by the utilization loss risk is: setting the adjustment interval to 10-30 minutes; all utilization loss risks obtained within an adjustment interval constitute a risk value set; the maximum value and average value of the risk value set are recorded as MU and EU respectively, and the risk overflow Ovs is calculated, Ovs = (MU-EU) / EU; if the risk overflow of the current risk value set is greater than or equal to the adjustment overflow threshold, the first deceleration condition is met; wherein the adjustment overflow threshold value range is 1.10-1.60; if the current risk value set is larger than the value of EU corresponding to the previous risk value set, the second deceleration condition is met. When both the first deceleration condition and the second deceleration condition are met, it is judged that underutilization risk occurs. Underutilization risk refers to the excessive release of hydrogen resulting in too low utilization rate. In the next adjustment interval, the hydrogen release rate will be reduced by 1%-10%; if the underutilization risk does not occur, the hydrogen release rate will be increased by 1%-10% in the next adjustment interval until the hydrogen release rate reaches the preset level.

[0031] Preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

[0032] The present invention also provides a mandarin fish hydrogenation aquaculture system based on historical data analysis, the mandarin fish hydrogenation aquaculture system based on historical data analysis includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the mandarin fish hydrogenation aquaculture method based on historical data analysis are implemented. The mandarin fish hydrogenation aquaculture system based on historical data analysis can be run on computing devices such as desktop computers, laptop computers, PDAs, and cloud data centers. The executable system may include, but is not limited to, a processor, a memory, and a server cluster. The processor executes the computer program to run in the following system units:

[0033] Collector installation unit, used to install the data collector in the breeding pond;

[0034] A hydrogen concentration monitoring unit, used to obtain hydrogen concentration;

[0035] A dynamic merging unit is used to calculate activity using video data obtained from cameras;

[0036] A data reorganization unit, used for treating a binary group consisting of activity and hydrogen concentration as a monitoring array;

[0037] A real-time evaluation unit for calculating utilization loss risk based on a monitoring array acquired in real time;

[0038] A hydrogen regulation unit is used to regulate the output rate of the hydrogen pump by taking into account the risk of loss of utilization.

[0039] The beneficial effects of the present invention are as follows: the present invention provides a hydrogen-filled farming method and system for mandarin fish based on historical data analysis, and calculates the risk of utilization loss by analyzing the binary group of activity and water hydrogen concentration in real-time data, thereby effectively quantifying and extracting the risk of hydrogen waste, and providing effective mathematical support for adjusting the hydrogen release rate to just meet the needs of mandarin fish; through dynamic adjustment, hydrogen waste and unstable utilization caused by traditional fixed hydrogen release rate are avoided, and the continuity and stability of hydrogen effect are ensured; through precise supply, the negative impact of insufficient hydrogen supply on the health of mandarin fish is avoided, which helps to reduce the decline in immunity and high incidence of diseases caused by oxidative stress in mandarin fish, and improve the health status and survival rate of mandarin fish; through automatic adjustment, manual intervention is reduced, making farming management more efficient and precise; and effective response to activity fluctuations caused by changes in environmental variables is achieved, so as to achieve the purpose of improving hydrogen efficiency and improving the quality of mandarin fish farming. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The above and other features of the present invention will become more apparent through a detailed description of the embodiments shown in conjunction with the accompanying drawings. In the drawings of the present invention, the same reference numerals represent the same or similar elements. Obviously, the drawings described below are only some embodiments of the present invention. It is possible for a person skilled in the art to derive other drawings based on these drawings without inventive effort. In the drawings:

[0041] Figure 1 Shown is a flow chart of a hydrogen-filled culture method for mandarin fish based on historical data analysis;

[0042] Figure 2 Shown is the structural diagram of the mandarin fish hydrogen breeding system based on historical data analysis. DETAILED DESCRIPTION

[0043] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other unless there is a conflict.

[0044] like Figure 1 The following is a flow chart of the hydrogen-filled culture method for mandarin fish based on historical data analysis. Figure 1 To illustrate a method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to an embodiment of the present invention, the method comprises the following steps:

[0045] Example 1:

[0046] A data collector is installed in the aquaculture pond, which includes a camera and a hydrogen concentration monitor. The activity is calculated using video data obtained from the camera. The hydrogen concentration is obtained through the hydrogen concentration monitor. The binary group consisting of the activity and hydrogen concentration is used as a monitoring array. The utilization loss risk is calculated based on the monitoring array obtained in real time. The output rate of the hydrogen pump is adjusted according to the utilization loss risk.

[0047] Furthermore, a data collector including a camera and a hydrogen concentration monitor is installed in the aquaculture pond. The method comprises: a water inlet of a circulating water treatment device is provided in the center of the aquaculture pond, and a hydrogen concentration monitor is arranged in the water inlet; the hydrogen concentration monitor is an infrared gas analyzer; and cameras are installed in the center and four corners of the aquaculture pond, respectively, at a height of 0.5 meters below the water surface.

[0048] Furthermore, the method for calculating activity level using video data obtained by a camera is as follows: the water body of the breeding pond is monitored by a camera and video data is obtained in real time; individual mandarin fish are identified from each frame of the video data using a target detection method, and the movement path of the individual mandarin fish is obtained using a tracking algorithm, where the constraint time of the movement path is 5 seconds; the ratio of the total length of the movement path to the constraint time is the corresponding activity level of the individual mandarin fish; wherein the target detection method is the YOLO algorithm, and the tracking algorithm is the DeepSORT tracking algorithm.

[0049] Furthermore, the method of using the binary group consisting of activity and hydrogen concentration as a monitoring array is: setting a time interval as the feedback interval TG, the feedback interval value is 2 minutes, and obtaining a monitoring array every TG, the monitoring array being a binary group consisting of hydrogen concentration and activity.

[0050] Furthermore, the method for calculating the utilization loss risk based on the real-time monitoring array is as follows: set a time period as the monitoring period KETW, which is set to 1 hour; record the moment when the activity reaches an extreme value in the current monitoring period as a Class I extreme point, and the moment when the hydrogen concentration reaches an extreme value as a Class II extreme point; record the number of moments between any Class I extreme point and the first Class II extreme point with the same polarity obtained by reverse time search as the concentration time lag interval selected value; and record the median value of each concentration time lag interval selected value as the concentration time lag interval GOMT;

[0051] Calculate the average value of the hydrogen concentration at any moment and its counterclockwise GOMT moments as the dynamic mean value of hydrogen at that moment Medrc. Standardize the activity and hydrogen concentration in the monitoring array sequence to form a normalized feature sequence. The activity of the normalized feature sequence is recorded as the activity normalization value Nrmac. The average value and range of all hydrogen concentrations in the normalized feature sequence are recorded as the first concentration mean Fstra and the first concentration range Fsonr respectively. At any moment, calculate the utilization loss risk Utzas based on the activity normalization value and the hydrogen dynamic mean value: Utzas=Nrmac -1 / 3 ·Fsonr -1 ·(Medrc-Fstra).

[0052] Furthermore, the method for adjusting the output rate of the hydrogen pump by the utilization loss risk is: setting the adjustment interval to 15 minutes; all utilization loss risks obtained within an adjustment interval constitute a risk value set; the maximum value and average value of the risk value set are recorded as MU and EU respectively, and the risk overflow Ovs is calculated, Ovs = (MU-EU) / EU; if the risk overflow of the current risk value set is greater than or equal to the adjustment overflow threshold, the first deceleration condition is met; wherein the adjustment overflow threshold is 1.2; if the current risk value set is larger than the value of EU corresponding to the previous risk value set, the second deceleration condition is met, and when both the first deceleration condition and the second deceleration condition are met, it is judged that insufficient utilization risk occurs, and the hydrogen release rate will be reduced by 3% in the next adjustment interval; if the insufficient utilization risk does not occur, the hydrogen release rate will be increased by 3% in the next adjustment interval until the hydrogen release rate reaches the preset level.

[0053] Example 2:

[0054] Example 2 uses the same hydrogen charging farming method as Example 1, the difference being that the method for calculating the utilization loss risk based on the monitoring array obtained in real time is as follows: set a time period as the monitoring period KETW, KETW takes a value of 1 hour; record the average value of the concentration equilibrium risk in the monitoring period as the concentration risk mean Cerik, record the monitoring arrays obtained at each moment in the current monitoring period to form a monitoring array sequence, record the activity as Vilve, and record the hydrogen concentration as Hygen; draw a time series curve for all hydrogen concentrations in the monitoring array sequence; search for the first extreme point in the reverse time direction on the time series curve starting from any moment; the number of moments between the obtained extreme point and any moment is the fitting delay Cceva, define the absolute value of the difference between the hydrogen concentration value at any moment and its previous moment as the concentration change value Fseca, and calculate the concentration equilibrium risk Tequm based on the fitting delay and the concentration change value:

[0055] ;

[0056] Among them, i1, i2 are cumulative variables, Fseca i1 is the concentration change value at the i1th moment in the counterclockwise direction at the current moment, Fseca i2 , Hygen i2 are the concentration change value and hydrogen concentration at the i2th moment counterclockwise from the current moment, respectively. Fseca0 and Hygen0 are the concentration change value and hydrogen concentration at the current moment, respectively.

[0057] The mean and standard deviation of all activity levels in the monitoring array sequence are respectively the first activity mean Fstvm and the first activity fluctuation value Fsfva, and the activity risk threshold at any moment is calculated as Acths = max(0, Fstvm - Fsfva - Vilve); the concentration equilibrium risk threshold at any moment is calculated as Azres = max(0, Tequm - Cerik),

[0058] The activity and concentration balance risk at any moment are combined into a binary pair and recorded as a derivation array. The sequence formed by each derivation array is recorded as a risk derivation sequence. The eigenvalue is calculated through the correlation coefficient matrix constructed by the risk derivation sequence. The ratio of the maximum eigenvalue to the sum of all eigenvalues ​​is recorded as the concentration risk load Cntla. The correlation coefficient between activity and concentration balance risk is the loss interaction coefficient Lsraf. The maximum values ​​of the activity risk threshold and concentration balance risk threshold at all moments in the monitoring period are recorded as the activity risk peak value Hcths and the concentration balance risk peak value Hzres, respectively. The utilization loss risk Utzas is calculated based on the concentration risk load, activity risk threshold, and concentration balance risk threshold:

[0059] .

[0060] Comparative Example 1 is a culture of mandarin fish without using hydrogen charging technology.

[0061] Comparative Example 2 is a mandarin fish farming method that adopts hydrogen filling technology but does not adjust the hydrogen pump for utilization loss risk.

[0062] Table 1

[0063] Weight gain rate (%) Feeding rate (g / tail) Average hydrogen charging rate (m³ / h) Adjustment frequency (ts / h) Comparative Example 1 90.26±33.1 42.82±0.38 0 0 Comparative Example 2 115.34±25.4 45.95±1.03 0.85 0 Example 1 118.43±23.9 48.57±2.12 0.77±1.79 3.63±0.35 Example 2 120.26±30.43 48.82±1.58 0.74±0.28 4.22±0.27

[0064] As shown in Table 1, the breeding performance data of mandarin fish cultured under different hydrogen filling conditions are shown. The growth rate and specific growth rate data in Comparative Example 1 and Comparative Example 2 can explain that hydrogen-filled mandarin fish has a significant breeding optimization effect. After the hydrogen filling rate is adjusted by this method, the hydrogen filling rate is adaptively adjusted, which not only improves the weight gain rate during the breeding process, but also slows down the average hydrogen filling rate, suggesting that the consumption of hydrogen is improved. The adjustment frequency of Example 2 has a higher frequency performance than that of Example 1. This is because the recognition sensitivity of Example 2 is better. It should be noted that in the performance of the average hydrogen filling rate, Example 1 has instability due to the hysteresis phenomenon, so its adjustment value is easy to sink excessively, which makes the hydrogen supply too small to a certain extent, reducing the triggering opportunity, so the adjustment frequency is insufficient. Among them, the higher the weight gain rate, the faster the animal grows in a specific period of time. The larger the feeding rate value, the more appetite the mandarin fish has and the better growth trend it has.

[0065] The embodiment of the present invention provides a mandarin fish hydrogenation breeding system based on historical data analysis, such as Figure 2 The figure shows a structural diagram of the mandarin fish hydrogenation breeding system based on historical data analysis of the present invention. The mandarin fish hydrogenation breeding system based on historical data analysis of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the embodiment of the mandarin fish hydrogenation breeding method based on historical data analysis are implemented.

[0066] The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to run in the following units of the system:

[0067] Collector installation unit, used to install the data collector in the breeding pond;

[0068] A hydrogen concentration monitoring unit, used to obtain hydrogen concentration;

[0069] A dynamic merging unit is used to calculate activity using video data obtained from cameras;

[0070] A data reorganization unit, used for treating a binary group consisting of activity and hydrogen concentration as a monitoring array;

[0071] A real-time evaluation unit for calculating utilization loss risk based on a monitoring array acquired in real time;

[0072] A hydrogen regulation unit is used to regulate the output rate of the hydrogen pump by taking into account the risk of loss of utilization.

[0073] The mandarin fish hydrogenation and aquaculture system based on historical data analysis can be run on computing devices such as desktop computers, laptops, PDAs, and cloud servers. The mandarin fish hydrogenation and aquaculture system based on historical data analysis can be operated on systems that may include, but are not limited to, processors and memories. Those skilled in the art will understand that the examples are merely examples of mandarin fish hydrogenation and aquaculture systems based on historical data analysis and do not constitute a limitation on the mandarin fish hydrogenation and aquaculture system based on historical data analysis. The system may include more or fewer components than the example, or a combination of certain components, or different components. For example, the mandarin fish hydrogenation and aquaculture system based on historical data analysis may also include input and output devices, network access devices, buses, and the like.

[0074] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the operating system of the mandarin fish hydrogenation aquaculture system based on historical data analysis, and utilizes various interfaces and lines to connect various parts of the operating system of the mandarin fish hydrogenation aquaculture system based on historical data analysis.

[0075] The memory can be used to store the computer programs and / or modules. The processor implements the various functions of the mandarin fish hydrogen aquaculture system based on historical data analysis by running or executing the computer programs and / or modules stored in the memory and accessing the data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data and a phone book). Furthermore, the memory may include high-speed random access memory (RAM) and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0076] Although the present invention has been described in considerable detail and with particularity with respect to several embodiments, it is not intended to limit the present invention to any of these details or embodiments or any particular embodiment, so as to effectively encompass the intended scope of the present invention. In addition, the present invention has been described above with respect to embodiments foreseen by the inventors for the purpose of providing a useful description, and those insubstantial modifications of the present invention that are not currently foreseen may still represent equivalent modifications of the present invention.

Claims

1. A hydrogen-filled culture method for mandarin fish based on historical data analysis, characterized in that: The method comprises the following steps: A data collector is installed in the aquaculture pond, and the data collector includes a camera and a hydrogen concentration monitor; the activity is calculated through the video data obtained by the camera; the hydrogen concentration is obtained through the hydrogen concentration monitor; the binary group consisting of the activity and the hydrogen concentration is used as a monitoring array; the utilization loss risk is calculated based on the monitoring array obtained in real time; the output rate of the hydrogen pump is adjusted based on the utilization loss risk; wherein the method for calculating the utilization loss risk based on the monitoring array obtained in real time is as follows: set a time period as the monitoring period KETW, KETW∈[0.25,2] hours; within the current monitoring period, the moment when the activity has an extreme value is recorded as a first-class extreme point, and the moment when the hydrogen concentration has an extreme value is recorded as a second-class extreme point, the number of moments between any one-class extreme point and the first second-class extreme point of the same polarity obtained by reverse time search is recorded as the concentration time lag interval selection value, and the median value of each concentration time lag interval selection value is recorded as the concentration time lag interval GOMT; Calculate the average value of the hydrogen concentration at any moment and its counterclockwise GOMT moments as the dynamic mean value of hydrogen at that moment Medrc; standardize the activity and hydrogen concentration in the monitoring array sequence to form a normalized feature sequence, record the activity of the normalized feature sequence as the activity normalized value Nrmac, record the average value and range of all hydrogen concentrations in the normalized feature sequence as the first concentration mean Fstra and the first concentration range Fsonr respectively; at any moment, calculate the utilization loss risk Utzas based on the activity normalized value and the hydrogen dynamic mean: Utzas=Nrmac -1 / 3 ·Fsonr -1 ·(Medrc-Fstra).

2. The method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to claim 1, characterized in that: The method for installing a data collector in a breeding pond, wherein the data collector includes a camera and a hydrogen concentration monitor, is as follows: a water inlet of a circulating water treatment device is provided in the center of the breeding pond, and a hydrogen concentration monitor is arranged in the water inlet; the hydrogen concentration monitor is any one of an infrared gas analyzer, a semiconductor hydrogen sensor or a gas chromatograph; and cameras are respectively installed in the center and at the four corners of the breeding pond, with the installation height being 0.5 to 1 meter below the water surface.

3. The method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to claim 1, characterized in that: The method for calculating the activity level using the video data obtained by the camera is as follows: the water body of the aquaculture pond is monitored by the camera and the video data is obtained in real time; Individual mandarin fish are identified from each frame of video data using a target detection method, and their movement paths are obtained using a tracking algorithm. The movement path is constrained within a time range of 2 to 20 seconds. The ratio of the total length of the movement path to the constraint time represents the activity level of the mandarin fish. The target detection method is any one of YOLO, SSD, or Faster R-CNN. The tracking algorithm includes any one of SORT, DeepSORT, or KLT tracking algorithms.

4. The method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to claim 1, characterized in that: The method of using the binary group consisting of activity and hydrogen concentration as a monitoring array is: set a time interval as the feedback interval TG, the feedback interval value range is 2 minutes to 10 minutes, and obtain a monitoring array every TG. The monitoring array is a binary group consisting of hydrogen concentration and activity.

5. The method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to claim 1, characterized in that: An alternative method for calculating the utilization loss risk based on the monitoring arrays obtained in real time is as follows: suppose a time period is defined as the monitoring period KETW, where KETW∈[0.5,3] hours; denote the average value of the concentration equilibrium risk in the monitoring period as the concentration risk mean Cerik; denote the monitoring arrays obtained at each moment in the current monitoring period as a monitoring array sequence, denote the activity as Vilve, and the hydrogen concentration as Hygen; draw a time series curve for all hydrogen concentrations in the monitoring array sequence; search for the first extreme point in the reverse time direction on the time series curve starting from any moment; the number of moments between the obtained extreme point and any moment is defined as the fitting delay Cceva; define the absolute value of the difference between the hydrogen concentration value at any moment and its previous moment as the concentration change value Fseca; and calculate the concentration equilibrium risk Tequm based on the fitting delay and the concentration change value: ； Among them, i1, i2 are cumulative variables, Fseca i1 is the concentration change value at the i1th moment in the counterclockwise direction at the current moment, Fseca i2 , Hygen i2 are the concentration change value and hydrogen concentration at the i2th moment counterclockwise from the current moment, respectively. Fseca0 and Hygen0 are the concentration change value and hydrogen concentration at the current moment, respectively. Let the mean and standard deviation of all activities in the monitoring array sequence be the first activity mean Fstvm and the first activity fluctuation value Fsfva, respectively, and calculate the activity risk threshold Acths = max(0, Fstvm - Fsfva - Vilve) at any moment; calculate the concentration equilibrium risk threshold Azres = max(0, Tequm - Cerik) at any moment; The activity and concentration equilibrium risk at any moment are formed into a pair and recorded as a derivation array. The sequence composed of each derivation array is recorded as a risk derivation sequence. The eigenvalue is calculated through the correlation coefficient matrix constructed by the risk derivation sequence. The ratio of the maximum eigenvalue to the sum of all eigenvalues ​​is recorded as the concentration risk load Cntla. The correlation coefficient between activity and concentration equilibrium risk is the loss interaction coefficient Lsraf; The maximum values ​​of the activity risk threshold and the concentration equilibrium risk threshold at all times during the monitoring period are respectively the activity risk peak value Hcths and the concentration equilibrium risk peak value Hzres; the utilization loss risk Utzas is calculated based on the concentration risk load, the activity risk threshold and the concentration equilibrium risk threshold: 。 6. The method for hydrogen-filled cultivation of mandarin fish based on historical data analysis according to claim 1, characterized in that: The method for adjusting the output rate of the hydrogen pump by the utilization loss risk is as follows: setting the adjustment interval to 10-30 minutes; all utilization loss risks obtained within an adjustment interval constitute a risk value set; the maximum value and average value of the risk value set are recorded as MU and EU respectively, and the risk overflow Ovs is calculated, Ovs = (MU-EU) / EU; if the risk overflow of the current risk value set is greater than or equal to the adjustment overflow threshold, the first speed reduction condition is met; the adjustment overflow threshold value range is 1.10-1.60; if the current risk value set is larger than the value of EU corresponding to the previous risk value set, the second speed reduction condition is met. When both the first speed reduction condition and the second speed reduction condition are met, it is judged that the underutilization risk occurs, and the hydrogen release rate will be reduced by 1%-10% in the next adjustment interval.

7. The hydrogen-filled aquaculture system for mandarin fish based on historical data analysis is characterized by: The mandarin fish hydrogenation aquaculture system based on historical data analysis includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the mandarin fish hydrogenation aquaculture method based on historical data analysis according to any one of claims 1 to 6 are implemented. The mandarin fish hydrogenation aquaculture system based on historical data analysis runs on a desktop computer, a laptop computer, a PDA, and a computing device in a cloud data center.

Images