Early warning method and system based on marine ranch pollution

Abstract

The invention relates to an early warning method and system based on marine ranch pollution, and relates to the technical field of signal processing, and the method comprises the steps: building a three-dimensional biosensor network matrix, and dynamically adjusting the sensor deployment density of a marine ranch; a biological indicator response function is constructed according to the sensor deployment density, and a comprehensive response value is calculated; performing multi-source fusion on the sensor network data, the biological response data and the traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index; calculating and evaluating the risk index of the comprehensive pollution risk of the marine ranch by combining the fused pollution index and the space-time parameter; dynamically optimizing the early warning threshold according to the change rate of the risk index and the historical data entropy to obtain a dynamic threshold; and based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold, generating a multi-stage early warning signal for marine ranch pollution early warning. The accuracy of marine ranch pollution early warning can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and particularly to a warning method, system, electronic device and non-transitory computer-readable storage medium for ocean ranch pollution. Background Art

[0002] As an important model for the sustainable utilization of marine resources, ocean ranches have been widely applied globally. Existing pollution warning methods mainly rely on physical and chemical monitoring technologies, such as regularly collecting seawater samples to detect key indicators such as dissolved oxygen, pH value, ammonia nitrogen, and heavy metal concentration. At the same time, satellite remote sensing and unmanned aerial vehicle inspection technologies are also used for large-scale and real-time monitoring of the surface pollution situation of the sea area.

[0003] However, there are still some limitations in existing methods. Firstly, physical and chemical monitoring relies on fixed-point and timed sampling, which cannot comprehensively cover the entire ranch area, resulting in monitoring blind spots in space and time. Secondly, although satellite remote sensing has a wide coverage range, it is limited by weather and sea conditions and is difficult to continuously and stably obtain high-resolution data. Summary of the Invention

[0004] The present invention aims at the technical problems existing in the prior art and provides a warning method, system, electronic device and non-transitory computer-readable storage medium for ocean ranch pollution that can xx the accuracy.

[0005] The technical solution of the present invention to solve the above technical problems is as follows: The present invention provides a warning method for ocean ranch pollution, and the method includes: Establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density of the ocean ranch; Construct a biological indicator response function according to the sensor deployment density and calculate a comprehensive response value; Perform multi-source fusion on sensor network data, biological response data and traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index; Combine the fused pollution index and spatio-temporal parameters to calculate and evaluate a risk index for the comprehensive pollution risk of the ocean ranch; Dynamically optimize the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold; Generate multi-level warning signals for the ocean ranch pollution warning based on the logarithmic relationship between the risk index of the ocean ranch and the dynamic threshold.

[0006] Optionally, the dynamically adjusting the sensor deployment density of the ocean ranch includes: Obtain the water depth at the location of each sensor at the current position; Obtain the reference pollution risk value corresponding to the reference state when there is no pollution risk in the marine ranch; Obtain the actual pollution risk value used to reflect the pollution risk degree at the current location; Determine the sensor deployment density according to the water depth, the reference pollution risk value, and the actual pollution risk value, in combination with the adjustment coefficient for adjusting the sensor deployment density.

[0007] Optionally, the sensor deployment density is expressed as: ; Wherein, is the sensor deployment density, are the spatial coordinates of the current location, h is the water depth, t is the time, and are the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient, and the fourth adjustment coefficient respectively, is the reference pollution risk value, and P is the pollution risk assessment value at the current location.

[0008] Optionally, constructing a biological indicator response function according to the sensor deployment density and calculating the comprehensive response value includes: Obtain the response values of various biological indicators in the marine ranch at time t; Obtain the weight of each response value; Obtain the time decay factor of the rate at which the response value decays over time; Obtain the response time interval from the start of the response of the biological indicator to the current moment; Obtain the periodic fluctuation factor representing the intensity of the periodic change of the biological indicator response and the periodic factor of the periodic change frequency.

[0009] Optionally, the comprehensive response value is expressed as: ; Wherein, is the comprehensive response value, is the response value of the i-th biological indicator at time t, is the weight of the response value of the i-th biological indicator, is the time decay coefficient, is the response time interval, is the periodic fluctuation amplitude, is the periodic factor.

[0010] Optionally, performing multi-source fusion on sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index, including: Obtain a spatial weight kernel function for considering the influence of spatial location on the pollution index; Perform fusion processing on data from different data sources to obtain data source fusion data; Perform triple integral processing on the spatial weight kernel function and the data source fusion data to obtain the fused pollution index.

[0011] Optionally, calculating a risk index for evaluating the comprehensive pollution risk of the marine ranch by combining the fused pollution index and spatio-temporal parameters includes: Calculate the weighted sum of the pollution indices of each region; Obtain the variance of the pollution index; Obtain a time scale coefficient for the degree of influence of the time cumulative effect on the risk index; Obtain the cumulative monitoring time and the reference time of pollution monitoring; Calculate the risk index based on the weighted sum of the pollution indices, the variance, the time scale coefficient, the cumulative monitoring time, and the reference time.

[0012] Optionally, dynamically optimizing the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold includes: Obtain a reference threshold of the risk index when there are no other influencing factors; Obtain a risk change sensitivity for the degree of influence of the change rate of the risk index on the warning threshold; Obtain the historical data entropy value representing the uncertainty of the historical data of the pollution data, and the reference entropy value of the reference level of the historical data quotient value; Obtain a fluctuation coefficient representing the intensity of the periodic fluctuation of the risk index, and a period parameter of the periodic fluctuation frequency; Determine the dynamic threshold according to the risk change sensitivity, the historical data entropy value, the reference entropy value, the fluctuation coefficient, and the period parameter.

[0013] Optionally, generating multi-level warning signals for the pollution warning of the marine ranch based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold includes: Obtain the risk ratio of the risk index to the dynamic threshold; Obtain a grading coefficient for adjusting the degree of influence of the risk ratio on the warning level; Obtain a verification index related to the pollution risk; Obtain an auxiliary factor weight representing the importance of each verification index in the warning level; Determine the multi-level warning signals of the pollution warning level of the marine ranch according to the risk ratio, the grading coefficient, the verification index, and the auxiliary factor weight.

[0014] The present invention also provides an early warning system based on the pollution of a marine ranch, and the system includes: A sensor deployment module, which is used to establish a three-dimensional biological sensor network matrix and dynamically adjust the deployment density of sensors in the marine ranch; An integrated response module, which is used to construct a biological indicator response function according to the sensor deployment density and calculate an integrated response value; A pollution index module, which is used to perform multi-source fusion on sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index; A pollution risk module, which is used to calculate and evaluate a risk index of the comprehensive pollution risk of the marine ranch by combining the fused pollution index and spatio-temporal parameters; A dynamic threshold module, which is used to dynamically optimize an early warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold; A pollution early warning module, which is used to generate multi-level early warning signals for the pollution early warning of the marine ranch based on the logarithmic relationship between the risk index and the dynamic threshold of the marine ranch.

[0015] In addition, to achieve the above object, the present invention also proposes an electronic device, including: a memory, which is used to store a computer software program; a processor, which is used to read and execute the computer software program, and further implement an early warning method based on the pollution of a marine ranch as described above.

[0016] In addition, to achieve the above object, the present invention also proposes a non-transitory computer-readable storage medium, in which a computer software program is stored, and when the computer software program is executed by a processor, an early warning method based on the pollution of a marine ranch as described above is implemented.

[0017] The beneficial effects of the present invention are: (1) Through the three-dimensional biological sensor network, the present invention dynamically adjusts the deployment density in combination with water depth, time, and pollution risk, realizes full coverage of the sea area, and completely eliminates the blind area problem of traditional discrete monitoring points; (2) The biological indicator of the present invention has a sensitive physiological response to pollutants. By combining real-time modeling and attenuation factors, it can quickly capture pollution change signals and detect pollution events earlier than simple physical and chemical monitoring; (3) Through the spatial kernel function and multi-source data weighted fusion, the present invention organically combines sensor data, biological response data, and traditional monitoring data, effectively reduces the error of a single data source, and improves the accuracy and robustness of pollution assessment.

[0018] In summary, the present invention can not only improve the pollution monitoring accuracy and response speed of the marine ranch, but also make the pollution warning more scientific and forward-looking through dynamic optimization and data fusion technologies. Ultimately, it can effectively protect the marine ecological environment and provide strong technical support for the sustainable development of the marine ranch. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of a warning method based on marine ranch pollution provided by the present invention; Figure 2 It is a schematic structural diagram of a warning system based on marine ranch pollution provided by the present invention; Figure 3 It is a schematic hardware structure diagram of a possible electronic device provided by the present invention; Figure 4 It is a schematic hardware structure diagram of a possible computer-readable storage medium provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0021] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0022] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. The following description is given to enable any person skilled in the art to make and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those skilled in the art can recognize that the present invention can be practiced without these specific details. In other instances, well-known structures and processes are not described in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0023] Please refer toFigure 1 , a flowchart of a warning method based on ocean ranch pollution according to the present invention is provided, including the following steps: Step 201, establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density in the ocean ranch.

[0024] In some embodiments, step 201 may include: Obtain the water depth at the location of each sensor at the current position; Obtain the reference pollution risk value corresponding to the reference state when there is no pollution risk in the ocean ranch; Obtain the actual pollution risk value reflecting the pollution risk degree at the current position; According to the water depth, the reference pollution risk value, and the actual pollution risk value, and in combination with the adjustment coefficient for adjusting the sensor deployment density, determine the sensor deployment density.

[0025] In some embodiments, the sensor deployment density can be expressed as: ; Wherein, is the sensor deployment density, is the spatial coordinate at the current position, h is the water depth, t is the time, and are the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient, and the fourth adjustment coefficient respectively, is the reference pollution risk value, and P is the pollution risk assessment value at the current position.

[0026] In specific implementation, is a constant, called the first adjustment coefficient. It determines the overall scale of the sensor deployment density and is a basic adjustment parameter.

[0027] In the formula, provides a reference density level. If has a large value, the deployment density of the entire sensor network will be relatively high; if has a small value, the deployment density will be low. In actual deployment, can be adjusted according to the scale, budget, and monitoring objectives of the ocean ranch. For example, if a denser monitoring network is required, can be increased.

[0028] , h is the water depth at the location of the sensor, is the second adjustment coefficient. This exponential decay function is used to simulate the influence of water depth on the sensor deployment density. As the water depth increases, the deployment density of the sensor will decay exponentially.

[0029] It is understandable that the deployment cost in deep - water areas is higher (e.g., longer cables and stronger anchoring devices are required). The environmental conditions in deep - water areas may be more complex (such as water flow, pressure, etc.), resulting in increased difficulty in sensor maintenance. The spread of pollution in deep - water areas may be relatively slow, so there is no need for as dense monitoring as in shallow - water areas.

[0030] Exponential function The value decreases as h increases. When h = 0, = 1; when h increases, approaches 0.

[0031] In a marine ranch, shallow - water areas (such as areas close to the coast) usually require denser sensor deployment because the pollution risk in these areas is higher and they are more vulnerable to human activities. By adjusting the value, the influence degree of water depth on the deployment density can be controlled.

[0032] Time - periodic influence , where t is time, is the third adjustment coefficient, is the fourth adjustment coefficient. This term introduces time - periodic changes to simulate the dynamic adjustment of sensor deployment density over time.

[0033] controls the intensity of the periodic change. If is larger, the influence of the periodic change on the deployment density is more significant. determines the frequency of the period. For example, if = 2π, the period is 1 time unit (such as hours, days, etc.).

[0034] is a periodic function whose value varies between - 1 and 1. Therefore, the value of to varies.

[0035] The value of to varies, indicating that the deployment density fluctuates around the reference value of 1.

[0036] In a marine ranch, some environmental factors (such as tides and seasonal changes) may affect the sensor deployment requirements. For example, denser monitoring may be required during tides because the spread rate of pollutants is faster.

[0037] By adjusting and , these periodic changes can be simulated. For example, if it is necessary to simulate daily periodic changes, can be set to 2π (assuming the time unit is hours); if it is necessary to simulate seasonal changes, can be set to 2π / 365 (assuming the time unit is days).

[0038] Pollution risk impact , is the reference pollution risk value, and P is the pollution risk assessment value at the current location. It reflects the impact of the pollution risk at the current location on the sensor deployment density.

[0039] is a reference level, indicating the reference state without pollution risk. P is the actual pollution risk value at the current location, reflecting the degree of pollution risk at this location. represents the ratio of the pollution risk at the current location to the reference risk. If P increases (i.e., the pollution risk is higher), this ratio will decrease; if P decreases (i.e., the pollution risk is lower), this ratio will increase.

[0040] The square root function is used to smooth the change of this ratio, making the change of the deployment density more gentle. When P = , = 1, indicating that the pollution risk at the current location is equivalent to the reference level, and the deployment density is not affected. If P increases, will decrease, and the deployment density will decrease; if P decreases, will increase, and the deployment density will increase.

[0041] In the marine ranch, areas with higher pollution risks may require more intensive monitoring, but at the same time, the deployment cost will also be higher. By introducing this term, the deployment density of sensors can be dynamically adjusted to match the pollution risk. For example, in areas with lower pollution risks, the deployment density of sensors can be appropriately reduced to save costs; in areas with higher pollution risks, the deployment density of sensors can be increased to improve the monitoring accuracy. Through this comprehensive consideration, the formula can more scientifically guide the deployment of sensors, enabling it to meet the monitoring requirements while reasonably controlling costs.

[0042] Step 202, construct a bioindicator response function according to the sensor deployment density, and calculate the comprehensive response value.

[0043] In some embodiments, step 202 may include: Obtain the response values of various bioindicators in the marine ranch at time t; Obtain the weight of each response value; Obtain the time decay factor of the rate at which the response value decays over time; ​​​​Obtain the response time interval from the start of the response of the biological indicator to the current moment; Obtain a periodic fluctuation factor representing the intensity of the periodic change in the response of the biological indicator and a periodic factor representing the periodic change frequency.

[0044] In some embodiments, the comprehensive response value can be expressed as:

[0045] where, is the comprehensive response value, is the response value of the i-th biological indicator at time t, is the weight of the response value of the i-th biological indicator, is the time decay coefficient, is the response time interval, is the periodic fluctuation amplitude, is the periodic factor.

[0046] In specific implementation, the weighted response of the biological indicator , is the response value of the i-th biological indicator at time t. A biological indicator refers to an organism (such as plankton, algae or benthos) that is sensitive to environmental changes, and the physiological, behavioral or population changes of which can reflect the changes in environmental quality.

[0047] is the weight coefficient of the i-th biological indicator. The weight coefficient is used to reflect the importance of different biological indicators in the comprehensive response. For example, some biological indicators may be more sensitive to pollution, so their weights will be higher. By multiplying the response value of each biological indicator by its weight and summing them up, a comprehensive biological response signal is obtained, which reflects the overall response of multiple biological indicators to environmental changes.

[0048] Time decay factor , is the time decay coefficient, representing the decay rate of the biological indicator response over time. This coefficient reflects the persistence of the biological response. If is very large, the response will decay rapidly; if is very small, the response will last for a long time.

[0049] is the response time interval, representing the duration from the start of the response of the biological indicator to the current time. This parameter is used to simulate the natural weakening of the biological response over time. As time goes by, the response intensity of the biological indicator will gradually weaken. This exponential decay function simulates this natural decay process, making the comprehensive response value R(t) more in line with the actual situation.

[0050] Periodic fluctuation factor , is the amplitude of the periodic fluctuation, representing the intensity of the periodic change in the biological indicator response. This parameter determines the maximum change amplitude of the response value during the periodic fluctuation.

[0051] is the period factor, representing the frequency of the periodic change. It determines the period length of the biological response. For example, if , the period is 1 time unit (such as hours, days, etc.).

[0052] is the cosine function, used to introduce the periodic change. The value of the cosine function varies between -1 and 1. Therefore, will vary between - and .

[0053] The response of the biological indicator is affected by periodic factors, such as circadian rhythm, tidal changes, or seasonal changes. This periodic fluctuation factor simulates the impact of these periodic changes on the biological response.

[0054] The formula R(t) is a comprehensive response value that combines the following three key factors: The weighted response of the biological indicator: By weighted summation, the response values of different biological indicators are integrated together, reflecting the comprehensive response of multiple organisms to environmental changes.

[0055] Time decay: Through the exponential decay function, it simulates the natural weakening of the biological response over time, making the response value more in line with the actual situation.

[0056] Periodic fluctuation: Through the cosine function, it introduces periodic changes and simulates the impact of periodic environmental factors on the biological response.

[0057] Finally, the obtained R(t) is a dynamic, time-dependent response value, which can more comprehensively reflect the comprehensive response of biological indicators to pollution changes in the marine environment.

[0058] In the pollution monitoring of marine ranches, the biological indicator response value R(t) can be used as important input data for subsequent multi-source data fusion and pollution risk assessment. For example: If R(t) increases significantly, it may indicate the occurrence of a pollution event in the marine ranch. If R(t) shows obvious periodic changes, it may be related to tides, circadian rhythm, or other periodic environmental factors. If R(t) decays rapidly over time, it may indicate the response of biological indicators to short-term pollution events. In the above ways, the present invention provides a dynamic index based on biological response for pollution early warning in marine ranches, which can more sensitively reflect environmental changes.

[0059] Step 203: Perform multi-source fusion on the sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain the fused pollution index.

[0060] In some embodiments, step 203 may include: Obtain a spatial weight kernel function for considering the influence of spatial location on the pollution index; Perform fusion processing on the data from different data sources to obtain data source fusion data; Perform triple integral processing on the spatial weight kernel function and the data source fusion data to obtain the fused pollution index.

[0061] In some embodiments, the fused pollution index can be expressed as: ; where, is the fused pollution index, is the spatial weight kernel function, is the traditional monitoring data, are the first data source weight, the second data source weight, and the third data source weight respectively.

[0062] In specific implementation, this formula is used to calculate the fused pollution index F(x, y, z, t), which combines the biological sensor network data, biological indicator response data, and traditional monitoring data.

[0063] K(x, y, z) is a spatial weight kernel function for considering the influence of spatial location on the pollution index. This function can be a distance-based weight function, such as a Gaussian kernel function or a spherical kernel function, or a function based on spatial correlation, such as a spatial autocorrelation function. The value of K(x, y, z) usually decreases with the increase of distance, indicating that the farther the spatial location, the smaller the influence on the current point.

[0064] are the first data source weight, the second data source weight, and the third data source weight respectively. D(x, y, z) is the deployment density of the biological sensor network. R(t) is the comprehensive response value of the biological indicator. S(t) is the traditional monitoring data. This weighted sum represents the contributions of different data sources to the pollution index. The weight coefficients a, b, and c are used to adjust the importance of different data sources.

[0065] Spatial integration , this triple integral represents integrating over all positions in three-dimensional space to obtain the fused pollution index F(x, y, z, t). The integration range usually depends on the size and resolution of the study area. The spatial weight kernel function K(x, y, z) is used to consider the influence of spatial positions on the pollution index. It can be selected according to the actual spatial correlation or distance decay relationship.

[0066] Data source fusion data , weight coefficients , b, and c are used to adjust the importance of different data sources, indicating the contributions of different data sources to the pollution index. For example, if the data from the biosensor network is more reliable, the value of can be increased; if the traditional monitoring data is more accurate, the value of c can be increased. By integrating over all positions in three-dimensional space, the fused pollution index F(x, y, z, t) can be obtained, which combines the data from the biosensor network, the bioindicator response data, and the traditional monitoring data.

[0067] Through this formula, the present invention can obtain a fused pollution index F(x, y, z, t), which takes into account the spatial positions and the contributions of different data sources, providing a more comprehensive and accurate indicator for pollution monitoring in the marine ranch.

[0068] Step 204: Combine the fused pollution index and spatio-temporal parameters to calculate a risk index for evaluating the comprehensive pollution risk of the marine ranch.

[0069] In some embodiments, step 204 may include: Calculate the weighted sum of the pollution indices for each region; Obtain the variance of the pollution index; Obtain the time scale coefficient for the degree of influence of the time cumulative effect on the risk index; Obtain the cumulative monitoring time and the reference time for pollution monitoring; Calculate the risk index according to the weighted sum of the pollution indices, the variance, the time scale coefficient, the cumulative monitoring time, and the reference time.

[0070] In some embodiments, the risk index can be expressed as: ; where The risk index is is the regional weight, is the variance, T is the cumulative monitoring time, is the reference time, is the time scale coefficient.

[0071] In a specific implementation, this formula is used to calculate the comprehensive pollution risk index RI, which synthesizes the pollution indices of different regions, the time accumulation effect, and the influence of the time scale.

[0072] Weighted sum of regional pollution indices , is the pollution index of the i-th region, representing the pollution level of this region at time t. is the weight of the i-th region, representing the importance of this region in the comprehensive risk assessment. This weighted sum represents the contribution of the pollution indices of all regions to the comprehensive risk index.

[0073] Variance adjustment factor , is the variance, representing the variability or uncertainty of the pollution index. This exponential function indicates that as the variance increases, the comprehensive risk index will increase exponentially. This is because a higher variance means greater uncertainty in the pollution level, and thus higher risk.

[0074] Time accumulation effect , where T is the cumulative monitoring time, representing the total duration from the start of monitoring to the current time. is the reference time, representing a reference time point. is the time scale coefficient, which determines the degree of influence of the time accumulation effect on the comprehensive risk index. This term indicates that as the monitoring time increases, the comprehensive risk index will increase. This is because longer-term monitoring can reveal more pollution events, thus increasing the risk.

[0075] By calculating the weighted sum of the pollution indices of all regions, the present invention can obtain a comprehensive pollution level, which reflects the contributions of different regions to the overall risk. By introducing the variance adjustment factor, the influence of the uncertainty of the pollution index on the comprehensive risk can be considered. A higher variance means greater uncertainty, and thus higher risk. By introducing the time accumulation effect, the influence of the monitoring time on the comprehensive risk can be considered. As the monitoring time increases, the comprehensive risk index will increase, reflecting that longer-term monitoring can reveal more pollution events.

[0076] In the above manner, the present invention can obtain a comprehensive pollution risk index RI, which synthesizes the pollution indices of different regions, the time accumulation effect, and the influence of the time scale, providing a more comprehensive and accurate indicator for the pollution risk assessment of the marine ranch.

[0077] Step 205: Dynamically optimize the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold.

[0078] In some embodiments, step 205 may include: Obtain the baseline threshold of the risk index when there are no other influencing factors; Obtain the risk change sensitivity that represents the degree of influence of the change rate of the risk index on the warning threshold; Obtain the historical data entropy value representing the uncertainty of the historical data of the pollution data, and the reference entropy value of the baseline level of the historical data quotient value; Obtain the fluctuation coefficient representing the intensity of the periodic fluctuation of the risk index, and the period parameter of the periodic fluctuation frequency; Determine the dynamic threshold according to the risk change sensitivity, the historical data entropy value, the reference entropy value, the fluctuation coefficient, and the period parameter.

[0079] In some embodiments, the dynamic threshold can be expressed as: ; Where, is the dynamic threshold, is the baseline threshold, is the fluctuation coefficient, is the period parameter, is the risk change sensitivity, is the change rate of the risk index, is the historical data entropy value, is the reference entropy value.

[0080] In specific implementation, this formula is used to determine the dynamic threshold , considering the periodic fluctuation, the change rate of the risk index, and the entropy value of the historical data.

[0081] Baseline threshold is a constant representing the baseline threshold, that is, the threshold of the risk index when there are no other influencing factors. Periodic fluctuation , t is the time, representing the current moment. is the fluctuation coefficient, representing the intensity of the periodic fluctuation. is the period parameter, which determines the frequency of the periodic fluctuation. This term introduces the periodic change, simulating the situation where the threshold may change over time (such as seasons, tides, etc.).

[0082] The change rate of the risk index has an impact , is the change rate of the risk index, representing the change speed of the risk index over time. is the risk change sensitivity, which determines the degree of influence of the change rate of the risk index on the threshold. This exponential function indicates that as the change rate of the risk index increases, the threshold will decay exponentially. This is because when the change rate of the risk index is relatively high, a more stringent threshold may be required for early warning.

[0083] The historical data entropy value has an impact where \(H\) is the entropy value of historical data, representing the uncertainty and complexity of the historical data of the contaminated data. is the reference entropy value, representing a baseline level. This item indicates that the threshold is proportional to the square root of the entropy value of the historical data. If the entropy value of the historical data is high, the threshold will increase accordingly.

[0084] The threshold will show periodic changes over time, simulating the influence of possible seasonal, tidal and other factors. As the change rate of the risk index increases, the threshold will show exponential decay, reflecting the sensitivity to rapidly changing risks. The threshold is proportional to the square root of the entropy value of the historical data, reflecting the influence of the uncertainty and complexity of the historical data on the threshold.

[0085] Through this formula, a dynamic threshold can be obtained which combines the periodic fluctuations, the change rate of the risk index, and the entropy value of the historical data, providing a more flexible and accurate threshold for the pollution warning of the marine ranch.

[0086] Step 206: Generate multi-level warning signals for the pollution warning of the marine ranch based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold.

[0087] In some embodiments, step 206 may include: Obtain the risk ratio of the risk index to the dynamic threshold; Obtain the grading coefficient for adjusting the influence degree of the risk ratio on the warning level; Obtain the verification index representing the pollution risk; Obtain the auxiliary factor weight representing the importance of each verification index in the warning level; Determine the multi-level warning signals of the pollution warning level of the marine ranch according to the risk ratio, the grading coefficient, the verification index, and the auxiliary factor weight.

[0088] In some embodiments, the multi-level warning signals of the pollution warning level of the marine ranch may be expressed as: ; where is the warning level, is the grading coefficient, is the auxiliary factor weight, is the \(i\)-th verification index.

[0089] In specific implementation, this formula is used to calculate the warning level \(AL\), which combines the ratio of the risk index to the dynamic threshold and the influence of the auxiliary verification index.

[0090] The ratio of the risk index to the dynamic threshold , RI is the comprehensive pollution risk index, representing the current monitored pollution risk level. is the dynamic threshold, representing the warning threshold level at the current time point. By calculating of the ratio, the current risk can be evaluated whether it exceeds the set threshold. If RI is much greater than , the ratio is larger, indicating a higher risk; conversely, the ratio is smaller.

[0091] The use of the logarithmic function log is to compress the range of the ratio into a more manageable interval, and at the same time the logarithmic function makes the result more sensitive to the relative change of the risk. For example, when increases from 1 to 2, the change in the logarithmic value is more significant than when it increases from 10 to 20.

[0092] Grading coefficient , used to adjust the influence degree of the risk ratio on the warning level. By adjusting value, the sensitivity of the risk ratio to the warning level can be controlled. A larger value will make the warning level stricter, and a smaller value will make the warning level looser.

[0093] In the actual warning system, value can be adjusted according to the strictness of the warning and the actual needs. For example, in high-risk areas or areas more sensitive to pollution, a larger value can be set to improve the sensitivity of the warning.

[0094] Auxiliary verification index , is the weight of the i-th auxiliary factor, representing the importance of each auxiliary verification index in the warning level. is the i-th verification index, representing other auxiliary information related to the pollution risk (such as meteorological data, water quality parameters, historical data, etc.).

[0095] By introducing the auxiliary verification index, other factors that may affect the pollution risk can be comprehensively considered, making the warning level more comprehensive and accurate. For example, even if the risk index RI does not exceed the threshold, some auxiliary indicators (such as meteorological conditions) may indicate that the pollution risk is increasing. The sum of each auxiliary indicator multiplied by its weight represents the comprehensive contribution of these auxiliary indicators to the warning level.

[0096] The ceiling function ⌈⋅⌉ ensures that the warning level is an integer, facilitating the hierarchical warning in practical applications. For example, the warning level can be divided into level 1 (low risk), level 2 (medium risk), and level 3 (high risk).

[0097] Formula AL is a comprehensive warning level, which consists of two parts: The logarithm of the risk ratio: By calculating and multiplying by the grading coefficient , it is evaluated whether the current risk exceeds the threshold.

[0098] The weighted sum of the auxiliary verification indicators: By introducing the auxiliary verification indicators and their weights , other factors that may affect the pollution risk are comprehensively considered.

[0099] Finally, the result is converted into an integer warning level through the ceiling function, which is convenient for hierarchical warning in practical applications.

[0100] Suppose a pollution warning system for an ocean ranch is running, and the specific parameters are as follows: The current risk index RI = 1.5; The dynamic threshold Th(t) = 1.2; The grading coefficient = 2; The auxiliary verification indicators: = 0.3 (meteorological conditions); = 0.2 (water quality parameters); The weights of the auxiliary factors: = 0.5, = 0.3.

[0101] Substitute into the formula: AL = ⌈0.446 + 0.21⌉ = 1.

[0102] Therefore, the warning level AL = 1, indicating that the current risk is at a low risk level. Through this method of comprehensively considering the risk index, dynamic threshold and auxiliary verification indicators, the formula AL of the present invention can more accurately reflect the pollution risk of the ocean ranch and provide a scientific basis for the actual warning system.

[0103] Please refer to Figure 2 , Figure 2 which is the structural schematic diagram of a warning system for ocean ranch pollution provided by the present invention.

[0104] As Figure 2 shown, a warning system for ocean ranch pollution proposed in an embodiment of the present invention includes: The sensor deployment module 301 is used to establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density of the ocean ranch; The comprehensive response module 302 is used to construct a bioindicator response function according to the sensor deployment density and calculate the comprehensive response value; A pollution index module 303, configured to perform multi-source fusion on sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index; A pollution risk module 304, configured to calculate and evaluate a risk index of the comprehensive pollution risk of the marine ranch by combining the fused pollution index and spatio-temporal parameters; A dynamic threshold module 305, configured to dynamically optimize an early warning threshold according to a change rate of the risk index and a historical data entropy value to obtain a dynamic threshold; A pollution early warning module 306, configured to generate multi-level early warning signals for the pollution early warning of the marine ranch based on a logarithmic relationship between the risk index of the marine ranch and the dynamic threshold.

[0105] Please refer to Figure 3 , Figure 3 , which is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 3 shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored on the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented: Establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density of the marine ranch; Construct a biological indicator response function according to the sensor deployment density and calculate a comprehensive response value; Perform multi-source fusion on sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain a fused pollution index; Combine the fused pollution index and spatio-temporal parameters to calculate and evaluate a risk index of the comprehensive pollution risk of the marine ranch; Dynamically optimize an early warning threshold according to a change rate of the risk index and a historical data entropy value to obtain a dynamic threshold; Generate multi-level early warning signals for the pollution early warning of the marine ranch based on a logarithmic relationship between the risk index of the marine ranch and the dynamic threshold.

[0106] Please refer to Figure 4 , Figure 4 , which is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. As Figure 4 shown, this embodiment provides a computer-readable storage medium 500, on which a computer program 411 is stored. When the computer program 411 is executed by a processor, the following steps are implemented: Establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density of the marine ranch; Construct a bioindicator response function according to the sensor deployment density and calculate the comprehensive response value; Perform multi-source fusion on the sensor network data, biological response data, and traditional monitoring data through a spatial weight kernel function to obtain the fused pollution index; Combine the fused pollution index and spatio-temporal parameters to calculate a risk index for evaluating the comprehensive pollution risk of the marine ranch; Dynamically optimize the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold; Generate multi-level warning signals for pollution warning of the marine ranch based on the logarithmic relationship between the risk index and the dynamic threshold of the marine ranch.

[0107] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0108] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a system for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0110] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction system that implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the steps specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 or steps of functions specified in a plurality of blocks.

[0112] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn the basic inventive concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0113] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An early warning method based on marine ranch pollution, characterized in that: The method comprises: Establish a three-dimensional biosensor network matrix to dynamically adjust the sensor deployment density of the marine ranch; Constructing a biological indicator response function according to the sensor deployment density and calculating a comprehensive response value; The sensor network data, biological response data and traditional monitoring data are fused through the spatial weight kernel function to obtain the fused pollution index. Combining the fused pollution index and spatiotemporal parameters, calculating a risk index for evaluating the comprehensive pollution risk of the marine ranch; Dynamically optimize the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold; Based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold, a multi-level warning signal for pollution warning of the marine ranch is generated.

2. The early warning method based on marine ranch pollution according to claim 1 is characterized in that: The method of dynamically adjusting the sensor deployment density of the marine ranch includes: Get the water depth of each sensor at the current location; Obtaining a baseline pollution risk value corresponding to a baseline state when the marine ranch has no pollution risk; Obtaining an actual pollution risk value that reflects the pollution risk level at the current location; The sensor deployment density is determined according to the water depth, the benchmark pollution risk value and the actual pollution risk value in combination with an adjustment coefficient for adjusting the sensor deployment density.

3. The early warning method based on marine ranch pollution according to claim 2 is characterized in that: The sensor deployment density is expressed as: ； in, is the sensor deployment density, is the spatial coordinate of the current position, h is the water depth, t is the time, and are the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient and the fourth adjustment coefficient, is the baseline pollution risk value, and P is the pollution risk assessment value of the current location.

4. The early warning method based on marine ranch pollution according to claim 3 is characterized in that: The step of constructing a biological indicator response function according to the sensor deployment density and calculating a comprehensive response value comprises: Obtaining response values ​​of various biological indicators in the marine ranch at time t; Obtaining a weight for each of the response values; Obtaining a time decay factor of the rate at which the response value decays over time; Obtaining a response time interval from when the biological indicator starts responding to the current moment; A periodic fluctuation factor representing the intensity of the periodic change in response of the biological indicator and a periodic factor representing the frequency of the periodic change are obtained.

5. The early warning method based on marine ranch pollution according to claim 4 is characterized in that: The comprehensive response value is expressed as: ； in, is the comprehensive response value, is the response value of the ith biological indicator at time t, is the weight of the response value of the i-th biological indicator, is the time decay coefficient, is the response time interval, is the periodic fluctuation amplitude, is the period factor.

6. The early warning method based on marine ranch pollution according to claim 5 is characterized in that: The multi-source fusion of sensor network data, biological response data and traditional monitoring data is performed through the spatial weight kernel function to obtain the fused pollution index, including: Obtaining a spatial weight kernel function for considering the impact of spatial location on the pollution index; Perform fusion processing on data from different data sources to obtain data source fusion data; The spatial weight kernel function and the data source fusion data are subjected to triple integration processing to obtain the fused pollution index.

7. The early warning method based on marine ranch pollution according to claim 6 is characterized in that: The step of combining the fused pollution index and spatiotemporal parameters to calculate a risk index for evaluating the comprehensive pollution risk of the marine ranch includes: Calculate the weighted sum of pollution index for each region; Get the variance of the pollution index; Obtaining the time scale coefficient of the degree of influence of the time accumulation effect on the risk index; Obtain the cumulative monitoring time and benchmark time for pollution monitoring; The risk index is calculated according to the weighted sum of the pollution indices, the variance, the time scale coefficient, the cumulative monitoring time and the reference time.

8. The early warning method based on marine ranch pollution according to claim 7 is characterized in that: The step of dynamically optimizing the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain the dynamic threshold includes: Obtain the benchmark threshold of the risk index when there are no other influencing factors; Obtaining the risk change sensitivity of the risk index change rate to the degree of influence of the early warning threshold; Obtaining a historical data entropy value representing uncertainty of historical data of pollution data and a reference entropy value of a benchmark level of the historical data quotient; Obtaining a fluctuation coefficient representing the intensity of the periodic fluctuation of the risk index and a period parameter representing the frequency of the periodic fluctuation; The dynamic threshold is determined according to the risk change sensitivity, the historical data entropy value, the reference entropy value, the fluctuation coefficient and the periodic parameter.

9. The early warning method based on marine ranch pollution according to claim 8 is characterized in that: The method of generating a multi-level warning signal for the marine ranch pollution warning based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold value includes: Obtaining a risk ratio of the risk index to the dynamic threshold; Obtaining a level division coefficient for adjusting the degree of influence of the risk ratio on the warning level; Obtain verification indicators that represent the risk associated with contamination; Obtaining auxiliary factor weights representing the importance of each of the verification indicators in the warning level; A multi-level warning signal of the marine ranch pollution warning level is determined based on the risk ratio, the level classification coefficient, the verification index and the auxiliary factor weight.

10. An early warning system based on marine ranch pollution, characterized in that: The system comprises: The sensor deployment module is used to establish a three-dimensional biosensor network matrix and dynamically adjust the sensor deployment density of the marine ranch; A comprehensive response module, used to construct a biological indicator response function according to the sensor deployment density and calculate a comprehensive response value; The pollution index module is used to perform multi-source fusion of sensor network data, biological response data and traditional monitoring data through the spatial weight kernel function to obtain the fused pollution index; A pollution risk module, used to calculate and evaluate the risk index of the comprehensive pollution risk of the marine ranch by combining the fused pollution index and spatiotemporal parameters; A dynamic threshold module is used to dynamically optimize the warning threshold according to the change rate of the risk index and the historical data entropy value to obtain a dynamic threshold; The pollution warning module is used to generate a multi-level warning signal for pollution warning of the marine ranch based on the logarithmic relationship between the risk index of the marine ranch and the dynamic threshold.

Images