Marine animal husbandry environment data collection method and system

Abstract

The present invention relates to the field of data processing technology, and in particular to a method and system for collecting marine animal husbandry environment data, comprising: obtaining a local window, obtaining the degree of fluctuation of the data amplitude in the local window of each moment data according to the difference between two adjacent moment data in the local window of each moment data in the marine salinity data, obtaining the local area difference of each moment data according to all moment data in the local window of each moment data in the marine salinity data, and obtaining the credibility factor of each moment data in the marine salinity data; obtaining the first credibility according to the marine temperature data and the marine salinity data, and obtaining the second credibility according to the marine salinity data and the marine conductivity data; obtaining the correction weight of each moment data in the marine salinity data, and completing the collection of the marine salinity data. The present invention improves the accuracy of the marine salinity data collection by analyzing the marine salinity data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for collecting marine animal husbandry environment data. Background Art

[0002] There is a close mutual influence between the marine environment and marine animal husbandry. The state of the marine environment will directly affect the health and sustainability of marine animal husbandry. The water quality in the marine environment is crucial to the growth and health of farmed animals. High-quality water can provide suitable growth conditions, while the presence of pollution or harmful substances may cause disease or death. Water temperature and salinity are important factors affecting the ecology of marine organisms. Different types of farmed animals have specific requirements for different water temperature and salinity conditions. Therefore, changes in the marine environment have a great impact on marine animal husbandry.

[0003] In order to analyze whether the marine environment has changed, it is necessary to collect data in the marine environment for analysis. In order to analyze whether the collected marine environment data is abnormal, the local outlier factor LOF algorithm can be used to analyze the marine environment. Under normal circumstances, abnormal data appear due to human sewage and industrial wastewater discharge, and these abnormal data can reflect that the marine environment has changed; however, in the process of collecting data, problems such as sound vibration or sensor performance cause the collected data to be abnormal, but these abnormal data cannot reflect that the marine environment has changed. Therefore, the present invention analyzes the noise point data that appears in the collection due to problems such as sound vibration or sensor performance in the process of collecting data. Summary of the invention

[0004] The present invention provides a method and system for collecting marine animal husbandry environment data to solve the existing problems.

[0005] A method and system for collecting marine animal husbandry environment data of the present invention adopts the following technical scheme:

[0006] An embodiment of the present invention provides a method for collecting marine animal husbandry environment data, the method comprising the following steps:

[0007] Collect ocean salinity data, ocean temperature data, and ocean conductivity data;

[0008] Obtain a local window of data at each moment in the ocean salinity data, obtain the degree of fluctuation of the data amplitude in the local window of each moment in the ocean salinity data according to the difference between two adjacent moment data in the local window of each moment in the ocean salinity data, obtain a local fitting curve of the data at each moment in the local window of the data at each moment in the ocean salinity data according to all moment data, obtain a local area difference of the data at each moment according to the local fitting curve of the data at each moment, and obtain a credibility factor of the data at each moment in the ocean salinity data according to the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of the data at each moment;

[0009] The first credibility is obtained based on the ratio of the ocean temperature data to the ocean salinity data, and the second credibility is obtained based on the ratio of the ocean salinity data to the ocean conductivity data;

[0010] The correction weight of the data at each moment in the ocean salinity data is obtained according to the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data. The distance between every two moments of data is corrected according to the correction weight of the data at each moment in the ocean salinity data to obtain the distance between every two moments of data after correction. The collection of ocean salinity data is completed according to the distance between every two moments of data after correction.

[0011] Furthermore, the specific steps of obtaining the local window of the ocean salinity data at each moment include the following:

[0012] The data at each moment in the ocean salinity data is taken as the center point of the local window, and A is taken as the number of data in the local window to obtain the local window of the data at each moment; wherein A is a preset threshold.

[0013] Furthermore, the method of obtaining the fluctuation degree of the data amplitude in the local window of each moment data according to the difference between two adjacent moment data in the local window of each moment data in the ocean salinity data includes the following specific steps:

[0014] The calculation formula for the fluctuation degree of the data amplitude in the local window of each moment is:

[0015]

[0016] In the formula, H i,j represents the value of the jth element in the local amplitude sequence of the i-th moment in the ocean salinity data, H i represents the mean value of all elements in the local amplitude sequence of the i-th moment data in the ocean salinity data, A represents the number of moment data in the local window, and Y i Indicates the degree of fluctuation of the data amplitude in the local window of the data at the i-th moment, || indicates the absolute value sign;

[0017] The difference between two adjacent data points in the local window of each moment in the ocean salinity data is calculated and then the absolute value is obtained to obtain the local amplitude sequence of each moment in the time sequence.

[0018] Further, the method of obtaining a local fitting curve of each moment data according to all moment data in a local window of each moment data in the ocean salinity data, and obtaining a local area difference of each moment data according to the local fitting curve of each moment data, comprises the following specific steps:

[0019] Record any time data in the ocean salinity data as the target time data, and obtain the local fitting curve of the target time data by the least square method according to all time data in the local window of the target time data;

[0020] The calculation formula for the local area difference of the data at each moment is:

[0021]

[0022] In the formula, f i (x) represents the function corresponding to the local fitting curve of the data at the i-th moment, x represents the independent variable of the function corresponding to the local fitting curve of the data at each moment, s i,c represents the c-th moment data in the local window of the i-th moment data, s i,c-1 represents the c-1th moment data in the local window of the i-th moment data, A represents the number of moment data in the local window, Z i It represents the local area difference of the data at the i-th moment, and || represents the absolute value sign.

[0023] Furthermore, the credibility factor of the data at each moment in the ocean salinity data is obtained according to the local area difference of the data at each moment and the fluctuation degree of the data in the local window of the data at each moment, and the specific steps include the following:

[0024] The calculation formula for the credibility factor of the ocean salinity data at each moment is:

[0025] T i =exp(-Y i ×Z i )

[0026] Where Y i Indicates the degree of fluctuation of the data amplitude in the local window of the i-th moment data, Z i Represents the local area difference of the data at the i-th moment, T i It represents the credibility factor of the data at the i-th moment in the ocean salinity data, and exp() represents an exponential function with a natural constant as the base.

[0027] Furthermore, the step of obtaining the first credibility according to the ratio of the ocean temperature data to the ocean salinity data comprises the following specific steps:

[0028] According to the ratio between the ocean temperature data and the ocean salinity data at each moment, a temperature-salinity ratio sequence is obtained, and the temperature-salinity ratio fitting curve is obtained by the least square method according to the ratio data of all moments in the temperature-salinity ratio sequence, and then the ratio data of the first moment and the ratio data of the last moment are connected to obtain a time straight line;

[0029] The calculation formula of the first credibility is:

[0030]

[0031] Where l(x) represents the function corresponding to the time line, g(x) represents the function corresponding to the temperature-salinity ratio fitting curve, x represents the independent variable of the function corresponding to the time line and the function corresponding to the temperature-salinity ratio fitting curve, t1 represents the time of the first moment, t2 represents the time of the last moment, M represents the number of intersections between the time line and the temperature-salinity ratio fitting curve, and δ1 represents the first credibility.

[0032] Furthermore, the step of obtaining the second credibility according to the ratio of the ocean salinity data to the ocean conductivity data comprises the following specific steps:

[0033] According to the ratio between the ocean salinity data and the ocean conductivity data at each moment, a conductivity-salinity ratio series is obtained;

[0034] The calculation formula of the second credibility is:

[0035] δ2=exp(-D)

[0036] Where D represents the variance of all data in the conductivity-salinity ratio series, exp() represents an exponential function with a natural constant as the base, and δ2 represents the second credibility.

[0037] Furthermore, the correction weight of the data at each moment in the ocean salinity data is obtained according to the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data, including the following specific steps:

[0038] The calculation formula for the correction weight of the data at each moment in the ocean salinity data is:

[0039] U i =1-norm(T i ×δ1×δ2)

[0040] Where, T irepresents the credibility factor of the data at the i-th moment in the ocean salinity data, δ1 represents the first credibility, δ2 represents the second credibility, and U i It represents the correction weight of the data at the i-th moment in the ocean salinity data, and norm() represents the linear normalization function.

[0041] Furthermore, the distance between every two moments of data is corrected according to the correction weight of each moment of data in the ocean salinity data to obtain the distance between every two moments of data after correction, and the collection of ocean salinity data is completed according to the distance between every two moments of data after correction, including the following specific steps:

[0042] The correction coefficient of the distance between each two moment data is equal to the average of the correction weights of each two moment data. The distance between each two moment data after correction is obtained by multiplying the correction coefficient of the distance between each two moment data by the distance between each two moment data.

[0043] The LOF algorithm is used to detect abnormal data in the ocean salinity data based on the distance between the data at every two moments after correction, and the number of abnormal data is obtained, which is recorded as G; when the number of abnormal data G is greater than or equal to the preset threshold B, the collected ocean salinity data is unreliable and the ocean salinity data should be recollected; when the number of abnormal data G is less than the preset threshold B, the collected ocean salinity data is credible.

[0044] The present invention also provides a marine livestock environment data acquisition system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the above steps when executing the computer program.

[0045] The technical solution of the present invention has the following beneficial effects: the present invention obtains a local window of data at each moment in the ocean salinity data by analyzing the ocean salinity data, obtains the degree of fluctuation of the data amplitude in the local window of each moment in the ocean salinity data according to the difference between two adjacent moment data in the local window of each moment in the ocean salinity data, obtains the local area difference of the data at each moment according to all moment data in the local window of each moment in the ocean salinity data, obtains the credibility factor of the data at each moment in the ocean salinity data according to the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of each moment, thereby improving the accuracy of judging the credibility of the ocean salinity data; obtains the first credibility according to the ratio of the ocean temperature data to the ocean salinity data, and obtains the second credibility according to the ratio of the ocean salinity data to the ocean conductivity data; obtains the correction weight of the data at each moment in the ocean salinity data according to the credibility factor of the data at each moment in the ocean salinity data, the first credibility and the second credibility, completes the collection of the ocean salinity data according to the correction weight of the data at each moment in the ocean salinity data, thereby improving the accuracy of collecting the ocean salinity data. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of the steps of a method and system for collecting marine livestock environment data according to the present invention;

[0048] Figure 2 A fitting curve diagram of ocean salinity data of a method and system for collecting marine animal husbandry environment data of the present invention;

[0049] Figure 3 The present invention provides a method and system for collecting marine animal husbandry environment data and a ratio fitting curve diagram of ocean temperature data and ocean salinity data. DETAILED DESCRIPTION

[0050] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method and system for collecting marine livestock environment data according to the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0051] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0052] The specific scheme of a method and system for collecting marine livestock environment data provided by the present invention is described in detail below with reference to the accompanying drawings.

[0053] See also Figure 1 , which shows a flow chart of the steps of a method for collecting marine animal husbandry environment data provided by an embodiment of the present invention, the method comprising the following steps:

[0054] Step S001: Collect ocean salinity data, ocean temperature data and ocean conductivity data.

[0055] It should be noted that when analyzing the marine environmental data, the collected data may be inaccurate due to interference from sound vibration or sensor performance during the data collection process. Therefore, the collected data is analyzed and processed to eliminate changes in the data caused by non-environmental factors.

[0056] Specifically, a salinity sensor is used to collect ocean salinity data within a day every 2 minutes, a temperature sensor is used to collect ocean temperature data within a day every 2 minutes, and an ocean conductivity sensor is used to collect industry conductivity data within a day every 2 minutes.

[0057] At this point, ocean salinity data, industry temperature data and ocean conductivity sensors are obtained.

[0058] Step S002: obtaining a local window of data at each moment in the ocean salinity data, obtaining the degree of fluctuation of the data amplitude in the local window of each moment in the ocean salinity data according to the difference between two adjacent moment data in the local window of each moment in the ocean salinity data, obtaining a local fitting curve of the data at each moment according to all moment data in the local window of each moment in the ocean salinity data, obtaining a local area difference of the data at each moment according to the local fitting curve of the data at each moment, and obtaining a credibility factor of the data at each moment in the ocean salinity data according to the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of each moment.

[0059] It should be noted that in the process of collecting ocean salinity data, if human pollution or industrial wastewater discharge causes the collected data to change, this change is gradual, that is, the amplitude difference does not change much during the change; when problems such as sound vibration or sensor performance cause the collected data to be abnormal during the data collection process, this data change is not gradual, that is, the amplitude difference changes randomly during the change, and there is no pattern to follow.

[0060] Specifically, a threshold value A is preset, wherein this embodiment is described by taking A=7 as an example, and this embodiment is not specifically limited, wherein A can be determined according to the specific implementation situation. Taking the data at each moment in the ocean salinity data as the center point of the local window, taking A as the number of data in the local window to obtain the local window of the data at each moment, the local window of the data at each moment in the ocean salinity data can be obtained; wherein, when the data on the left and right sides of the ocean salinity data do not meet the number of data in the local window, only the data that exists is used as the data of the local window.

[0061] (1) The fluctuation degree of the data amplitude in the local window of each moment data is obtained according to the difference between the two adjacent moment data in the local window of each moment data in the ocean salinity data.

[0062] It should be noted that when the marine environment changes due to human pollution or industrial wastewater discharge, the change in ocean salinity data is gradually increasing. Therefore, the changes in ocean salinity data caused by the two situations can be analyzed based on the fluctuations in the incremental increase of data at adjacent moments.

[0063] Specifically, the local amplitude sequence of each moment data is obtained according to the difference of adjacent moment data in the local window of each moment data in the ocean salinity data. The specific acquisition process of the local amplitude sequence of each moment data is: the difference between two adjacent moment data in the local window of each moment data in the ocean salinity data is calculated and the absolute value is obtained, and the local amplitude sequence of each moment data is obtained in chronological order. The degree of fluctuation of the data amplitude in the local window of each moment data is obtained according to the local amplitude sequence of each moment data in the ocean salinity data, which is expressed by the formula:

[0064]

[0065] In the formula, H i,j represents the value of the jth element in the local amplitude sequence of the i-th moment in the ocean salinity data, H i represents the mean value of all elements in the local amplitude sequence of the i-th moment data in the ocean salinity data, A represents the number of moment data in the local window, and Y i It represents the degree of fluctuation of the data amplitude in the local window of the data at the i-th moment, and || represents the absolute value sign.

[0066] Among them, the smaller the fluctuation of the data amplitude in the local window of the data at each moment, the more reliable the collected ocean salinity data is.

[0067] At this point, the fluctuation degree of the data amplitude in the local window of the ocean salinity data at each moment is obtained.

[0068] (2) A local fitting curve of each moment data is obtained based on all moment data in a local window of each moment data in the ocean salinity data, and a local area difference of each moment data is obtained based on the local fitting curve of each moment data.

[0069] It should be noted that when the monitored ocean salinity data is gradually increasing, it can be judged that the change in ocean salinity data is caused by human pollution or industrial wastewater discharge, and the marine environment is considered to have changed; when the monitored ocean salinity data changes randomly, it can be judged that the change in ocean salinity data is caused by sound vibration or sensor performance problems. Due to the interference of other factors, it is unclear whether the marine environment has changed. The ocean salinity data caused by human pollution or industrial wastewater discharge is gradually increasing, while the change in ocean salinity data caused by sound vibration or sensor performance is random. Therefore, by curve fitting the collected ocean salinity data, the curve fitted by the gradually increasing data is close to a straight line, while the change in ocean salinity data is random. The data is fitted as a curve, so the two situations can be analyzed and processed based on the two curves fitted by the data.

[0070] It should be further explained that when the collected data is gradually increasing, the fitted curve is close to Figure 2 The second one, that is, the more credible the data collected at this time; when the collected data changes irregularly, the fitted curve is Figure 2 The first one is caused by problems such as sound vibration or sensor performance, which means that the data collected at this time is unreliable. Figure 2 The curves fitted in different situations are quite different. Therefore, the area difference of the data at two adjacent moments can be obtained by subtracting the trapezoidal area of ​​the line connecting the data at adjacent moments from the curve area of ​​the data at adjacent moments. That is, the area difference of the reliable data collected is very small, while the area difference of the unreliable data collected is very large.

[0071] Specifically, any moment data in the ocean salinity data is recorded as the target moment data, and a quintic polynomial is used to perform curve fitting based on all moment data in a local window of the target moment data through the least squares method to obtain a local fitting curve of the target moment data; wherein, the least squares method is a well-known technology and will not be described in detail here; in this embodiment, a quintic polynomial is used for curve fitting, but the polynomial fitting is not specifically limited.

[0072] The local area difference of the data at each moment is obtained according to the local fitting curve of the data at each moment, which is expressed as follows:

[0073]

[0074] In the formula, f i (x) represents the function corresponding to the local fitting curve of the data at the i-th moment, and x represents the independent variable of the function corresponding to the local fitting curve of the data at each moment, where the independent variable is time, s i,c represents the c-th moment data in the local window of the i-th moment data, s i,c-1 represents the c-1th moment data in the local window of the i-th moment data, A represents the number of moment data in the local window, Z i It represents the local area difference of the data at the i-th moment, and || represents the absolute value sign.

[0075] Among them, when the difference between the curve area of ​​the data at adjacent moments and the trapezoidal area of ​​the line connecting the data at adjacent moments is smaller, the local area difference of the data at each moment is smaller, which means that the collected ocean salinity data is more reliable.

[0076] At this point, the local area difference of the ocean salinity data at each moment is obtained.

[0077] (3) The credibility factor of the data at each moment in the ocean salinity data is obtained based on the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of the data at each moment.

[0078] The credibility factor of the data at each moment in the ocean salinity data is obtained based on the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of the data at each moment, which is expressed by the formula:

[0079] T i =exp(-Y i ×Z i )

[0080] Where Y i Indicates the degree of fluctuation of the data amplitude in the local window of the i-th moment data, Z i Represents the local area difference of the data at the i-th moment, T i It represents the credibility factor of the data at the i-th moment in the ocean salinity data, and exp() represents an exponential function with a natural constant as the base.

[0081] Among them, when the fluctuation degree of the data amplitude in the local window of the data at each moment is greater and the local area difference of the data at each moment is greater, the credibility factor of the data at each moment in the ocean salinity data is smaller, which means that the collected ocean salinity data is less credible.

[0082] So far, the credibility factor of the data at each moment in the ocean salinity data.

[0083] Step S003: obtaining a first credibility according to a ratio of the ocean temperature data to the ocean salinity data, and obtaining a second credibility according to a ratio of the ocean salinity data to the ocean conductivity data.

[0084] It should be noted that when the water temperature rises, the thermal motion of water molecules increases, which increases the molecular distance of solutes (salts), resulting in a decrease in solubility, that is, the number of salts dissolved in seawater decreases relatively, resulting in a decrease in salinity. Therefore, the change in ocean temperature data and the change in salinity data show a negative correlation, that is, the more negatively correlated the temperature data and salinity data are within the corresponding range, the higher the confidence level of the salinity data.

[0085] It should be further explained that when there is a negative correlation between ocean salinity data and ocean temperature data, when the temperature gradually rises, the ratio between the temperature data and the salinity data at each moment will gradually increase, and the growth is close to exponential growth. The curve fitted by the ratio between the temperature data and the salinity data is as follows: Figure 3 (a) in the figure; when the collected ocean salinity data is due to the performance of the sensor equipment, there is no negative correlation between the ocean salinity data and the ocean temperature data. That is, when the temperature gradually rises, the ratio between the temperature data and the salinity data at each moment changes randomly without any specific rules. The curve fitted by the ratio between the temperature data and the salinity data is as follows: Figure 3 (b) in the figure. The ratio of the temperature data to the salinity data at the first moment and the last moment is connected as a time straight line, and the analysis is performed based on the area difference between the time straight line and the fitting curve of the ratio of the temperature data to the salinity data.

[0086] Specifically, a temperature-salinity ratio sequence is obtained based on the ratio between the ocean temperature data and the ocean salinity data at each moment, and a temperature-salinity ratio fitting curve is obtained by the least square method based on the ratio data at all moments in the temperature-salinity ratio sequence. In this embodiment, a fifth-order polynomial is used for curve fitting, but polynomial fitting is not specifically limited. Then, the ratio data at the first moment and the ratio data at the last moment are connected to obtain a time straight line, and the difference between the areas respectively formed with the horizontal axis is obtained based on the temperature-salinity ratio fitting curve and the time straight line, and the first credibility is obtained based on the difference between the areas, which is expressed as:

[0087]

[0088] Wherein, l(x) represents the function corresponding to the time line, g(x) represents the function corresponding to the temperature-salinity ratio fitting curve, x represents the independent variable of the function corresponding to the time line and the function corresponding to the temperature-salinity ratio fitting curve, where the independent variable is time, t1 represents the time of the first moment, t2 represents the time of the last moment, M represents the number of intersections between the time line and the temperature-salinity ratio fitting curve, and δ1 represents the first credibility.

[0089] Among them, the greater the difference between the areas, the more credible the collected ocean salinity data is; and the greater the number of intersections between the straight line and the curve, the less credible the collected ocean salinity data is.

[0090] At this point, the first credibility is obtained.

[0091] It should be noted that ocean conductivity refers to the ability of seawater to conduct electricity. The main dissolved substances in the ocean are salts, which decompose into charged ions in water. The presence of these ions makes seawater have good conductivity. Therefore, there is a close relationship between ocean conductivity data and ocean salinity data. Generally speaking, there is an approximately linear relationship between ocean conductivity and salinity. When the salinity in seawater increases, the amount of dissolved salt ions increases, and the conductivity also increases accordingly, that is, there is a positive correlation between the two.

[0092] It should be further explained that when there is a positive correlation between the ocean conductivity data and the ocean salinity data, it indicates that the collected ocean salinity data is more reliable. Therefore, when there is a positive correlation between the ocean conductivity data and the ocean salinity data, the fluctuation range of the ratio of the two data should not be large. The smaller the fluctuation range of the data ratio is, the more reliable the collected ocean salinity data is. Therefore, the reliability of the collected ocean salinity data can be analyzed according to the fluctuation range of the data ratio, which is the second reliability.

[0093] Specifically, according to the ratio between the ocean salinity data and the ocean conductivity data at each moment, a conductivity-salinity ratio sequence is obtained, and the second credibility is obtained according to the conductivity-salinity ratio sequence, which is expressed by the formula:

[0094] δ2=exp(-D)

[0095] Where D represents the variance of all data in the conductivity-salinity ratio series, exp() represents an exponential function with a natural constant as the base, and δ2 represents the second credibility.

[0096] Among them, when the variance of all data in the conductivity-salinity ratio sequence is smaller, the second credibility is greater, indicating that the collected ocean salinity data is more credible.

[0097] At this point, the second credibility is obtained.

[0098] Step S004: Obtain a correction weight for each moment of the ocean salinity data according to the credibility factor, the first credibility and the second credibility of the data at each moment of the ocean salinity data; correct the distance between every two moments of the data according to the correction weight of the data at each moment of the ocean salinity data to obtain the distance between every two moments of the data after correction; and complete the collection of the ocean salinity data according to the distance between every two moments of the data after correction.

[0099] The correction weight of the data at each moment in the ocean salinity data is obtained according to the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data, which is expressed by the formula:

[0100] U i =1-norm(T i ×δ1×δ2)

[0101] Where, T i represents the credibility factor of the data at the i-th moment in the ocean salinity data, δ1 represents the first credibility, δ2 represents the second credibility, and U i It represents the correction weight of the data at the i-th moment in the ocean salinity data, and norm() represents the linear normalization function.

[0102] Among them, when the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data are larger, it means that the data collected at this moment is more credible, that is, when detecting abnormal data through the LOF algorithm, the distance between it and other data should be reduced, and it should be classified as normal data; when the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data are smaller, it means that the data collected at this moment is less credible, that is, when detecting abnormal data through the LOF algorithm, the distance between it and other data should be increased, and it should be classified as abnormal data.

[0103] When using the LOF algorithm for anomaly detection, the correction coefficient of the distance between every two moments of data is equal to the average of the correction weights of every two moments of data. The distance between every two moments of data after correction is obtained by multiplying the correction coefficient of the distance between every two moments of data by the distance between every two moments of data.

[0104] A threshold value B is preset, wherein this embodiment is described by taking B=5 as an example, and this embodiment is not specifically limited, wherein B can be determined according to the specific implementation situation. The LOF algorithm is used to detect abnormal data in the ocean salinity data based on the distance between the data at every two moments after correction, and the number of abnormal data is obtained, which is recorded as G. When the number of abnormal data G is greater than or equal to the preset threshold value B, the collected ocean salinity data is unreliable, and the ocean salinity data should be recollected; when the number of abnormal data G is less than the preset threshold value B, the collected ocean salinity data is credible, and the collected ocean salinity data can be used to analyze changes in the marine livestock environment. Among them, the LOF algorithm is a well-known technology and will not be described in detail here.

[0105] This embodiment provides a marine livestock environment data acquisition system, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements steps S001 to S004 when executing the computer program.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for collecting marine animal husbandry environment data, characterized in that: The method comprises the following steps: Collect ocean salinity data, ocean temperature data, and ocean conductivity data; Obtain a local window of data at each moment in the ocean salinity data, obtain the degree of fluctuation of the data amplitude in the local window of each moment in the ocean salinity data according to the difference between two adjacent moment data in the local window of each moment in the ocean salinity data, obtain a local fitting curve of the data at each moment in the local window of the data at each moment in the ocean salinity data according to all moment data, obtain a local area difference of the data at each moment according to the local fitting curve of the data at each moment, and obtain a credibility factor of the data at each moment in the ocean salinity data according to the local area difference of the data at each moment and the degree of fluctuation of the data in the local window of the data at each moment; The first credibility is obtained based on the ratio of the ocean temperature data to the ocean salinity data, and the second credibility is obtained based on the ratio of the ocean salinity data to the ocean conductivity data; Obtaining a correction weight of the data at each moment in the ocean salinity data according to the credibility factor, the first credibility and the second credibility of the data at each moment in the ocean salinity data, correcting the distance between every two moments of data according to the correction weight of the data at each moment in the ocean salinity data, obtaining the distance between every two moments of data after correction, and completing the collection of the ocean salinity data according to the distance between every two moments of data after correction; Among them, the method for obtaining the local area difference of the data at each moment is: Record any time data in the ocean salinity data as the target time data, and obtain the local fitting curve of the target time data by the least square method according to all time data in the local window of the target time data; The calculation formula for the local area difference of the data at each moment is: In the formula, f i (x) represents the function corresponding to the local fitting curve of the data at the i-th moment, x represents the independent variable of the function corresponding to the local fitting curve of the data at each moment, s i,c represents the c-th moment data in the local window of the i-th moment data, s i,c-1 represents the c-1th moment data in the local window of the i-th moment data, A represents the number of moment data in the local window, Z i represents the local area difference of the data at the i-th moment, and || represents the absolute value sign; The method for obtaining the first credibility is as follows: According to the ratio between the ocean temperature data and the ocean salinity data at each moment, a temperature-salinity ratio sequence is obtained, and the temperature-salinity ratio fitting curve is obtained by the least square method according to the ratio data of all moments in the temperature-salinity ratio sequence, and then the ratio data of the first moment and the ratio data of the last moment are connected to obtain a time straight line; The calculation formula of the first credibility is: Wherein, l(x) represents the function corresponding to the time line, g(x) represents the function corresponding to the temperature-salinity ratio fitting curve, x represents the independent variable of the function corresponding to the time line and the function corresponding to the temperature-salinity ratio fitting curve, t1 represents the time of the first moment, t2 represents the time of the last moment, M represents the number of intersections between the time line and the temperature-salinity ratio fitting curve, and δ1 represents the first credibility; The second credibility is obtained by: According to the ratio between the ocean salinity data and the ocean conductivity data at each moment, a conductivity-salinity ratio series is obtained; The calculation formula of the second credibility is: δ2=exp(-D) Where D represents the variance of all data in the conductivity-salinity ratio series, exp() represents an exponential function with a natural constant as the base, and δ2 represents the second credibility.

2. A method for collecting marine animal husbandry environment data according to claim 1, characterized in that: The specific steps of obtaining the local window of the ocean salinity data at each moment are as follows: The data at each moment in the ocean salinity data is taken as the center point of the local window, and A is taken as the number of data in the local window to obtain the local window of the data at each moment; wherein A is a preset threshold.

3. A method for collecting marine animal husbandry environment data according to claim 1, characterized in that: The method of obtaining the fluctuation degree of the data amplitude in the local window of each moment data according to the difference between two adjacent moment data in the local window of each moment data in the ocean salinity data includes the following specific steps: The calculation formula for the fluctuation degree of the data amplitude in the local window of each moment is: In the formula, H i,j represents the value of the jth element in the local amplitude sequence of the i-th moment in the ocean salinity data, H i represents the mean value of all elements in the local amplitude sequence of the i-th moment data in the ocean salinity data, A represents the number of moment data in the local window, and Y i Indicates the degree of fluctuation of the data amplitude in the local window of the data at the i-th moment, || indicates the absolute value sign; The difference between two adjacent data points in the local window of each moment in the ocean salinity data is calculated and the absolute value is obtained to obtain the local amplitude sequence of each moment in time sequence.

4. A method for collecting marine animal husbandry environment data according to claim 1, characterized in that: The specific steps of obtaining the credibility factor of the data at each moment in the ocean salinity data according to the local area difference of the data at each moment and the fluctuation degree of the data in the local window of the data at each moment are as follows: The calculation formula for the credibility factor of the ocean salinity data at each moment is: T i =exp(-Y i ×Z i ) Where Y i Indicates the degree of fluctuation of the data amplitude in the local window of the i-th moment data, Z i Represents the local area difference of the data at the i-th moment, T i It represents the credibility factor of the data at the i-th moment in the ocean salinity data, and exp() represents an exponential function with a natural constant as the base.

5. The method for collecting marine animal husbandry environment data according to claim 1, characterized in that: The specific steps of obtaining the correction weight of the data at each moment in the ocean salinity data according to the credibility factor of the data at each moment in the ocean salinity data, the first credibility and the second credibility are as follows: The calculation formula for the correction weight of the data at each moment in the ocean salinity data is: U i =1-norm(T i ×δ1×δ2) Where, T i represents the credibility factor of the data at the i-th moment in the ocean salinity data, δ1 represents the first credibility, δ2 represents the second credibility, and U i It represents the correction weight of the data at the i-th moment in the ocean salinity data, and norm() represents the linear normalization function.

6. A method for collecting marine animal husbandry environment data according to claim 1, characterized in that: The distance between each two moments of data is corrected according to the correction weight of each moment of data in the ocean salinity data to obtain the distance between each two moments of data after correction, and the collection of ocean salinity data is completed according to the distance between each two moments of data after correction, including the following specific steps: The correction coefficient of the distance between each two moment data is equal to the average of the correction weights of each two moment data. The distance between each two moment data after correction is obtained by multiplying the correction coefficient of the distance between each two moment data by the distance between each two moment data. The LOF algorithm is used to detect abnormal data in the ocean salinity data based on the distance between the data at every two moments after correction, and the number of abnormal data is obtained, which is recorded as G; when the number of abnormal data G is greater than or equal to the preset threshold value B, the collected ocean salinity data is unreliable and the ocean salinity data should be recollected; when the number of abnormal data G is less than the preset threshold value B, the collected ocean salinity data is credible.

7. A marine livestock environment data collection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a marine livestock environment data collection method as described in any one of claims 1-6 are implemented.

Images