CN115879750A - Aquatic seedling raising environment monitoring management system and method - Google Patents
Aquatic seedling raising environment monitoring management system and method Download PDFInfo
- Publication number
- CN115879750A CN115879750A CN202310187538.7A CN202310187538A CN115879750A CN 115879750 A CN115879750 A CN 115879750A CN 202310187538 A CN202310187538 A CN 202310187538A CN 115879750 A CN115879750 A CN 115879750A
- Authority
- CN
- China
- Prior art keywords
- prediction
- index
- reference index
- category
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
- Y02A40/81—Aquaculture, e.g. of fish
Abstract
The invention relates to the technical field of data processing, in particular to an aquatic breeding environment monitoring and management system and method, which classify all historical reference indexes according to an environment influence factor set to obtain matching index classes; classifying the historical reference indexes of the matching index categories according to the frequency values and the linear offset degrees to obtain two characteristic categories, and obtaining adjustment parameters by combining the frequency values of the historical reference indexes in the matching index categories; and constructing a predictive linear regression equation according to the reference index data variation and the adjusting parameters, carrying out nonlinear conversion on the predictive linear regression equation to obtain the predictive probability value of each predictive reference index, and carrying out weighting processing on all the predictive probability values to obtain the final predictive reference index, thereby realizing accurate monitoring on the aquatic seedling raising condition at the future date. According to the invention, the problem of nonlinear water quality parameter prediction accuracy is optimized, the practicability of nonlinear data is higher, and the accuracy of predicting the water quality change of the aquatic seedling raising environment is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an aquatic seedling raising environment monitoring and management system and method.
Background
In the aquatic breeding and culture, the quality of water quality management not only influences the production performance of fishes and shrimps, but also influences the composition and richness of bait organisms. With the popularization of aquaculture and the aggravation of water pollution, the survival rate of aquaculture seedlings is low and the yield of aquaculture is reduced due to the fact that many aquaculture farms are lack of water quality management conditions and experiences and management efficiency is limited. The current environmental data that influences the quality of water change through various thing networking sensors replace manual monitoring to utilize artificial intelligence to carry out analysis, processing and feedback to environmental data, realize the high-efficient management of aquaculture, plant's quality of water, nevertheless real-time regulation can't satisfy the promptness of water quality management, regulation, need change the quality of water and predict and prevent, just can reduce the aquatic products loss and the influence that the quality of water change leads to minimumly. The existing water quality change prediction mode is to construct a prediction model according to an influence relation through autocorrelation analysis between influence factors and water quality parameters to obtain a predicted value, but the method is only suitable for linear water quality parameters such as water temperature and transmittance, and partial water quality parameters are nonlinear changes, so that the practicability of the prediction result of the partial water quality parameters is poor.
In the prior art, all historical data are analyzed to obtain experience data of the fishes to be predicted, and no matching data matched with the experience data of the fishes to be predicted in the historical data is specifically analyzed, so that a large error is generated, and the accuracy of a final prediction result is poor. In the prior art, the association degree between each water quality environment data variable and a fish health growth index is obtained in a mean value mode, and the water quality environment data variable with higher association degree with the fish health growth index is selected as a key water quality environment data variable influencing the growth of fish; and regulating and controlling the water quality in the aquaculture system according to the trained predictive regulation and control model. When the correlation degree is obtained, an averaging method is used, the prediction of the nonlinear data is rough, so that the final prediction result has errors, and the prediction result is not accurate enough.
Disclosure of Invention
In order to solve the technical problem that the reference index prediction of water quality is not accurate enough in the prior art, the invention aims to provide an aquatic seedling environment monitoring and management system and method, and the adopted technical scheme is as follows:
the invention provides an aquatic seedling raising environment monitoring and management method, which comprises the following steps:
obtaining a historical reference index and a corresponding environmental influence factor set of each historical date of aquaculture; the historical reference indexes of each historical date correspond to the environmental influence factor sets one by one; obtaining a set of environmental impact factors for a future date;
classifying all historical reference indexes according to the environmental influence factor set of each historical date and the future date to obtain at least two reference index categories, and taking the reference index category to which the environmental influence factor set of the future date belongs as a matching index category;
classifying all historical reference indexes according to the frequency value of each historical reference index in the matching index class to obtain two characteristic classes; obtaining adjusting parameters of corresponding historical reference indexes according to the feature class to which each historical reference index in the matching index class belongs and the frequency value;
obtaining numerical value categories of all historical reference indexes in the matching index categories; taking the numerical value corresponding to each numerical value category as a prediction reference index, and sequentially inserting the prediction reference index into the matching index category to obtain the reference index data variation of the matching index category before and after insertion; for any one of the prediction reference indexes, constructing a prediction linear regression equation according to the adjusting parameters of the prediction reference indexes and the reference index data variation;
carrying out nonlinear transformation on the predictive linear regression equation to obtain the predictive probability value of each predictive reference index; weighting each prediction reference index according to the prediction probability value to obtain a final prediction reference index; and monitoring and managing the aquatic seedling raising environment in the future date according to the final prediction reference index.
Further, the set of environmental impact factors includes:
the environmental indexes comprise air temperature, water vapor pressure, cloud cover, air pressure, wind speed and sunshine hours; and taking all the environmental indexes of each date as the environmental influence factor set of the corresponding date.
Further, the method for acquiring the feature category comprises the following steps:
the feature classes comprise a first feature class and a second feature class;
sorting the frequency values of all historical reference indexes in the matching index category according to the sequence of the frequency values from large to small to obtain a frequency sequence; accumulating the first frequency value in the frequency sequence in sequence until the accumulated value is larger than a preset ideal threshold value;
dividing all historical reference indexes corresponding to the frequency values participating in the accumulation operation into categories to be classified, eliminating the historical reference index corresponding to the last frequency value participating in the accumulation operation from the categories to be classified, and taking the updated categories to be classified as the first characteristic categories; and classifying all historical reference indexes belonging to the non-first characteristic category in the matching index category into the second characteristic category.
Further, the method for obtaining the adjustment parameter includes:
for any historical reference index in the matching index class, if the historical reference index is in a first characteristic class, the adjusting parameter corresponding to the historical reference index is a preset stable parameter;
and if the historical reference index is in the second characteristic category, taking the ratio of the frequency value of the corresponding historical reference index to the ideal threshold value as a negative adjusting parameter, and subtracting the negative adjusting parameter from the constant one to obtain the adjusting parameter of the corresponding historical reference index.
Further, the method for acquiring the reference index data variation includes:
taking the standard deviation of all historical reference indexes in the matching index category as a first discrete degree of the matching index category before insertion;
taking the numerical value corresponding to each numerical value category as a prediction reference index, sequentially inserting the prediction reference index into the matching index category, forming a prediction category by all historical reference indexes in the matching index category and the prediction reference index, and taking the standard deviation of the prediction category as the prediction dispersion degree of the inserted matching index category;
obtaining a difference absolute value between the predicted dispersion degree of each inserted matching index category and the first dispersion degree of each inserted matching index category, and taking the difference absolute value as the reference index data variation of the matching index categories before and after insertion; each of the prediction reference indexes corresponds to one of the reference index data variation amounts.
Further, the construction method of the predictive linear regression equation comprises the following steps:
acquiring the variable quantity of reference index data corresponding to the prediction reference index of the matching index type after each insertion, and taking the frequency value of the matching index type after the insertion, which is subtracted from the constant 1, as the independent variable of the prediction linear regression equation; and taking the adjusting parameters as error terms, and taking the reference index data variation as linear regression parameters.
Further, the method for obtaining the prediction probability value comprises the following steps:
carrying out nonlinear conversion on the prediction linear regression equation by using a Sigmoid function to obtain a corresponding first prediction equation;
taking the numerical value corresponding to each numerical value category as a prediction reference index, inputting each prediction reference index into the first prediction equation, and outputting the sub-prediction probability of the prediction reference index; and obtaining the prediction probability value of the prediction reference index according to the absolute value of the difference value between the constant one and the sub prediction probability.
Further, the method for obtaining the final prediction reference index includes:
accumulating the prediction probability values of all the prediction reference indexes to obtain a comprehensive prediction probability value;
obtaining the ratio of the prediction probability value of each prediction reference index to the comprehensive prediction probability value, and taking the ratio as the sub-weight of the prediction reference index; taking the product of the sub-weight and the prediction reference index value as a sub-prediction index of the prediction reference index; and accumulating the sub prediction indexes of all the prediction reference indexes to obtain the final prediction reference index.
The invention also provides an aquatic product seedling raising environment monitoring and management system which comprises a memory and a processor, wherein the processor executes a computer program stored in the memory so as to realize the aquatic product seedling raising environment monitoring and management method.
The invention has the following beneficial effects:
classifying all historical reference indexes according to an environment influence factor set, and taking a reference index class to which the environment influence factor set of a future date belongs as a matching index class; according to the method, similar multidimensional environment influence factors are introduced, the relevance of the prediction reference index and the historical reference index is improved, and the prediction result is more accurate. All historical reference indexes are classified according to the frequency values to obtain two characteristic classes, the frequency values reflect the influence of the historical reference indexes on the prediction reference indexes, therefore, the influence weight between the two characteristic classes is different through classification, the prediction reference indexes are analyzed by combining the weight of the influence of the historical reference indexes on the prediction reference indexes, and the obtained prediction effect is more accurate. And acquiring an adjusting parameter according to the characteristic category and the frequency value of each historical reference index in the matching index category, wherein the adjusting parameter can predict the degree of fit between the reference index and the historical reference index of each historical date, and provides a basis for a subsequently acquired predictive linear regression equation. The reference index data variation reflects the difference of the discrete degree between the matching index category after insertion and the matching index category before insertion, and the structural change of the matching index category before and after insertion can be more intuitively represented. And constructing a predictive linear regression equation according to the adjustment parameters and the reference index data variation, carrying out nonlinear conversion on the predictive linear regression equation to obtain a predictive probability value, and carrying out weighting processing on each predictive reference index according to the predictive probability value to obtain a final predictive reference index, thereby realizing accurate monitoring of the aquatic seedling raising condition at the future date. According to the invention, by optimizing the problem of the prediction accuracy of the nonlinear water quality parameters, the practicability of nonlinear data is higher, the prediction result is more accurate, and the accuracy of predicting the water quality change of the aquatic seedling raising environment is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flow chart of a method for monitoring and managing an aquatic breeding environment according to an embodiment of the present invention.
Detailed Description
In order to further illustrate the technical means and effects of the present invention adopted to achieve the predetermined purpose, the following detailed description will be given to the system and method for monitoring and managing aquatic product seedling environment according to the present invention, and the specific implementation manner, structure, characteristics and effects thereof, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
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.
The concrete scheme of the system and the method for monitoring and managing the aquatic breeding environment provided by the invention is concretely described below by combining the attached drawings.
Referring to fig. 1, it shows a flow chart of a monitoring and management method for aquatic breeding environment according to an embodiment of the present invention, the method includes:
step S1: obtaining a historical reference index and a corresponding environmental influence factor set of each historical date of aquaculture; the historical reference indexes of each historical date correspond to the environmental influence factor sets one by one; a set of environmental impact factors for a future date is obtained.
In the embodiment of the invention, the concrete implementation scene is the regulation and control of the water quality parameters in the aquaculture process. The water quality parameter indexes of aquaculture include temperature, pH value, salinity, ammonia nitrogen, hydrogen sulfide, nitrite, available phosphorus, transparency, dissolved oxygen and the like, and water quality management needs to adjust the water quality according to the water quality parameter indexes.
In order to enhance the timeliness and accuracy of water quality parameter prediction, the characteristics of basic water quality parameter indexes in each period need to be analyzed, the change of the water quality parameter indexes is influenced by external factors and internal factors of the seedling raising pond, but the internal factors of the seedling raising pond have a plurality of water quality parameter indexes which are mutually interfered, and if the internal factors are used as the influencing factors for subsequent prediction, the reliability of the internal factors can be greatly reduced. Therefore, in the embodiment of the invention, a certain water quality parameter index is analyzed, the dissolved oxygen concentration is selected as the historical reference index of analysis, the historical reference index of each historical date of aquaculture and the corresponding environmental influence factor set are obtained, and an implementer can determine the historical reference index of analysis according to a specific implementation scene. Further, the environmental indexes of each historical period comprise air temperature, water vapor pressure, cloud cover, air pressure, wind speed and sunshine hours, all the environmental indexes of each date are used as the environmental influence factor sets of the corresponding date, and the historical reference indexes of each historical date correspond to the environmental influence factor sets one by one. In other embodiments, the operator may select other types of indexes as the reference indexes, and the other types of indexes are the same as the method for analyzing the dissolved oxygen concentration according to the embodiment of the present invention, and are not described again.
In order to improve the timeliness and accuracy of water quality parameter prediction, the prediction reference index of the future date can be predicted only by acquiring the environmental influence factor set of the future date, and the environmental influence factor set of the future date can be directly acquired from the weather website.
Step S2: and classifying all the historical reference indexes according to the environmental influence factor sets of each historical date and the future date to obtain at least two reference index categories, and taking the reference index category to which the environmental influence factor set of the future date belongs as a matching index category.
The historical reference indices and the environmental influence factor sets for all the historical dates, and the environmental influence factor sets for the future dates are obtained by step S1. For the reference indexes, the reference indexes are influenced by environmental indexes such as air temperature, water vapor pressure, cloud cover, wind speed and the like, the environmental influence factor set of the future date is obtained through a meteorological website, and therefore when the prediction reference indexes of the future date of the nursery pond are predicted, the causal connection between the prediction reference indexes of the future date and the historical reference indexes of the historical date is established in a time series analysis mode.
The prediction reference indexes of the future date are predicted, and the history reference indexes corresponding to the environmental influence factor sets of the history dates which are the same as or similar to the future date need to be classified, so that the prediction reference indexes of the future date can be accurately predicted by analyzing the history reference indexes corresponding to the history dates under the condition that the multidimensional environmental influence factors are the same. Therefore, all historical reference indicators are classified according to the environmental influence factor set of each historical date and future date, and at least two reference indicator categories are obtained. In the embodiment of the invention, the method for classifying the historical reference indexes of all historical dates is a Gaussian mixture model, and the Gaussian mixture model is generally used for clustering multidimensional data. The gaussian mixture model is a technical means well known to those skilled in the art, and the specific process is not described herein, and only the clustering process of the gaussian mixture model provided by one embodiment of the present invention is briefly described here:
the first step is as follows: and inputting the historical reference indexes of each historical date and the corresponding environmental influence factor sets, adding the environmental influence factor sets of the future dates into the environmental influence factor sets corresponding to the historical reference indexes of the historical dates, and classifying the historical reference indexes of all the historical dates according to all the environmental influence factor sets. The method comprises the steps of taking an environment influence factor set as an independent variable x and historical reference indexes as a dependent variable y, establishing a plane rectangular coordinate system according to the independent variable and the dependent variable, obtaining data points corresponding to each historical date by the independent variable x and the dependent variable y, and classifying all the data points according to a single-Gaussian density distribution condition which is met by each data point in a Gaussian mixture model, namely classifying all the historical reference indexes.
The second step: setting the number of components of a preset initialized Gaussian mixture model; the gaussian distribution parameters, i.e. mean and variance, of each reference index class are randomly initialized. In the embodiment of the present invention, the number of components of the initialized gaussian mixture model is preset to be 6, and the specific numerical value may be specifically set according to the specific implementation manner.
The third step: and calculating the probability of each Gaussian model to which the historical reference index of each historical date belongs, namely calculating the posterior probability. The closer the historical reference index of each historical date is to the center of the Gaussian distribution, the higher the posterior probability is, that is, the higher the probability that the historical reference index of the corresponding historical date belongs to the reference index category.
The fourth step: calculating prior distribution probability, mean and covariance parameters to maximize the probability of the historical reference index of each historical date, calculating new parameters by using the probability weighting of the historical reference indexes of the historical dates, wherein the weight is the probability that the historical reference index of the corresponding historical date belongs to the reference index category.
The fifth step: and repeating the second step and the third step until convergence.
And a sixth step: and outputting a plurality of reference index category results, wherein the environmental influence factor sets corresponding to the historical reference indexes of all historical dates in each reference index category are similar. The predicted reference indices for the future date are then also categorized into a reference index category that matches the set of environmental impact factors for the future date.
The environmental influence factor sets of the historical reference indexes in each reference index category obtained through the classification of the Gaussian mixture model are similar, but the corresponding historical reference indexes are different, namely the historical reference indexes in the reference index categories are nonlinear data, and although the prediction reference indexes are not yet known, the corresponding prediction reference indexes are also classified into the matching index categories according to the clustering mode of the multidimensional environmental influence factor sets no matter the numerical values of the reference indexes at the future date are. Therefore, under the condition that the prediction reference index of the future date is unknown, the prediction reference index of the future date is supposed to be inserted into the matching index category, but when the prediction reference index is required to be limited to be inserted, the structural change of the matching index category needs to be as small as possible, so that the maximum posterior probability of the history reference index of the history date in the matching index category to the prediction reference index can be ensured to the maximum extent, and the more accurate prediction reference index of the future date can be obtained. Since the prediction reference index of the future date is unknown, the prediction reference index of the future date needs to be predicted according to the historical reference index of the historical date in the same reference index category.
Therefore, the reference index category to which the environmental influence factor set of the future date belongs is taken as the matching index category, and the prediction reference index of the future date needs to be predicted subsequently according to the historical reference index of the historical date in the matching index category, so that the accuracy of the prediction reference index of the future date can be improved.
And step S3: classifying all historical reference indexes according to the frequency value of each historical reference index in the matching index class to obtain two characteristic classes; and obtaining the adjusting parameters of the corresponding historical reference indexes according to the feature class and the frequency value of each historical reference index in the matching index class.
Step S2 shows that the prediction reference index of the future date needs to be predicted according to the historical reference index of the historical date in the matching index category, and a logistic regression model is introduced to predict the value of the prediction reference index. It should be noted that the logistic regression model is a technical means well known to those skilled in the art, and details are not repeated herein, and only the basic principle of the logistic regression model is briefly described: the nature of the logistic regression model is that the linear regression model is subjected to nonlinear conversion through a Sigmoid function to obtain a probability value between 0 and 1. The nature of the logistic regression model is the prediction probability, and the linear regression equation obtained according to the original algorithm is the causal relationship between the historical reference index of the historical date and a certain environmental index, but the environmental influence factor set is a multi-dimensional parameter in the invention, and even if the multi-dimensional environmental influence factor set is changed into one-dimensional data in a data dimension reduction mode, the final output probability value is inaccurate. Therefore, in the embodiment of the present invention, the prediction reference index is iterated according to all the historical reference indexes in the matching index category, the prediction reference index is inserted into the matching index category, the data change of the reference index in the matching index category after the prediction reference index is inserted is obtained, and the causal relationship of the linear regression equation is reconstructed.
In the embodiment of the invention, the argument value constructed in the linear regression equation is the frequency value of a certain type of historical reference index in the matching index class, when the prediction reference index is any one type of historical reference index in the matching index class, compared with the original matching index class, the structure of the matching index class containing the prediction reference index changes, the influence of the prediction reference indexes with different values on the structure of the matching index class containing the prediction reference index is different, and compared with the matching index class not containing the prediction reference index, the linear regression line corresponding to the matching index class containing the prediction reference index shifts.
Therefore, the change of the matching index class structure containing the prediction reference index can feed back the degree of matching between the prediction reference index and the historical reference index of each historical date. Because the traditional logistic regression carries out likelihood estimation on the output result after carrying out nonlinear transformation on the linear equation by using the Sigmoid function to obtain the two-classification probability result, although the invention does not need to carry out two-classification, the invention can also be regarded as that the data in the matching index category is divided into two types with higher influence weight and lower influence weight on the prediction reference index. Therefore, in order to make the prediction effect on the prediction reference index more accurate and determine the accurate range of the value, all the history reference indexes are classified according to the frequency value of each history reference index in the matching index category to obtain two characteristic categories, the embodiment of the invention specifically comprises the following steps:
firstly, the frequency value of each historical reference index in the original matching index category needs to be obtained, and the larger the frequency value is, the more the times of occurrence of the corresponding historical reference index in the original matching index category are. The weight of influence of the historical reference indexes with higher frequency values in the original matching index category on the prediction reference indexes is higher; the historical reference indexes with smaller frequency values in the original matching index category have lower weight on the influence of the prediction reference indexes. When the prediction reference index is a historical reference index with a large frequency value in the original matching index category, the linear regression offset change of the matching index category after the prediction reference index is inserted is small; when the prediction reference index is a historical reference index with a small frequency value in the original matching index category, the offset of linear regression of the matching index category after the prediction reference index is inserted may be large, the frequency value corresponding to the historical reference index with the small frequency value is changed from small to large due to the fact that the same type of prediction reference index is inserted into the historical reference index with the small frequency value, and the type with the low weight and the influence on the prediction reference index is changed into the type with the high weight and the influence on the prediction reference index.
Secondly, determining screening conditions of different feature categories according to the relationship between the frequency values of each historical reference index in the original matching index category, wherein the screening conditions specifically comprise the following steps: sequencing the frequency values of all historical reference indexes in the matching index category according to the sequence of the frequency values from large to small to obtain a frequency sequence; and accumulating the frequency sequences in sequence from the first frequency value in the frequency sequences until the accumulated value is greater than a preset ideal threshold value. Classifying all historical reference indexes corresponding to the frequency values participating in the accumulation operation into categories to be classified, eliminating the historical reference index corresponding to the last frequency value participating in the accumulation operation from the categories to be classified, and taking the updated categories to be classified as first feature categories; and classifying all historical reference indexes belonging to non-first characteristic classes in the matching index classes into second characteristic classes, wherein the characteristic classes comprise the first characteristic class and the second characteristic class. The first characteristic category is a category with higher weight which has influence on the prediction reference index in the original matching index category; the second feature class is a class having a low weight that affects the prediction reference index within the original matching index class. In the embodiment of the present invention, the preset ideal threshold is 50%, and the value of the specific ideal threshold may be specifically set according to a specific implementation scenario. In order to more intuitively express all the reference index classifications in the matching index categories, the conditional formula of the reference index classification specifically includes:
in the formula (I), the compound is shown in the specification,
representing the number of historical reference indexes contained in an accumulated value which is accumulated from the first frequency value in sequence and is smaller than an ideal threshold value;
representing historical reference index
The frequency value in the original match indicator category,
it is indicative of an ideal threshold value for the threshold,
indicating that a maximum function is to be found. For example, the ideal threshold value is preset to be 50%, and the first three values are accumulated in sequence from the first frequency value in the frequency sequenceThe accumulated value of the historical reference index is 48%, and the accumulated values of the first four historical reference indexes are 51%. Therefore, the maximum accumulated value smaller than the ideal threshold value at this time contains 3 historical reference index numbers, and at this time
I.e. by
Then, the weight of the influence of the first three historical reference indexes on the prediction reference index is higher, namely the first characteristic category; the remaining historical reference indicators in the frequency series have a lower weight on the influence of the prediction reference indicator, i.e. the second characteristic class.
Since the prediction reference index is of the first feature type or the second feature type, the influence on the structure of the matching index type including the prediction reference index is also different. And the change of the matching index class structure containing the prediction reference index can feed back the degree of fit between the prediction reference index and the historical reference index of each historical date, so that an adjusting factor of the prediction reference index can be constructed according to the feature class to which each historical reference index belongs, and a prediction linear regression equation can be obtained in the subsequent process only by combining the adjusting factor. Therefore, the obtaining of the adjustment parameter corresponding to the historical reference index according to the feature class and the frequency value to which each historical reference index in the matching index class specifically includes:
for any historical reference index in the matching index category, if the historical reference index is in a first characteristic category, the adjusting parameter corresponding to the historical reference index is a preset stable parameter; when the prediction reference index is of the first characteristic type, the linear equation of the matching index type containing the prediction reference index is almost not shifted or shifted slightly, the weight of the influence of the historical reference index of the first characteristic type on the prediction reference index is originally large, and the influence weight of the similar prediction reference index after insertion is only larger, so that the design expectation is met. Therefore, in the embodiment of the present invention, the preset stability parameter is set to be a value 0, and the specific value of the preset stability parameter may be specifically set according to a specific implementation scenario.
If the historical reference index is in the second characteristic category, the ratio of the frequency value of the corresponding historical reference index to the ideal threshold value is used as a negative adjusting parameter, and the negative adjusting parameter is subtracted from the constant one to obtain the adjusting parameter of the corresponding historical reference index. And when the prediction reference index is in the second characteristic type, the linear equation of the matching index type containing the prediction reference index has larger offset, and the adjustment coefficient is reversely corrected, namely the influence weight of the second characteristic type on the final prediction reference index is increased. The weight of the second feature type influencing the prediction reference index is small, the frequency value of the historical reference index in the second feature type is changed from small to large due to the insertion of the similar prediction reference index, and the type with the lower weight influencing the prediction reference index is changed into the type with the higher weight influencing the prediction reference index. Therefore, when the prediction reference index is in the second feature class, the linear offset generated by the inserted matching index class directly affects the accuracy of the subsequent final prediction reference index and deviates from the expectation. The obtaining formula of the adjusting parameter of the historical reference index is specifically expressed as follows:
in the formula (I), the compound is shown in the specification,
an adjustment coefficient representing the historical reference index j,
representing historical reference index
The frequency value in the original match indicator category,
representing an ideal threshold. The adjusting coefficient of each historical reference index represents the possibility of deviation of the corresponding linear equation after the corresponding historical reference index is used as a prediction reference index and inserted into the matched index category, and the weight of the influence of the corresponding historical reference index on the prediction reference index is weightedAnd (6) adjusting the rows. When the prediction reference index is of the first characteristic type, the linear regression offset of the matching index type after the prediction reference index is inserted is small in change, and adjustment is not needed; when the prediction reference index is of the second feature type, the offset of the linear regression of the matching index type after the insertion of the prediction reference index may be large, and the weight affecting the prediction reference index needs to be adjusted.
And step S4: obtaining the numerical value categories of all historical reference indexes in the matching index categories; taking the numerical value corresponding to each numerical value category as a prediction reference index, sequentially inserting the prediction reference index into the matching index categories, and obtaining the reference index data variation of the matching index categories before and after insertion; and for any prediction reference index, constructing a prediction linear regression equation according to the adjustment parameters of the prediction reference index and the data variation of the reference index.
As can be seen from step S3, each historical reference index in the matching index category has a corresponding adjustment coefficient. The adjustment coefficient represents the possibility of offset of the corresponding linear equation after the corresponding historical reference index is used as a prediction reference index and inserted into the matching index category, and further the weight of the corresponding historical reference index influencing the prediction reference index is adjusted. Because the values of the historical reference indexes in the matching index categories without the inserted prediction reference indexes are repeated, the adjustment coefficients corresponding to the historical reference indexes corresponding to the repeated values are the same, namely, the adjustment degrees of the weights of the historical reference indexes corresponding to the repeated values, which influence the prediction reference indexes, are also the same. Therefore, it is necessary to obtain the numerical categories of all the historical reference indicators in the matching indicator category, and then analyze the numerical values of the prediction reference indicators according to each numerical category.
Further, taking a numerical value corresponding to each numerical value category as a prediction reference index, sequentially inserting the prediction reference index into the matching index categories, and obtaining the reference index data variation of the matching index categories before and after insertion, the embodiment of the invention specifically comprises:
and taking the standard deviation of all historical reference indexes in the matching index category as the first discrete degree of the matching index category before insertion. The first discrete degree reflects the discrete degree among all the historical reference indexes in the matching index category before insertion, and the larger the first discrete degree is, the more dispersed the numerical distribution of the historical reference indexes in the corresponding matching index category before insertion is.
And taking the numerical value corresponding to each numerical value category as a prediction reference index, sequentially inserting the prediction reference index into the matching index category, forming a prediction category by all historical reference indexes and the prediction reference index in the matching index category, and taking the standard deviation of the prediction category as the prediction dispersion degree of the inserted matching index category. The prediction dispersion degree reflects the dispersion degree among all reference indexes in the matching index category after the prediction reference indexes of different numerical value categories are inserted into the matching index category, and the larger the prediction dispersion degree is, the more dispersed the distribution of the reference index values in the matching index category of the inserted prediction reference indexes is.
Obtaining a difference absolute value of the predicted dispersion degree of each post-insertion matching index category and the first dispersion degree of each pre-insertion matching index category, and taking the difference absolute value as the reference index data variation of the pre-insertion and post-insertion matching index categories; each prediction reference index corresponds to one reference index data variation. The reference index data variation reflects the difference in the degree of dispersion between the post-insertion matching index category and the pre-insertion matching index category, and the larger the reference index data variation, the larger the variation in the degree of dispersion in the matching index category after insertion of the corresponding prediction reference index.
The method comprises the steps of taking the numerical value of each numerical value category in the matching index category as a prediction reference index, sequentially inserting the prediction reference index into the matching index category, and knowing the frequency value of the prediction reference index, the first discrete degree of the matching index category before insertion and the prediction discrete degree of the matching index category after insertion to design a linear regression model. When the numerical value of the prediction reference index is any numerical value category, the smaller the variation of the reference index data after the prediction reference index is inserted into the matching index category is, the higher the fitness of the final prediction reference index which is the corresponding prediction reference index is, the maximum posterior probability of the historical reference index to the final prediction reference index can be ensured to the maximum extent, and simply, the maximum reliability of the prediction reference index obtained by the historical reference index can be ensured.
According to the embodiment of the invention, under the influence factors of similar multidimensional environment, the predicted reference index is used as a dynamic numerical value to influence the structure variation of the matching index category, so that the relevance between the predicted reference index and the historical reference index is improved, the prediction result is more accurate, and the linear regression equation offset generated by the matching index category after the predicted reference index is inserted into the matching index category needs to be corrected according to the adjustment parameter of each historical reference index in the matching index category. The prediction reference index of each numerical category is provided with a corresponding historical reference index in the matching index category, and as each historical reference index in the matching index category is provided with a corresponding adjusting parameter, and after the prediction reference index of each numerical category is inserted into the matching index category, each index category corresponds to a reference index data variation of the matching index category before and after insertion. Therefore, the prediction reference index of each numerical category has a corresponding adjusting parameter and a reference index data variation. Further, for any prediction reference index, a prediction linear regression equation is constructed according to the adjustment parameters of the prediction reference index and the reference index data variation, and the embodiment of the invention specifically comprises the following steps:
acquiring the variable quantity of reference index data corresponding to the prediction reference index of the matching index category after each insertion, and subtracting the frequency value of the prediction reference index in the matching index category after the insertion from the constant 1 to be used as an independent variable of the prediction linear regression equation; and taking the adjusting parameters as error terms, and taking the reference index data variable quantity as a linear regression parameter. It should be noted that the method for constructing the linear regression equation is a technical means well known to those skilled in the art, and details are not repeated herein, and only a predictive linear regression equation provided by an embodiment of the present invention is briefly described herein as follows:
in the formula (I), the compound is shown in the specification,
the dependent variable representing the predictive linear regression equation,
indicating the number of all value classes in the matching index class,
denotes the first
A prediction reference index of the class value class,
denotes the first
The adjustment parameters corresponding to the prediction reference indexes of the class value category,
indicates the insertion of
The degree of prediction dispersion of the class value class of prediction reference indicators matching the indicator class,
a first degree of dispersion is indicated and,
is shown as
The variation of the reference index data corresponding to the prediction reference index of the class value category,
denotes the first
Matching the frequency values of the prediction reference indexes of the class value category in the index category after the insertion;
representing an absolute value function.
In the predictive linear regression equation, the variation of the reference index data reflects the difference of the discrete degrees between the matching index types after insertion and the matching index types before insertion, and the larger the variation of the reference index data is, the lower the degree of engagement between the final predicted reference index and the corresponding predicted reference index is. The adjusting parameter is used for correcting the linear regression equation offset generated by the matching index class after the prediction reference index is inserted.
And the logic relation correction is performed on the frequency value of the prediction reference index in the matching index category after the insertion. When the inserted prediction reference index is a numerical value category in the matching index category, the prediction dispersion degree of the inserted matching index category is used as the prediction dispersion degree of the matching index category
Can be related to
And dependent variable
The linear regression equation of (a) is,
to insert into
The linear relation variable quantity of the matching index class after the prediction reference index of the class value class is smaller when the variable quantity of the reference index data is smaller
The larger the dependent variable
The smaller the change, i.e. the first
Prediction reference index and final prediction reference of class value categoryThe higher the fitness of the index.
Step S5: carrying out nonlinear transformation on the prediction linear regression equation to obtain the prediction probability value of each prediction reference index; weighting each prediction reference index according to the prediction probability value to obtain a final prediction reference index; and monitoring and managing the aquatic seedling raising environment in the future date according to the final prediction reference index.
And (5) obtaining a predictive linear regression equation through the analysis of the step (S4), wherein the logistic regression model is essentially that the predictive linear regression equation is subjected to nonlinear conversion through a Sigmoid function. Therefore, the nonlinear conversion is performed on the predictive linear regression equation to obtain the predictive probability value of each predictive reference index, which specifically includes:
and carrying out nonlinear transformation on the predictive linear regression equation by using a Sigmoid function to obtain a corresponding first predictive equation. Taking the numerical value corresponding to each numerical value category as a prediction reference index, inputting each prediction reference index into a first prediction equation, and outputting the sub-prediction probability of the prediction reference index; and obtaining the prediction probability value of the prediction reference index according to the absolute value of the difference value between the constant one and the sub prediction probability. It should be noted that, the method of using Sigmoid function to perform nonlinear transformation on linear regression equation is a technical means well known to those skilled in the art, and will not be described herein again. The formula for obtaining the prediction probability value specifically includes:
in the formula (I), the compound is shown in the specification,
denotes the first
The prediction probability value of the prediction reference index of the class value category,
a dependent variable representing a predictive linear regression equation, i.e. a predictive probability equation,
which is a representation of a natural constant of,
denotes the first
Sub-prediction probabilities of prediction reference indices of class value classes,
representing an absolute value function.
When the predicted probability value is smaller, i.e. when
Inserting the prediction reference index of the class value class into the matching index class
The smaller the amount of change of the linear relationship of the prediction reference index of the class value class in the matching index class is, that is
The larger the size of the tube is,
the smaller, the following will follow
Logic that the prediction reference indicators of class-value classes should have less influence on the structure of the inserted matching indicator classes, i.e.
The smaller. Thus making use of
Correct the logical relationship when
The smaller the size of the product is,
the larger.
If the prediction probability value of the prediction reference index is larger, the possibility that the corresponding prediction reference index is the final prediction reference index is higher; if the prediction probability value of the prediction reference index is smaller, the probability that the corresponding prediction reference index is the final prediction reference index is lower.
In order to obtain a more accurate numerical value of the prediction reference index and realize accurate monitoring of the aquatic seedling raising condition at a future date, each prediction reference index is weighted according to the prediction probability value to obtain a final prediction reference index, and the embodiment of the invention specifically comprises the following steps:
accumulating the prediction probability values of all the prediction reference indexes to obtain a comprehensive prediction probability value; obtaining the ratio of the prediction probability value of each prediction reference index to the comprehensive prediction probability value, and taking the ratio as the sub-weight of the prediction reference index; taking the product of the sub-weight and the prediction reference index value as a sub-prediction index of the prediction reference index; and accumulating the sub prediction indexes of all the prediction reference indexes to obtain the final prediction reference index. The formula for obtaining the final prediction reference index specifically includes:
in the formula (I), the compound is shown in the specification,
which represents the final prediction reference index (ref),
indicating the number of all numerical categories in the matching index category,
is shown as
The prediction reference index of the class value class,
denotes the first
The value corresponding to the prediction reference index of the class value category,
is shown as
The prediction probability values of the prediction reference indicators of the class value classes,
representing the integrated prediction probability value.
In the formula of the final prediction reference index,
denotes the first
Sub-weights of prediction reference indices for class-valued classes, use
Will be provided with
And carrying out normalization, and multiplying the numerical values of the corresponding prediction reference indexes to obtain the sub prediction indexes of the corresponding prediction reference indexes.
And for weighted calculation, obtaining a final prediction reference index, wherein the final prediction reference index represents a reference index of a future date obtained by prediction, and the monitoring and management of the aquatic seedling raising environment of the future date can be realized through the final prediction reference index.
In the embodiment of the invention, the dissolved oxygen concentration is selected as a historical reference index for water body analysis, and the dissolved oxygen of the known water body is kept between 5mg/L and 8mg/L, at least more than 4mg/L and not less than 3mg/L; the later period of the high-density intensive culture is not less than 4mg/L. Therefore, in the embodiment of the invention, the preset reference index threshold is 4mg/L.
Further, monitoring and managing the aquatic breeding environment at a future date according to the final prediction reference index, wherein the specific monitoring and managing process in the embodiment of the invention specifically comprises the following steps: if the final predicted reference index is smaller than the preset reference index threshold value, the reference index of the water body is improved by improving the water pumping power of a manual pump, prolonging the water pumping time or increasing oxygen through micropores at the bottom of the fry raising pool, so that the problems that the concentration of dissolved oxygen in the fry raising pool is low and the fry generates blister diseases due to environmental indexes such as air temperature, cloud cover, air speed and the like in the future date are prevented.
In summary, in the present invention, all historical reference indicators are classified according to the environmental influence factor set, and the reference indicator category to which the environmental influence factor set of the future date belongs is taken as the matching indicator category; according to the method, similar multidimensional environment influence factors are introduced, the relevance of the prediction reference index and the historical reference index is improved, and the prediction result is more accurate. And classifying all the historical reference indexes according to the frequency values to obtain two characteristic classes, wherein the frequency values reflect the influence of the historical reference indexes on the prediction reference indexes, so that the influence weights of the two characteristic classes are different through classification, the prediction reference indexes are analyzed by combining the weight of the influence of the historical reference indexes on the prediction reference indexes, and the obtained prediction effect is more accurate. And acquiring an adjusting parameter according to the characteristic category and the frequency value of each historical reference index in the matching index category, wherein the adjusting parameter can predict the degree of fit between the reference index and the historical reference index of each historical date, and provides a basis for a subsequently acquired predictive linear regression equation. The reference index data variation reflects the difference of the discrete degree between the matching index category after insertion and the matching index category before insertion, and the structural change of the matching index category before and after insertion can be more intuitively represented. And constructing a predictive linear regression equation according to the adjustment parameters and the variable quantity of the reference index data, carrying out nonlinear conversion on the predictive linear regression equation to obtain a predictive probability value, carrying out weighting processing on each predictive reference index according to the predictive probability value to obtain a final predictive reference index, and realizing accurate monitoring on the aquatic seedling raising condition at a future date. By optimizing the problem of the nonlinear water quality parameter prediction accuracy, the method has higher practicability on nonlinear data and more accurate prediction result, and improves the accuracy of predicting the water quality change of the aquatic seedling raising environment.
The invention also provides an aquatic breeding environment monitoring and managing system which comprises a memory and a processor, wherein the processor executes a computer program stored in the memory so as to realize the aquatic breeding environment monitoring and managing method.
It should be noted that: the sequence of the above embodiments of the present invention is only for description, and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
All the embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from other embodiments.
Claims (9)
1. An aquatic seedling raising environment monitoring and management method is characterized by comprising the following steps:
obtaining a historical reference index and a corresponding environmental influence factor set of each historical date of aquaculture; the historical reference indexes of each historical date correspond to the environmental influence factor sets one by one; obtaining a set of environmental impact factors for a future date;
classifying all historical reference indexes according to the environmental influence factor set of each historical date and the future date to obtain at least two reference index categories, and taking the reference index category to which the environmental influence factor set of the future date belongs as a matching index category;
classifying all historical reference indexes according to the frequency value of each historical reference index in the matching index class to obtain two characteristic classes; acquiring an adjusting parameter of the corresponding historical reference index according to the characteristic category to which each historical reference index in the matching index category belongs and the frequency value;
obtaining numerical value categories of all historical reference indexes in the matching index categories; taking the numerical value corresponding to each numerical value category as a prediction reference index, and sequentially inserting the prediction reference index into the matching index categories to obtain the reference index data variation of the matching index categories before and after insertion; for any one of the prediction reference indexes, constructing a prediction linear regression equation according to the adjusting parameters of the prediction reference indexes and the reference index data variation;
carrying out nonlinear conversion on the prediction linear regression equation to obtain the prediction probability value of each prediction reference index; weighting each prediction reference index according to the prediction probability value to obtain a final prediction reference index; and monitoring and managing the aquatic seedling raising environment in the future date according to the final prediction reference index.
2. An aquatic breeding environment monitoring and management method according to claim 1, wherein the environmental influence factor set comprises:
the environmental indexes comprise air temperature, water vapor pressure, cloud cover, air pressure, wind speed and sunshine hours; and taking all the environmental indexes of each date as the environmental influence factor set of the corresponding date.
3. An aquatic breeding environment monitoring and management method according to claim 1, wherein the characteristic category obtaining method comprises:
the feature classes comprise a first feature class and a second feature class;
sorting the frequency values of all historical reference indexes in the matching index category according to the sequence of the frequency values from large to small to obtain a frequency sequence; accumulating the frequency sequences in sequence from the first frequency value of the frequency sequences until the accumulated value is larger than a preset ideal threshold value;
classifying all historical reference indexes corresponding to the frequency values participating in the accumulation operation into categories to be classified, eliminating the historical reference index corresponding to the last frequency value participating in the accumulation operation from the categories to be classified, and taking the updated categories to be classified as the first characteristic categories; and classifying all historical reference indexes belonging to the non-first characteristic category in the matching index category into the second characteristic category.
4. An aquatic breeding environment monitoring and managing method according to claim 3, wherein the adjusting parameter obtaining method comprises:
for any historical reference index in the matching index category, if the historical reference index is in a first characteristic category, the adjusting parameter corresponding to the historical reference index is a preset stable parameter;
and if the historical reference index is in the second characteristic category, taking the ratio of the frequency value of the corresponding historical reference index to the ideal threshold value as a negative adjusting parameter, and subtracting the negative adjusting parameter from the constant one to obtain the adjusting parameter of the corresponding historical reference index.
5. An aquatic breeding environment monitoring and managing method according to claim 1, wherein the method for obtaining the reference index data variation comprises:
taking the standard deviation of all historical reference indexes in the matching index category as a first discrete degree of the matching index category before insertion;
taking the numerical value corresponding to each numerical value category as a prediction reference index, sequentially inserting the prediction reference index into the matching index category, forming a prediction category by all historical reference indexes in the matching index category and the prediction reference index, and taking the standard deviation of the prediction category as the prediction dispersion degree of the inserted matching index category;
obtaining a difference absolute value between the predicted dispersion degree of each matching index type after insertion and a first dispersion degree of the matching index type before insertion, and taking the difference absolute value as a reference index data variation of the matching index types before and after insertion; each of the prediction reference indices corresponds to one of the reference index data variation amounts.
6. The aquatic breeding environment monitoring and management method according to claim 5, wherein the construction method of the predictive linear regression equation comprises:
acquiring the variable quantity of reference index data corresponding to the prediction reference index of the matching index type after each insertion, and taking the frequency value of the matching index type after the insertion, which is subtracted from the constant 1, as the independent variable of the prediction linear regression equation; and taking the adjusting parameters as error terms, and taking the reference index data variation as linear regression parameters.
7. The aquatic breeding environment monitoring and managing method as claimed in claim 6, wherein the obtaining method of the predicted probability value comprises:
carrying out nonlinear conversion on the prediction linear regression equation by using a Sigmoid function to obtain a corresponding first prediction equation;
taking the numerical value corresponding to each numerical value category as a prediction reference index, inputting each prediction reference index into the first prediction equation, and outputting the sub-prediction probability of the prediction reference index; and obtaining the prediction probability value of the prediction reference index according to the absolute value of the difference value between the constant one and the sub prediction probability.
8. An aquatic breeding environment monitoring and managing method according to claim 7, wherein the method for obtaining the final prediction reference index comprises:
accumulating the prediction probability values of all the prediction reference indexes to obtain a comprehensive prediction probability value;
obtaining the ratio of the prediction probability value of each prediction reference index to the comprehensive prediction probability value, and taking the ratio as the sub-weight of the prediction reference index; taking the product of the sub-weight and the prediction reference index value as a sub-prediction index of the prediction reference index; and accumulating the sub prediction indexes of all the prediction reference indexes to obtain a final prediction reference index.
9. An aquatic breeding environment monitoring and management system, comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to realize an aquatic breeding environment monitoring and management method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187538.7A CN115879750B (en) | 2023-03-02 | 2023-03-02 | Aquatic seedling environment monitoring management system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187538.7A CN115879750B (en) | 2023-03-02 | 2023-03-02 | Aquatic seedling environment monitoring management system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115879750A true CN115879750A (en) | 2023-03-31 |
| CN115879750B CN115879750B (en) | 2023-05-05 |
Family
ID=85761742
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310187538.7A Active CN115879750B (en) | 2023-03-02 | 2023-03-02 | Aquatic seedling environment monitoring management system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115879750B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116204566A (en) * | 2023-04-28 | 2023-06-02 | 深圳市欣冠精密技术有限公司 | Digital factory monitoring big data processing system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104392391A (en) * | 2014-11-14 | 2015-03-04 | 国家电网公司 | Power grid running safety risk quantification method |
| US20160007426A1 (en) * | 2011-04-14 | 2016-01-07 | Suntracker Technologies Ltd. | Predictive daylight harvesting system |
| EP3239735A1 (en) * | 2016-04-27 | 2017-11-01 | Sigma Space Corporation | Methods and apparatus for adaptive multisensor analysis and aggregation |
| CN113344972A (en) * | 2021-05-25 | 2021-09-03 | 浙江工业大学 | Intensive culture-based fish track prediction method |
| CN114997775A (en) * | 2022-05-19 | 2022-09-02 | 深圳市展之信供应链管理有限公司 | Intelligent regulation and control method and system for improving transportation survival rate of aquatic animals |
-
2023
- 2023-03-02 CN CN202310187538.7A patent/CN115879750B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160007426A1 (en) * | 2011-04-14 | 2016-01-07 | Suntracker Technologies Ltd. | Predictive daylight harvesting system |
| CN104392391A (en) * | 2014-11-14 | 2015-03-04 | 国家电网公司 | Power grid running safety risk quantification method |
| EP3239735A1 (en) * | 2016-04-27 | 2017-11-01 | Sigma Space Corporation | Methods and apparatus for adaptive multisensor analysis and aggregation |
| CN113344972A (en) * | 2021-05-25 | 2021-09-03 | 浙江工业大学 | Intensive culture-based fish track prediction method |
| CN114997775A (en) * | 2022-05-19 | 2022-09-02 | 深圳市展之信供应链管理有限公司 | Intelligent regulation and control method and system for improving transportation survival rate of aquatic animals |
Non-Patent Citations (1)
| Title |
|---|
| 于盟盟: "水产育苗养殖海水过滤系统" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116204566A (en) * | 2023-04-28 | 2023-06-02 | 深圳市欣冠精密技术有限公司 | Digital factory monitoring big data processing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115879750B (en) | 2023-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Comparison of six statistical approaches in the selection of appropriate fish growth models | |
| CN112906298B (en) | Blueberry yield prediction method based on machine learning | |
| CN110782658A (en) | Traffic prediction method based on LightGBM algorithm | |
| CN115879750B (en) | Aquatic seedling environment monitoring management system and method | |
| CN113537639A (en) | Method and system for establishing fish growth prediction model as culture environment parameter | |
| Jasmin et al. | An intelligent framework for prediction and forecasting of dissolved oxygen level and biofloc amount in a shrimp culture system using machine learning techniques | |
| WO2020124232A1 (en) | Systems and methods for predicting growth of a population of organisms | |
| CN113159456A (en) | Water quality prediction method, device, electronic device, and storage medium | |
| CN115907195A (en) | Photovoltaic power generation power prediction method, system, electronic device and medium | |
| CN111915097A (en) | Water quality prediction method for optimizing LSTM neural network based on improved genetic algorithm | |
| CN116882562A (en) | Agricultural carbon emission influence factor analysis method and system | |
| Tian et al. | Stock assessment for Atlantic yellowfin tuna based on extended surplus production model considering life history | |
| CN116804668B (en) | Salt iodine content detection data identification method and system | |
| CN109472320B (en) | Method for automatically correcting variety parameters of crop growth period model under uncertain conditions | |
| CN116228455B (en) | Method for processing test data of flow culture of Notopterygium longituba in leaf | |
| CN116630080A (en) | Method and system for determining capacity of aquatic product intensive culture feed based on image recognition | |
| Tahvonen et al. | Optimizing high-dimensional stochastic forestry via reinforcement learning | |
| McDonald et al. | A TUTORIAL ON EVALUATING EXPECTED RETURNS FROM RESEARCH FOR FISHERY MANAGEMENT USING BAYES'THEOREM | |
| CN115526117A (en) | Step type landslide displacement rate prediction method based on LSO-RF model | |
| McGhee et al. | Density‐dependent harvest modeling for the eastern wild turkey | |
| Mulema et al. | Quality and productivity in aquaculture: Prediction of oreochromis mossambicus growth using a transfer function arima model | |
| CN113539386A (en) | CLMVO-ELM-based dissolved oxygen concentration prediction method, device, equipment and storage medium | |
| CN117196881B (en) | Intelligent cultivation information management system based on big data | |
| Haiyang et al. | An improved Canopy-FFCM clustering algorithm for ocean data analysis | |
| Marsh et al. | The 2016 stock assessment of paua (Haliotis iris) for PAU 5D |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |