Intelligent monitoring system and method for aquaculture environment

Through artificial intelligence-based data analysis technology, the key parameters in the aquaculture environment are monitored in real time, and the fish health problems caused by deterioration of water quality are solved, real-time monitoring of the aquaculture environment and water quality improvement are achieved, and the immunity of fish are improved. Survival rate.

CN118939049AInactive Publication Date: 2024-11-12SHENZHEN ZIBEI DIGITAL TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411142126.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The harsh water quality in aquaculture environments leads to a decrease in fish immunity, increasing the risk of morbidity and survival, and it is difficult for the prior art to achieve real-time monitoring and improvement of water quality.

Method used

Using artificial intelligence-based data analysis technology, by analyzing dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll and water level data in the aquaculture water environment in real time, a multi-scale characteristic matrix is ​​constructed to judge the water quality status, and real-time monitoring and abnormality of the aquaculture environment are realized. Timely discovery of the situation.

Benefits of technology

Real-time monitoring of the aquaculture environment has been achieved, abnormal situations are discovered in a timely manner and measures have been taken to improve water quality, protect fish health and improve survival rates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939049A_ABST
    Figure CN118939049A_ABST
Patent Text Reader

Abstract

The invention provides an intelligent monitoring system and method for an aquaculture environment, relates to the field of intelligent aquatic product management, and adopts a data analysis technology based on artificial intelligence. Data of dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll and water level in the aquaculture water environment are analyzed in real time to judge whether the water quality state of the aquaculture water environment at the current time point is abnormal or not. Thus, real-time monitoring of the aquaculture environment can be achieved, abnormal conditions can be found in time, measures are taken to improve water quality, and therefore fish health is protected, and the survival rate is increased.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent aquaculture management, and more specifically, to an intelligent monitoring system and method for an aquaculture environment. Background Art

[0002] With the continuous economic development and significant improvement in the living standards of residents in recent years, aquaculture has gradually formed a new economic model, which has greatly promoted the rapid development of the aquaculture industry. However, in the process of artificial aquaculture, the most difficult problem is the bad water environment. Once an unsuitable water environment appears, the immunity of fish will decrease, which will not only easily cause pathological changes in the fish's body, resulting in increased morbidity, but will also seriously affect the survival of fish, making it difficult for them to survive well.

[0003] Therefore, an intelligent monitoring solution for aquaculture environment is needed. Summary of the invention

[0004] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides an intelligent monitoring system and method for an aquaculture environment, which uses artificial intelligence-based data analysis technology to perform real-time analysis of dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll, and water level data in the aquaculture water environment to determine whether there is an abnormality in the water quality of the aquaculture water environment at the current time point. In this way, real-time monitoring of the aquaculture environment can be achieved, which is conducive to timely detection of abnormal conditions and taking measures to improve water quality, thereby protecting fish health and improving survival rates.

[0005] According to one aspect of the present application, there is provided an intelligent monitoring system for an aquaculture environment, comprising:

[0006] Aquaculture environment monitoring related data collection module, used to obtain dissolved oxygen content, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level value in the aquaculture water environment at multiple predetermined time points within a predetermined time period;

[0007] A data structuring module related to aquaculture environment monitoring is used to input the dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​at the plurality of predetermined time points into a time-series matrix of aquaculture water environment monitoring parameters according to the time dimension;

[0008] A feature coding module for aquaculture environment monitoring related data, used for feature coding the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix;

[0009] A module for enhancing the characteristics of data related to aquaculture environment monitoring, used to enhance the characteristics of the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced characteristic matrix;

[0010] The aquaculture environment monitoring result generation module is used to obtain the aquaculture water environment water quality status monitoring result at the current time point based on the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

[0011] According to another aspect of the present application, there is provided an intelligent monitoring method for an aquaculture environment, comprising:

[0012] Obtaining dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​in the aquaculture water environment at multiple predetermined time points within a predetermined time period;

[0013] The dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level values ​​at the plurality of predetermined time points are input into a time series matrix of aquaculture water environment monitoring parameters according to the time dimension;

[0014] Performing feature encoding on the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix;

[0015] Performing feature enhancement on the multi-scale aquaculture water environment monitoring parameter time series feature matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix;

[0016] Based on the multi-scale aquaculture water environment monitoring parameter enhanced characteristic matrix, the aquaculture water environment water quality status monitoring result at the current time point is obtained.

[0017] Compared with the prior art, the intelligent monitoring system and method for aquaculture environment provided by the present application adopts data analysis technology based on artificial intelligence, and performs real-time analysis on dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll, and water level data in the aquaculture water environment to determine whether there is any abnormality in the water quality of the aquaculture water environment at the current time point. In this way, real-time monitoring of the aquaculture environment can be achieved, which is conducive to timely detection of abnormal conditions and taking measures to improve water quality, thereby protecting fish health and increasing survival rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 4 is a system block diagram of an intelligent monitoring system for an aquaculture environment according to an embodiment of the present application.

[0020] Figure 2 Schematic diagram of the architecture of an intelligent monitoring system for an aquaculture environment according to an embodiment of the present application.

[0021] Figure 3 It is a block diagram of a feature encoding module for data related to monitoring of an aquaculture environment in an intelligent monitoring system for an aquaculture environment according to an embodiment of the present application.

[0022] Figure 4 This is a block diagram of an aquaculture environment monitoring result generation module in an aquaculture environment intelligent monitoring system according to an embodiment of the present application.

[0023] Figure 5 Flow chart of the intelligent monitoring method for an aquaculture environment according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0025] In recent years, with the booming economy and the steady growth of residents' income levels, people's demand for food quality and diversity has increased, which has greatly promoted the transformation and upgrading of the aquaculture industry, making it evolve from a traditional production model to a new economic model that integrates technology, ecology and efficiency. Aquaculture has not only become an important pillar of many family economies, but also plays an irreplaceable role in promoting employment, improving dietary structure, and increasing farmers' income.

[0026] However, behind this prosperous scene, aquaculture is also facing unprecedented challenges, the most serious of which is the deterioration of water environment quality. With the acceleration of industrialization and the intensification of agricultural non-point source pollution, water bodies in many aquaculture areas have suffered varying degrees of pollution, such as excessive heavy metals, eutrophication, organic pollution, etc. These problems directly threaten the health and stability of the aquaculture water environment.

[0027] When fish are in such an unsuitable water environment, their physiological functions and immunity will suffer a serious blow. Harmful substances in the water will destroy the physiological balance of fish, affect their normal metabolism and physiological functions, and lead to a decline in fish physique and increased susceptibility. Once attacked by pathogens, fish will find it difficult to resist and are prone to various diseases, such as bacterial sepsis and parasitic infections. The outbreak of these diseases will not only cause a large number of fish deaths, but also increase breeding costs and reduce economic benefits. More seriously, the bad water environment will also have a profound impact on the reproduction and growth of fish. It is difficult for fish to reproduce normally in poor water quality, and the quality of their offspring will be greatly reduced, which poses a huge threat to the sustainable development of aquaculture. In addition, when fish live in a bad environment, their meat quality and taste will also be affected, which in turn affects consumers' willingness to buy and market acceptance. Therefore, an intelligent monitoring and identification solution for aquaculture environment is expected.

[0028] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also shown a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks provides new solutions and solutions for intelligent monitoring and identification of aquaculture environments.

[0029] Figure 1 4 is a system block diagram of an intelligent monitoring system for an aquaculture environment according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of an intelligent monitoring system for an aquaculture environment according to an embodiment of the present application. Figure 1 and Figure 2As shown, in the aquaculture environment intelligent monitoring system 100, it includes: an aquaculture environment monitoring related data collection module 110, which is used to obtain the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level value at multiple predetermined time points in the aquaculture water environment within a predetermined time period; an aquaculture environment monitoring related data structuring module 120, which is used to convert the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level value at the multiple predetermined time points into an aquaculture water environment monitoring parameter time series input matrix according to the time dimension; an aquaculture environment monitoring related data structuring module 120, which is used to convert the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level value at the multiple predetermined time points into an aquaculture water environment monitoring parameter time series input matrix according to the time dimension; The measurement-related data feature encoding module 130 is used to feature encode the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix; the aquaculture environment monitoring-related data feature enhancement module 140 is used to feature enhance the multi-scale aquaculture water environment monitoring parameter time series feature matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix; the aquaculture environment monitoring result generation module 150 is used to obtain the aquaculture water environment water quality status monitoring result at the current time point based on the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

[0030] Specifically, in the technical solution of the present application, the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level height value at multiple predetermined time points within a predetermined time period in the aquaculture water environment are first obtained. It should be understood that the dissolved oxygen in the aquaculture water is the basis for the survival of aquatic organisms, and fish and other aquatic organisms need dissolved oxygen to breathe. Low dissolved oxygen levels can cause fish hypoxia, reduce immunity, and even cause death. The pH value of the aquaculture water affects chemical reactions and biological activities in the water. Too low or too high pH values ​​can cause stress on fish and other aquatic organisms, affecting their growth and reproduction. The temperature of the aquaculture water directly affects the metabolic rate and growth rate of fish. Too high or too low temperature can cause fish stress and reduce immunity. The ammonia nitrogen content value of the aquaculture water is usually produced by fish excrement and decomposed organic matter. High ammonia nitrogen levels can cause fish poisoning and affect their health. The chlorophyll in the aquaculture water reflects the growth of algae in the water body. Excessive algae consume oxygen in the water, leading to hypoxia. The water level of aquaculture water affects the fluidity and oxygen exchange capacity of the water body. Too low a water level will lead to eutrophication and hypoxia. By obtaining water quality monitoring indicator data at multiple time points, a baseline model of water quality can be established to identify fluctuations within the normal range. When the parameter value at a certain time point deviates significantly from the historical data, abnormal conditions can be identified in a timely manner. In general, dissolved oxygen, pH value, temperature, ammonia nitrogen, chlorophyll and water level values ​​are interrelated and constitute the overall health of the water environment. Monitoring these parameters can detect potential water quality abnormalities in a timely manner.

[0031] In an embodiment of the present application, a possible implementation method for obtaining dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​in an aquaculture water environment at multiple predetermined time points within a predetermined time period may be: respectively installing a dissolved oxygen sensor, a pH sensor, a temperature sensor, an ammonia nitrogen sensor, a chlorophyll sensor and a water level sensor in the aquaculture water environment, so as to collect relevant aquaculture water environment monitoring indicator data in real time.

[0032] Taking into account that these aquaculture water quality monitoring parameter indicators are affected by many factors and change dynamically, in order to clearly show the changing trends of these water quality monitoring parameters over time, so as to understand the dynamic changes of aquaculture water, in the technical solution of the present application, the dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​at the multiple predetermined time points are input into the aquaculture water environment monitoring parameter time series matrix according to the time dimension. By arranging according to the time dimension, multiple aquaculture water quality monitoring parameter indicators can be integrated into a matrix to form a unified data structure, which simplifies the data processing process. Specifically, each row of data in the arranged aquaculture water environment monitoring parameter time series input matrix is ​​the value of all monitoring parameter indicators at a time point, and each column is the data of a specific monitoring parameter indicator data at multiple time points.

[0033] Figure 3 FIG. 1 is a block diagram of a feature encoding module for aquaculture environment monitoring related data in an aquaculture environment intelligent monitoring system according to an embodiment of the present application. Figure 3 As shown, the aquaculture environment monitoring related data feature encoding module 130 includes: a first-scale aquaculture environment monitoring feature extraction unit 131, used to perform first-scale time series feature extraction on the aquaculture water environment monitoring parameter time series input matrix to obtain a first-scale aquaculture water environment monitoring parameter time series feature matrix; a second-scale aquaculture environment monitoring feature extraction unit 132, used to perform second-scale time series feature extraction on the aquaculture water environment monitoring parameter time series input matrix to obtain a second-scale aquaculture water environment monitoring parameter time series feature matrix; a multi-scale feature fusion unit 133, used to fuse the first-scale aquaculture water environment monitoring parameter time series feature matrix and the second-scale aquaculture water environment monitoring parameter time series feature matrix to obtain the multi-scale aquaculture water environment monitoring parameter time series feature matrix.

[0034] The aquaculture water environment monitoring parameter time series input matrix contains multidimensional water quality parameters at multiple time points, which contains the long-term trend and periodic change characteristics of the water quality monitoring parameters. In order to extract these potential and important features from the original time series input matrix, it is necessary to pass the aquaculture water environment monitoring parameter time series input matrix through the first aquaculture water environment monitoring parameter time series feature encoder to obtain the first scale aquaculture water environment monitoring parameter time series feature matrix. After being processed by the first aquaculture water environment monitoring parameter time series feature encoder, the original redundant time series data is converted into the time series characteristics of the change trend of water quality monitoring parameters. In this application, the first aquaculture water environment monitoring parameter time series feature encoder is a convolutional neural network model using a first scale two-dimensional convolution kernel. Ordinary technicians in this field should understand that the convolutional neural network model (CNN) can effectively extract local features in the input matrix through convolution operations. In aquaculture water environment monitoring, local changes in certain water quality parameters may have an important impact on the overall water quality state. By using a two-dimensional convolution kernel, these local features can be captured, thereby improving the sensitivity of the model to identify water quality anomalies. Specifically, the two-dimensional convolution kernel can not only extract features in the time series, but also capture the spatial relationship between different water quality parameters, which is conducive to a more comprehensive analysis of water quality changes and thus improves the accuracy of judging abnormal aquaculture water quality.

[0035] At the same time, the aquaculture water environment monitoring parameter time series input matrix is ​​passed through the second aquaculture water environment monitoring parameter time series feature encoder to obtain a second scale aquaculture water environment monitoring parameter time series feature matrix. In the present application, the second aquaculture water environment monitoring parameter time series feature encoder is a convolutional neural network model using a second scale two-dimensional convolution kernel, and the first scale is different from the second scale. It should be understood that in a convolutional neural network, the scale (or size) of the convolution kernel refers to the size of the area covered when the convolution kernel slides on the input data. In the present application, the two-dimensional convolution kernel of the first scale is 3×3 in size, while the convolution kernel of the second scale is 5×5 in size. Different scales can capture features at different levels. The convolution kernel of the first scale is usually used to extract fine-grained local features, while the convolution kernel of the second scale can capture features in a larger range and identify more macroscopic trends of change, which can help the model fully understand water quality changes.

[0036] Considering that feature matrices of different scales can provide multi-level information. The first-scale aquaculture water environment monitoring parameter time series feature matrix can capture small changes in water quality parameters and is suitable for identifying short-term abnormalities. The second-scale aquaculture water environment monitoring parameter time series feature matrix can capture the long-term change trend of water quality and is suitable for analyzing seasonal changes. In order to effectively combine the information in feature matrices of different scales, the first-scale aquaculture water environment monitoring parameter time series feature matrix and the second-scale aquaculture water environment monitoring parameter time series feature matrix are fused in the technical solution of the present application to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix. The multi-scale aquaculture water environment monitoring parameter time series feature matrix obtained by fusion contains richer water quality change information. Based on the analysis of richer water quality change information, the accuracy of water quality anomaly identification can be improved. In addition, the multi-scale feature fusion makes the model more robust in the face of complex water quality changes. Even if the features of a certain scale are subject to noise or interference, the features of another scale can still provide useful information to ensure the stability of the model.

[0037] In an embodiment of the present application, a possible implementation method of fusing the first-scale aquaculture water environment monitoring parameter time series characteristic matrix and the second-scale aquaculture water environment monitoring parameter time series characteristic matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series characteristic matrix may be: using the following fusion formula to fuse the first-scale aquaculture water environment monitoring parameter time series characteristic matrix and the second-scale aquaculture water environment monitoring parameter time series characteristic matrix to obtain the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix; wherein the fusion formula is:

[0038] M c =αM a ⊕βM b

[0039] Among them, M c is the time series characteristic matrix of the multi-scale aquaculture water environment monitoring parameters, M a is the time series characteristic matrix of the aquaculture water environment monitoring parameters at the first scale, M b is the second-scale aquaculture water environment monitoring parameter time series characteristic matrix, ⊕ represents the addition of the elements at corresponding positions of the first-scale aquaculture water environment monitoring parameter time series characteristic matrix and the second-scale aquaculture water environment monitoring parameter time series characteristic matrix, α and β are weighted parameters for controlling the balance between the first-scale aquaculture water environment monitoring parameter time series characteristic matrix and the second-scale aquaculture water environment monitoring parameter time series characteristic matrix in the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix.

[0040] Next, the multi-scale aquaculture water environment monitoring parameter time series feature matrix is ​​passed through the aquaculture water environment monitoring parameter feature enhancer to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix. In the present application, the aquaculture water environment monitoring parameter feature enhancer is a convolutional neural network model using a spatial attention mechanism. The core of this structure is that it can adaptively focus on important areas in the input data, thereby improving the expression ability of features. Specifically, the input feature matrix is ​​processed using the aquaculture water environment monitoring parameter feature enhancer, which can enhance the expression of important features of the input feature matrix for identifying abnormal water quality states, and suppress those features that are relatively unimportant for identifying abnormal water quality states. The enhanced feature matrix can provide breeders with more accurate water quality status analysis and help them monitor water quality changes.

[0041] In an embodiment of the present application, a possible implementation method of passing the multi-scale aquaculture water environment monitoring parameter time series feature matrix through an aquaculture water environment monitoring parameter feature enhancer to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix can be: using each layer of the aquaculture water environment monitoring parameter feature enhancer to perform the following on the input data in the forward transfer of the layer: convolution processing is performed on the input data based on the convolution kernel to obtain a convolution feature map; the convolution feature map is passed through a spatial attention module to obtain a spatial attention score matrix; the spatial attention score matrix and each feature matrix of the convolution feature map along the channel dimension are multiplied by the spatial attention feature map at each position point; the spatial attention feature map is pooled based on a local feature matrix to obtain a pooled feature map; the pooled feature map is nonlinearly activated to obtain an activated feature map; wherein the input of the first layer of the aquaculture water environment monitoring parameter feature enhancer is the multi-scale aquaculture water environment monitoring parameter time series feature matrix, and the output of the last layer of the aquaculture water environment monitoring parameter feature enhancer is the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

[0042] Figure 4 FIG. 1 is a block diagram of an aquaculture environment monitoring result generation module in an aquaculture environment intelligent monitoring system according to an embodiment of the present application. Figure 4 As shown, the aquaculture environment monitoring result generation module 150 includes: a feature optimization unit 151, which is used to perform feature value expression back-propagation compensation on the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix based on feature offset information to obtain an optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix; a feature matrix classification unit 152, which is used to pass the optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix through a water quality classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the water quality status of the aquaculture water environment at the current time point.

[0043] In particular, in the technical solution of the present application, a multi-scale aquaculture water environment monitoring parameter time series feature matrix is ​​extracted from the multi-scale aquaculture water environment monitoring parameters, and these parameters may include temperature, dissolved oxygen, pH value, etc. The multi-scale aquaculture water environment monitoring parameter time series feature matrix is ​​processed using an aquaculture water environment monitoring parameter feature enhancer to highlight important features and form an enhanced feature matrix. The feature enhancer may give too high weights to certain monitoring parameters when extracting features, resulting in a shift in the distribution of these parameters in the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix. The processing of multi-scale features may overemphasize the changes in certain time scales or parameters, while ignoring other equally important changes, resulting in an imbalance of features. The shift deviation of feature distribution may cause the classifier to rely too much on certain features during training, resulting in extreme deviations in actual classification, such as mistakenly classifying normal water quality as abnormal or missing real abnormal situations. If the classifier learns the wrong pattern, it may not be able to accurately distinguish between normal and abnormal states of water quality, resulting in a decrease in the accuracy of the classification results. In order to solve this problem, in the technical solution of the present application, the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix is ​​subjected to eigenvalue expression back-propagation compensation based on feature offset information to obtain an optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix.

[0044] Specifically, in an embodiment of the present application, the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​subjected to eigenvalue expression back propagation compensation based on feature offset information to obtain an optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix, including: calculating a shift difference representation matrix between the multi-scale aquaculture water environment monitoring parameter time series feature matrix and the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix; probabilizing each eigenvalue of the shift difference representation matrix based on a probability activation function to obtain a probabilistic shift difference representation matrix; inputting the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix into the water quality classifier to obtain an initial classification probability value; based on Based on the comparison between the eigenvalues ​​of each position of the probabilistic shifted difference representation matrix and the initial classification probability value and the weight matrix of the water quality classifier, a backpropagation compensation representation matrix is ​​determined; the natural exponential function values ​​of each eigenvalue of the backpropagation compensation representation matrix are calculated to obtain a backpropagation compensation class representation matrix; the position-by-position point multiplication between the backpropagation compensation class representation matrix and the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix is ​​calculated to obtain a point multiplication result matrix, and the weighted sum between the point multiplication result matrix and the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix is ​​calculated to obtain the optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix.

[0045] In an embodiment of the present application, another expression method of performing back propagation compensation based on feature value expression of feature offset information on the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix to obtain an optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix can be: the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​processed by the following formula to obtain the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix; wherein the formula is:

[0046]

[0047] Among them, x 1(i,j) represents the eigenvalue of the (i, j)th position of the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix, X 2 represents the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix, x 2(i,j) represents the eigenvalue of the (i, j)th position of the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix, x 3(i,j) represents the eigenvalue of the (i, j)th position of the probabilistic shift difference representation matrix, X 4 represents the back propagation compensation representation matrix, δ and γ represent the first weighted hyperparameter and the second weighted hyperparameter respectively, Q represents the weight matrix of the water quality classifier, p represents the initial classification probability value, ⊙ represents the point multiplication by position, represents matrix multiplication, sgn represents the sign function, X 2 ' represents the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

[0048] That is, in view of the distribution shift deviation problem between feature matrices, the feature deviation information between the multi-scale aquaculture water environment monitoring parameter time series feature matrix and the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​used to perform feature value expression back propagation compensation based on feature offset information. The difference between the multi-scale aquaculture water environment monitoring parameter time series feature matrix and the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​calculated to obtain a shift difference representation matrix, so as to quantify the deviation of feature distribution. The shift difference representation matrix reveals the alignment of the two feature matrices in the feature space, providing a basis for subsequent compensation. Then, the probability activation function is applied to each eigenvalue of the shift difference representation matrix for probabilistic processing, so that the differential value is converted into a probability distribution, so that the model can understand the deviation between features in a probabilistic form. Then, the original multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​input into the water quality classifier to obtain the initial classification probability value, wherein the initial classification probability value provides the model with a basic classification performance indicator before feature compensation. Next, the back propagation compensation representation matrix is ​​determined by combining the probabilistic shift difference representation matrix and the initial classification probability value, as well as the weight matrix of the water quality classifier. That is, the feature matrix is ​​compensated by using the model's own weight and bias information to reduce the classification error. Next, the natural exponential function value is calculated for each eigenvalue of the back propagation compensation representation matrix, thereby enhancing the expression range of the eigenvalue and providing a mechanism for enlarging or reducing the feature compensation. Finally, the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​obtained by calculating the point product result between the back propagation compensation class representation matrix and the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix, and then calculating the weighted sum with the original multi-scale aquaculture water environment monitoring parameter enhanced feature matrix. It should be understood that the point product and weighted sum operations combine the compensation information and the original feature information to generate an adjusted feature matrix, aiming to reduce the extreme deviation of the category and improve the accuracy of classification. The entire data processing process achieves the goal of optimizing the feature representation from identifying the shift deviation of the feature distribution to compensating for these deviations. This process not only improves the adaptability of the water quality classifier to changes in feature distribution, but also enhances the model's discrimination between different categories. By expressing the back-propagation compensation mechanism based on the eigenvalues ​​of feature offset information, the model can capture the feature matrix more accurately.

[0049] Finally, the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix is ​​passed through a water quality classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the water quality state of the aquaculture water environment at the current time point. It should be understood that the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix integrates information from different time scales and monitoring parameter indicator characteristics, and can more comprehensively reflect water quality changes. Through the water quality classifier, these enhanced features can be converted into specific water quality state classifications, so as to accurately identify whether the current water quality is abnormal. The water quality classifier is a machine learning that aims to classify the water quality state according to the input feature data. The classification result directly indicates the water quality state of the aquaculture water environment at the current time point, helping the breeder to quickly understand whether the water quality is abnormal. The indication in this application is a binary classification (normal / abnormal). By obtaining the classification results in a timely manner, the breeder can take timely measures to better manage potential risks and reduce losses caused by abnormal water quality.

[0050] In summary, the intelligent monitoring system 100 for aquaculture environment based on the embodiment of the present application is explained, which uses artificial intelligence-based data analysis technology to perform real-time analysis of dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll, and water level data in the aquaculture water environment to determine whether there is an abnormality in the water quality of the aquaculture water environment at the current time point. In this way, real-time monitoring of the aquaculture environment can be achieved, which is conducive to timely detection of abnormal conditions and taking measures to improve water quality, thereby protecting fish health and improving survival rates.

[0051] As described above, the intelligent monitoring system 100 for an aquaculture environment according to an embodiment of the present application can be implemented in various terminal devices, such as a server for intelligent monitoring of an aquaculture environment. In one example, the intelligent monitoring system 100 for an aquaculture environment according to an embodiment of the present application can be integrated into a terminal device as a software module and / or a hardware module. For example, the intelligent monitoring system 100 for an aquaculture environment can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the intelligent monitoring system 100 for an aquaculture environment can also be one of the many hardware modules of the terminal device.

[0052] Alternatively, in another example, the intelligent monitoring system 100 for aquaculture environment and the terminal device may be separate devices, and the intelligent monitoring system 100 for aquaculture environment may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0053] Figure 5 FIG. 1 is a flow chart of an intelligent monitoring method for an aquaculture environment according to an embodiment of the present application. Figure 5As shown, in the intelligent monitoring method for aquaculture environment, the method includes: S110, obtaining the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level height value in the aquaculture water environment at multiple predetermined time points within a predetermined time period; S120, converting the dissolved oxygen content value, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level height value of the multiple predetermined time points into an aquaculture water environment monitoring parameter time series input matrix according to the time dimension; S130, feature encoding the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix; S140, feature enhancement of the multi-scale aquaculture water environment monitoring parameter time series feature matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix; S150, based on the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix, obtaining the aquaculture water environment water quality status monitoring result at the current time point.

[0054] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned aquaculture environment intelligent monitoring method have been described in detail above. Figures 1 to 4 The invention has been introduced in detail in the description of the intelligent monitoring system for aquaculture environment, and therefore, its repeated description will be omitted.

[0055] In summary, the intelligent monitoring method of the aquaculture environment based on the embodiment of the present application is explained, which adopts data analysis technology based on artificial intelligence, and performs real-time analysis of dissolved oxygen, pH, temperature, ammonia nitrogen, chlorophyll, and water level data in the aquaculture water environment to determine whether there is an abnormality in the water quality state of the aquaculture water environment at the current time point. In this way, real-time monitoring of the aquaculture environment can be achieved, which is conducive to timely detection of abnormal conditions and taking measures to improve water quality, thereby protecting fish health and improving survival rates.

[0056] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application.

[0057] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present application is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present application. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0058] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The second and other words are used to indicate names, but not to indicate any particular order.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical solution of the present application may be modified or replaced by equivalents without departing from the spirit of the technical solution of the present application.

Claims

1. An intelligent monitoring system for aquaculture environment, It is characterized in that include: Aquaculture environment monitoring related data collection module, used to obtain dissolved oxygen content, pH value, temperature value, ammonia nitrogen content value, chlorophyll content value and water level value in the aquaculture water environment at multiple predetermined time points within a predetermined time period; A data structuring module related to aquaculture environment monitoring is used to input the dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​at the plurality of predetermined time points into a time-series matrix of aquaculture water environment monitoring parameters according to the time dimension; A feature coding module for aquaculture environment monitoring related data, used for feature coding the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix; A module for enhancing the characteristics of data related to aquaculture environment monitoring, used to enhance the characteristics of the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced characteristic matrix; The aquaculture environment monitoring result generation module is used to obtain the aquaculture water environment water quality status monitoring result at the current time point based on the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

2. The intelligent monitoring system for aquaculture environment according to claim 1, It is characterized in that The aquaculture environment monitoring related data feature coding module includes: A first-scale aquaculture environment monitoring feature extraction unit is used to extract first-scale time series features from the aquaculture water environment monitoring parameter time series input matrix to obtain a first-scale aquaculture water environment monitoring parameter time series feature matrix; A second-scale aquaculture environment monitoring feature extraction unit is used to perform second-scale time series feature extraction on the aquaculture water environment monitoring parameter time series input matrix to obtain a second-scale aquaculture water environment monitoring parameter time series feature matrix; The multi-scale feature fusion unit is used to fuse the first-scale aquaculture water environment monitoring parameter time series feature matrix and the second-scale aquaculture water environment monitoring parameter time series feature matrix to obtain the multi-scale aquaculture water environment monitoring parameter time series feature matrix.

3. The intelligent monitoring system for aquaculture environment according to claim 2, It is characterized in that The first-scale aquaculture environment monitoring feature extraction unit is used to: pass the aquaculture water environment monitoring parameter time series input matrix through the first aquaculture water environment monitoring parameter time series feature encoder to obtain the first-scale aquaculture water environment monitoring parameter time series feature matrix.

4. The intelligent monitoring system for aquaculture environment according to claim 3, It is characterized in that The second-scale aquaculture environment monitoring feature extraction unit is used to: pass the aquaculture water environment monitoring parameter time series input matrix through a second aquaculture water environment monitoring parameter time series feature encoder to obtain the second-scale aquaculture water environment monitoring parameter time series feature matrix.

5. The intelligent monitoring system for aquaculture environment according to claim 4, It is characterized in that The aquaculture environment monitoring related data feature enhancement module is used to: pass the multi-scale aquaculture water environment monitoring parameter time series feature matrix through the aquaculture water environment monitoring parameter feature enhancer to obtain the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix.

6. The intelligent monitoring system for aquaculture environment according to claim 5, It is characterized in that The first aquaculture water environment monitoring parameter time series feature encoder is a convolutional neural network model using a first scale two-dimensional convolution kernel, and the second aquaculture water environment monitoring parameter time series feature encoder is a convolutional neural network model using a second scale two-dimensional convolution kernel, wherein the first scale is different from the second scale.

7. The intelligent monitoring system for aquaculture environment according to claim 6, It is characterized in that The aquaculture water environment monitoring parameter feature enhancer is a convolutional neural network model using a spatial attention mechanism.

8. The intelligent monitoring system for aquaculture environment according to claim 7, It is characterized in that The aquaculture environment monitoring result generation module comprises: A feature optimization unit, used for performing back propagation compensation of the feature value expression based on feature offset information on the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix to obtain an optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix; The feature matrix classification unit is used to pass the optimized multi-scale aquaculture water environment monitoring parameter enhanced feature matrix through a water quality classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the water quality state of the aquaculture water environment at the current time point.

9. The intelligent monitoring system for aquaculture environment according to claim 8, It is characterized in that The feature optimization unit is used to: Calculate the shift difference representation matrix between the multi-scale aquaculture water environment monitoring parameter time series characteristic matrix and the multi-scale aquaculture water environment monitoring parameter enhanced characteristic matrix; Probabilize each eigenvalue of the shifted difference representation matrix based on a probability activation function to obtain a probabilized shifted difference representation matrix; Inputting the multi-scale aquaculture water environment monitoring parameter enhanced feature matrix into the water quality classifier to obtain an initial classification probability value; Determine a back-propagation compensation representation matrix based on a comparison between the eigenvalues ​​of each position of the probabilistic shift difference representation matrix and the initial classification probability value and a weight matrix of the water quality classifier; Calculating the natural exponential function value of each eigenvalue of the back-propagation compensation representation matrix to obtain a back-propagation compensation class representation matrix; Calculate the positional point multiplication between the back-propagation compensation class representation matrix and the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix to obtain a point product result matrix, and calculate the weighted sum between the point product result matrix and the multi-scale aquaculture water environment monitoring parameter enhancement feature matrix to obtain the optimized multi-scale aquaculture water environment monitoring parameter enhancement feature matrix.

10. An intelligent monitoring method for aquaculture environment, It is characterized in that include: Obtaining dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level height values ​​in the aquaculture water environment at multiple predetermined time points within a predetermined time period; The dissolved oxygen content values, pH values, temperature values, ammonia nitrogen content values, chlorophyll content values ​​and water level values ​​at the plurality of predetermined time points are input into a time series matrix of aquaculture water environment monitoring parameters according to the time dimension; Performing feature encoding on the aquaculture water environment monitoring parameter time series input matrix to obtain a multi-scale aquaculture water environment monitoring parameter time series feature matrix; Performing feature enhancement on the multi-scale aquaculture water environment monitoring parameter time series feature matrix to obtain a multi-scale aquaculture water environment monitoring parameter enhanced feature matrix; Based on the multi-scale aquaculture water environment monitoring parameter enhanced characteristic matrix, the aquaculture water environment water quality status monitoring result at the current time point is obtained.

Citation Information

Cited By

  • Ecological aquaculture system

    CN119924242A

  • A breeding water environment monitoring and early warning system based on biological indication

    CN123087557A