Method for improving disease prevention and control capability of aquaculture through artificial vibration wave data reconstruction

Abstract

The invention discloses a method for improving the disease prevention and control capability of aquaculture through artificial vibration wave data reconstruction. The method comprises the following steps: collecting artificial vibration wave data; carrying out special processing on the acquired artificial vibration wave data; solving different fish disease parameters; establishing a prediction mathematical model for identifying different fish diseases; the method for automatically identifying different fish diseases is realized. According to the method for improving the aquaculture disease prevention and control capability through artificial vibration wave data reconstruction, the application of an aquaculture disease prevention and control technology is improved through big data artificial intelligence, healthy development of aquaculture is ensured, and great significance is achieved in reducing aquaculture diseases; the large-scale and high-speed development of aquaculture can also drive the development of the industry, the related industries can bring huge benefits, more working posts can also be provided, and huge positive influences are brought to the improvement of the employment rate and the improvement of regional economic development.

Description

Technical Field

[0001] This invention relates to the field of aquaculture technology, and in particular to a method for improving the disease prevention and control capabilities of aquaculture through artificial vibration wave data reconstruction. Background Technology

[0002] my country's aquaculture industry faces numerous challenges. For most species, only 20% are profitable, while many others break even or incur losses. Sometimes, a single disease outbreak can wipe out the entire crop, making life extremely difficult for farmers. Therefore, to prevent and control aquaculture diseases and ensure the industry's effective and healthy development, it is necessary to further develop advanced aquaculture concepts and technologies, and improve my country's disease prevention and control capabilities in aquaculture.

[0003] Traditional methods of disease control in aquaculture are no longer adequate for the demands of modern aquaculture. Fish diseases are diverse and unpredictable, including viral, bacterial, and parasitic diseases. Therefore, simply observing fish under a microscope to check for parasites, injuries, loss of appetite, and abnormal feces, and then administering medication based on these observations, is no longer sufficient to meet the needs of modern aquaculture.

[0004] Therefore, it is necessary to provide a new method for reconstructing artificial vibration wave data to improve the disease prevention and control capabilities of aquaculture and solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for improving disease prevention and control capabilities in aquaculture through artificial vibration wave data reconstruction.

[0006] The method for improving disease prevention and control in aquaculture through artificial vibration wave data reconstruction provided by this invention includes the following steps: S1: Collect artificial vibration wave data; S2: Special processing is performed on the collected artificial vibration wave data; S3: Calculate parameters for different fish diseases; S4: Establish predictive mathematical models to identify different fish diseases; S5: A method for automatically identifying different fish diseases.

[0007] Preferably, the method for acquiring artificial vibration wave data in step S1 includes: By artificially generating vibration waves, we can comprehensively collect reflected waves from different fish species underwater, and collect and record them in real time. The convolutional neural network algorithm in deep learning is used to identify and analyze the amplitude, frequency, phase, and waveform structural features of vibration wave data.

[0008] Preferably, the special processing of the artificial vibration wave data in step S2 includes: The collected vibration wave data is subjected to denoising, filtering, normalization, and feature enhancement to improve data usability and recognition accuracy.

[0009] Preferably, the method for obtaining different fish disease parameters in step S3 includes extracting waveform feature parameters related to the disease based on the processed vibration wave data, and establishing a disease feature database.

[0010] Preferably, the method for establishing a predictive mathematical model for different fish diseases in step S4 includes: By combining machine learning algorithms, including random forest, support vector machine, and recurrent neural network, a comprehensive analysis of the waveform structure features of fish is performed. Establish a GM calculation model based on grey system theory and optimize the prediction algorithm.

[0011] Preferably, the method for automatically identifying different fish diseases in step S5 includes: By utilizing established predictive models, we can achieve big data processing and automatic identification of different fish species and diseases. Train the model on a standard dataset; Accelerate the training and inference process of deep learning models using high-performance computing environments.

[0012] Compared with related technologies, the method for improving disease prevention and control in aquaculture through artificial vibration wave data reconstruction provided by this invention has the following beneficial effects: This invention provides a method for improving disease control in aquaculture through artificial vibration wave data reconstruction. By leveraging big data and artificial intelligence to enhance disease control technology, it significantly contributes to the healthy development of aquaculture and reduces aquatic diseases. The rapid and large-scale development of aquaculture also drives the growth of related industries, generating substantial economic benefits and creating more jobs, thus positively impacting employment and regional economic development. This method directly and quantitatively characterizes the spatial distribution of fish populations from research results on different fish diseases; it objectively and quantitatively evaluates different fish diseases, reducing subjective human factors and increasing the reliability of research results; it improves the accuracy of disease evaluation for different fish species, reduces diseases in aquaculture, and protects the economic benefits of specialized aquaculture farmers. Attached Figure Description

[0013] Figure 1 The flowchart of the aquaculture disease prevention and control method based on artificial vibration wave data reconstruction provided by the present invention is shown. Detailed Implementation

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] In the specific implementation process, such as Figure 1 As shown, a method for disease control in aquaculture based on artificial vibration wave data reconstruction is proposed. This method automatically identifies fish diseases by collecting, processing, and analyzing underwater vibration wave data and establishing a predictive model. The method includes the following steps: S1: Artificial vibration wave data acquisition; S2: Special processing of vibration wave data; S3: Parameters for different fish diseases; S4: Establishment of mathematical models for disease prediction; S5: Automatic recognition method implementation.

[0016] S1: Artificial vibration wave data acquisition: This step involves acquiring artificial vibration wave data, and the specific method is as follows: Vibration wave excitation and acquisition: Artificial vibration waves are generated at a specific underwater location (such as the center of an aquaculture pond) using a specialized vibration wave generator. The vibration wave frequency range is controlled within... The sampling frequency covers the frequency range in which fish may respond. A high-sensitivity hydrophone array (with at least 8 sampling points) is used to comprehensively collect reflected waves from different underwater fish, with a sampling frequency of at least [missing information]. It collects and records vibration wave data in real time.

[0017] Data identification and analysis: The Convolutional Neural Network (CNN) algorithm in deep learning is used to identify and analyze the amplitude, frequency, phase, and waveform structural features of vibration wave data.

[0018] The basic formula for convolution operation is expressed as: Convolution operation:

[0019] In the discrete case, for vibration wave signals and convolution kernel Convolution output for: Discrete convolution:

[0020] S2: Special processing of artificial vibration wave data Preprocessing the collected vibration wave data improves data quality and usability: Denoising and filtering: Wavelet thresholding denoising is used to eliminate environmental noise interference, and a Butterworth bandpass filter is used. Filter out irrelevant frequency components. The threshold for wavelet denoising is selected using the following formula: Wavelet threshold:

[0021] in, This is an estimate of the noise standard deviation. This is the signal length.

[0022] S3: Normalization and Feature Enhancement The vibration wave data is subjected to min-max normalization to map the data to... Interval: Data normalization:

[0023] Determining parameters for different fish diseases: Based on the processed vibration wave data, waveform feature parameters related to the disease are extracted: Feature parameter extraction: The extracted disease-related features include: Time-domain characteristics: mean amplitude, variance, peak factor, impulse factor of waveform. Frequency domain characteristics: spectral centroid, frequency standard deviation, spectral skewness, spectral kurtosis Time-frequency domain characteristics: wavelet packet energy distribution, short-time Fourier transform characteristics Disease-specific characteristics: Waveform characteristic parameters for different fish diseases (such as bacterial gill rot, saprolegniasis, and parasitic diseases): S4: Establish predictive mathematical models for identifying different fish diseases By combining multiple machine learning algorithms, a disease prediction model is established: Grey system GM model optimization: A GM(1,1) computational model based on grey system theory is established for disease trend prediction. GM(1,1) model:

[0024] in, Generate a sequence by accumulating the original data once. For development coefficient, This represents the gray effect quantity.

[0025] S5: Methods for automatically identifying different fish diseases: Using established predictive models, an automatic disease identification system can be implemented: Big data processing workflow: A distributed data processing architecture can be constructed to handle large-scale vibration wave data from multiple fish species and diseases. High-performance computing acceleration: Accelerating the training and inference process of deep learning models using GPU clusters: The following is a specific implementation example illustrating the practical application of this method: Example: Detection of bacterial gill rot in grass carp 1. Data Collection: The excitation frequency in the grass carp breeding pond was [missing information]. The vibration wave was collected, and the reflected wave data was acquired.

[0026] 2. Data processing: Perform wavelet denoising and normalization on the collected data.

[0027] 3. Feature extraction: The spectral centroid (significantly lower than that of healthy fish by 30%) and waveform peak factor (higher than that of healthy fish by 25%) were extracted as key features.

[0028] 4. Model recognition: Input feature vectors into the trained RF-SVM ensemble model and output disease recognition results.

[0029] 5. Results output: The system outputs a diagnosis of bacterial gill rot, mid-stage, with a confidence level of 92%.

[0030] Tests have shown that this method achieves an accuracy rate of 94.7% in identifying common fish diseases, which is 85% more efficient than traditional manual observation methods and can provide early warning of diseases (detecting disease signs 3-5 days in advance).

[0031] The present invention provides a method for disease control in aquaculture based on artificial vibration wave data reconstruction. This invention is the first to propose the introduction of artificial vibration wave data reconstruction theory into improving disease control in aquaculture. It presents the basic principles and algorithms of this method, thereby improving the reliability and operability of disease control in aquaculture.

[0032] In the application of aquaculture disease prevention and control technology, a mathematical prediction model based on artificial vibration wave data reconstruction has been realized. This demonstrates a new technology for quantitative aquaculture disease prevention and control, which allows for qualitative to quantitative calibration and comparison of graphs and tables, thus broadening the technical means for aquaculture disease prevention and control.

[0033] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for improving disease prevention and control capabilities in aquaculture through artificial vibration wave data reconstruction, characterized in that, Includes the following steps: S1: Collect artificial vibration wave data; S2: Special processing is performed on the collected artificial vibration wave data; S3: Calculate parameters for different fish diseases; S4: Establish predictive mathematical models to identify different fish diseases; S5: A method for automatically identifying different fish diseases.

2. The method according to claim 1, characterized in that, The method for acquiring artificial vibration wave data in step S1 includes: By artificially generating vibration waves, we can comprehensively collect reflected waves from different fish species underwater, and collect and record them in real time. The convolutional neural network algorithm in deep learning is used to identify and analyze the waveform structural features of vibration wave data, including amplitude, frequency, and phase.

3. The method according to claim 1, characterized in that, The special processing of the artificial vibration wave data in step S2 includes: The collected vibration wave data is processed by denoising, filtering, normalization, and feature enhancement to improve the usability and recognition accuracy of the data.

4. The method according to claim 1, characterized in that, The method for obtaining different fish disease parameters in step S3 includes extracting waveform feature parameters related to the disease based on the processed vibration wave data and establishing a disease feature database.

5. The method according to claim 1, characterized in that, The method for establishing a mathematical model for predicting different fish diseases as described in step S4 includes: By combining machine learning algorithms, including random forest, support vector machine, and recurrent neural network, a comprehensive analysis of the waveform structure features of fish is performed. Establish a GM calculation model based on grey system theory and optimize the prediction algorithm.

6. The method according to claim 1, characterized in that, The method for automatically identifying different fish diseases described in step S5 includes: By utilizing established predictive models, we can achieve big data processing and automatic identification of different fish species and diseases. Train the model on a standard dataset; Accelerate the training and inference process of deep learning models using high-performance computing environments.

Images