Squid resource dynamic estimation method based on Bayesian depth inversion network

Abstract

The invention discloses a dynamic squid resource estimation method based on a Bayesian depth inversion network. The method comprises the following steps: S1, forming a preprocessed multi-source data set; s2, constructing a Bayesian depth inversion network model; s3, training the Bayesian depth inversion network model by using the preprocessed multi-source data set to form a trained model parameter set; s4, obtaining a dynamic estimation result of the squid resources and an uncertainty quantitative index of the dynamic estimation result; s5, post-processing the dynamic estimation result of the squid resources and the uncertainty quantitative indexes to form a dynamic estimation report of the squid resources with spatial and temporal distribution characteristics; and S6, carrying out interface docking on the dynamic estimation report of the squid resources and a fishery management system, and carrying out dynamic monitoring and real-time decision making on the squid resources. According to the method, the error prediction rate of the low-confidence prediction area is remarkably reduced, the prediction stability of the high-confidence area is improved, and more reliable support data is provided for fishery administration ship scheduling and fishing path planning.

Description

Technical Field

[0001] The invention relates to the technical field of squid resources, and in particular to a squid resource dynamic estimation method based on a Bayesian deep inversion network. Background Art

[0002] As the pressure on global fishery resource monitoring and management continues to grow, intelligent estimation technology based on marine environmental data and target population activity characteristics has gradually become an important support for marine fishery management. As a high-value, widely distributed, and highly active economic marine organism, the dynamic assessment of squid resources is of great significance to ensuring sustainable fishing and avoiding over-exploitation of resources. At present, the estimation and prediction of squid resources mainly rely on two types of technical paths: one is based on acoustic detection and sampling statistics; the other is regression prediction methods based on traditional neural networks or machine learning models.

[0003] In the first type of method, acoustic detection is usually combined with multi-beam echo sounders and underwater trawling to scan the sea area and estimate the squid distribution density. However, this method is limited by the equipment deployment range, operating sea conditions and seabed topography, making it difficult to achieve wide-area, high-frequency and real-time resource monitoring, and cannot effectively reflect the behavioral migration and distribution change trends of squid groups under the influence of complex environmental variables. More importantly, the acoustic data itself has high noise and is seriously affected by the temperature and salt structure, which can easily cause distortion of resource estimation results and lacks reliable uncertainty characterization methods.

[0004] Although the second type of estimation method based on deep learning can improve the estimation accuracy of squid resources to a certain extent, there are several prominent problems in the current technology: First, the model generally adopts a static training paradigm, which fails to effectively combine the temporal linkage characteristics between squid behavior and the marine environment, resulting in a delayed response to the dynamic evolution pattern; second, there is a lack of effective prior knowledge fusion mechanism, the model generalization ability is weak, and it is very easy to produce incorrect fitting of noise or edge scenes; third, most deep models are "black box" structures and lack the ability to quantify the uncertainty of the output results, which limits the credibility of the prediction results in actual fishery scheduling and risk management. In addition, existing methods usually simply splice marine environmental parameters with squid behavior parameters, ignoring the heterogeneous coupling characteristics between the two, and it is difficult to fully extract the interaction mechanism between potential behavior driving factors and key environmental variables.

[0005] Therefore, there is an urgent need for a more robust, interpretable and practical intelligent estimation method to solve the problem. Summary of the invention

[0006] One purpose of the present invention is to propose a method for dynamic estimation of squid resources based on a Bayesian deep inversion network. The present invention significantly reduces the error prediction rate in low-confidence prediction areas and improves the prediction stability in high-confidence areas, providing more reliable supporting data for fishery vessel scheduling and fishing route planning.

[0007] A method for estimating squid resource dynamics based on a Bayesian deep inversion network according to an embodiment of the present invention comprises the following steps: S1. Collecting a real-time multi-source data set including marine environment parameters and squid biological activity parameters, preprocessing the real-time multi-source data set to form a preprocessed multi-source data set; S2. Construct a Bayesian deep inversion network model; S3. Use the preprocessed multi-source data set to train the Bayesian deep inversion network model, use the dynamic prior update module to update the dynamic prior data in real time during the training process, and optimize the model parameters by maximizing the likelihood function and minimizing the prediction error to form a set of trained model parameters; S4. Input the multi-source data set collected in real time into the trained Bayesian deep inversion network model, and use the dynamic prior update module to update the prior data of the real-time multi-source data set to obtain the dynamic estimation results of squid resources and their uncertainty quantitative indicators; S5. Post-process the squid resource dynamic estimation results and uncertainty quantitative indicators to form a squid resource dynamic estimation report with temporal and spatial distribution characteristics; S6. Interface the squid resource dynamic estimation report with the fishery management system to conduct dynamic monitoring of squid resources and real-time decision-making.

[0008] Optionally, the S1 includes the following steps: S11. Collect real-time multi-source data including marine environmental parameters and squid biological activity parameters from ocean observation platforms, satellite remote sensing devices and underwater sensors to build a real-time multi-source data set : ; in, Indicates Original data records, Indicates the total number of real-time collected data. For the The collection timestamp of the data. For the The ocean environment parameter vectors recorded in ,salinity , chlorophyll concentration , current speed , For the The squid biological activity parameter vectors recorded in the , respectively, represent the distribution location coordinates Biomass density index ; S12. Real-time multi-source datasets Perform time synchronization processing, uniformly map multi-source data from different devices and different time scales to a standard time series set, use a unified time step to time align all multi-source data, and construct a time-synchronized multi-source data set so that all parameters have a consistent time reference benchmark; S13. Perform spatial interpolation processing on the time-synchronized multi-source datasets. For areas with missing measurements in geographic space, interpolation estimation is performed based on the neighboring values ​​of the observed locations and their spatial positional relationships to fill in the missing spatial locations, thereby forming a spatially completed multi-source dataset, so that the marine environmental parameters and squid biological activity parameters have complete spatial coverage in the target area; S14. Fill missing values ​​in the spatially completed multi-source dataset. For the marine environmental parameters or squid biological activity parameter vectors that are still missing in non-spatial dimensions, use the sliding time window method to fill in the data based on the historical observations at the nearest moment. The filled data form the multi-source dataset after missing values ​​are processed. S15. Remove outliers from the multi-source data set after missing values ​​processing, calculate the historical mean and standard deviation of each parameter, and determine whether the current observation value of the multi-source data set after missing values ​​processing is within the preset reasonable range. If the deviation exceeds the set threshold, the record is marked as abnormal data and removed, and finally the pre-processed multi-source data set is formed. .

[0009] Optionally, S2 includes the following steps: S21. Constructing a Bayesian deep inversion network model The Bayesian deep inversion network model includes an input layer, a dual-channel feature extraction structure, a fusion nesting layer, an output layer, and a dynamic prior model. The input layer receives the preprocessed multi-source data set. , used to study the dynamic distribution behavior characteristics of squid resources under different marine environmental conditions; S22. Construct a dual-channel feature extraction structure for the Bayesian depth inversion network model. The dual-channel feature extraction structures are the ocean environment channel Channel with Squid Behavior , used to simulate the response mechanism of squid to environmental stimuli; S23. Set a fusion nesting layer to perform feature-level connection operations on the feature vectors of the ocean environment channel and the squid behavior channel to obtain a fusion feature representation. ; S24. Set the squid resource estimation output of the output layer of the Bayesian deep inversion network model : ; in, To predict the spatial coordinates of the squid, To predict squid density estimates and dynamically estimate the spatial distribution and quantity status of squid resources; S25. Constructing a dynamic prior model , based on historical marine environmental parameters and squid biological activity parameters, a dynamic prior model with time weight is constructed: ; in, Represents the current moment generated by the dynamic prior model Dynamic prior data of squid resources, The history The ocean environment parameters and squid biological activity parameters at each moment, is the time decay weight, Extraction function for statistical features; S27. Dynamic prior data of squid resources and fusion feature representation Perform Bayesian posterior modeling and calculate the posterior output distribution of the dynamic estimation of squid resources: ; in, Indicates that in the fusion feature representation Given the conditions, the squid resource estimation output is The Bayesian posterior distribution of Represents the estimated output for a given squid resource When , the fusion feature representation The likelihood function of .

[0010] Optionally, in the marine environment channel The residual structure is constructed in the process to extract stable background change features: ; in, For the Layer environmental feature representation, which represents the implicit response characteristics of squid to marine environmental parameters, Indicates Layer environment feature representation, and are weights and biases, The Sigmoid function represents the attention weight, and f is the activation function; In the Squid Behavior Channel The attention mechanism is introduced to enhance the recognition ability of squid gathering area features: ; in, The Sigmoid function represents the attention weight, is the Hadamard product, For the Layer attention map parameters, For the The output feature vector of the squid behavior channel represents the spatial behavior characteristics of individual or group squids. For the Output feature vector of the layer squid behavior channel.

[0011] Optionally, S3 includes the following steps: S31. Using preprocessed multi-source datasets Bayesian Deep Inversion Network Model Training is performed, and the training samples are composed of multiple input and output pairs, where each input vector is composed of the concatenation of marine environmental parameters and squid biological activity parameters, and the output label is composed of the corresponding squid real distribution position coordinates and density values; S32. During the training process, construct a Bayesian log-likelihood loss function including a dynamic prior constraint; S33. Before each training batch starts, the dynamic prior model is called to calculate the prior model based on the current training time step. The historical ocean environment parameters and squid biological activity parameters at each time point are used to update the squid resource dynamic prior data. The updated squid resource dynamic prior data is the squid resource dynamic prior data at the current time step, which is used to participate in the calculation of the posterior probability in the training loss, so that the training process continues to reflect the temporal evolution characteristics of the ocean environment and the dynamic changes of squid behavior. S34. In each training iteration, the Bayesian log-likelihood loss function is calculated using the training sample input of the current batch and the corresponding output label, and all trainable parameters in the Bayesian deep inversion network model are optimized by gradient descent based on the Bayesian log-likelihood loss function. The parameter update process uses the learning rate as the control factor, and the model weights are gradually corrected according to the current gradient direction; S35. Continue to perform training iterations until the Bayesian log-likelihood loss function becomes stable on the validation set or reaches the preset upper limit of the number of training rounds. After the training is completed, a set of trained model parameters is formed.

[0012] Optionally, the Bayesian log-likelihood loss function consists of two parts: the first part is the log-likelihood term of the posterior distribution, which is used to evaluate the degree of matching of the model output results under the guidance of dynamic priors; the second part is the regularization term, which is used to constrain the amplitude of the network weight parameters to prevent the model from overfitting in marine multi-source data. The minimization goal of the Bayesian log-likelihood loss function is to maximize the credibility of the estimation results.

[0013] Optionally, S4 includes the following steps: S41. The input vector in the real-time multi-source data set is sent to the input layer of the trained Bayesian deep inversion network model, and processed in sequence through the ocean environment channel, the squid behavior channel and the fusion nested layer to generate a fused feature representation, which comprehensively reflects the nonlinear coupling relationship between the current ocean environment changes and the squid individual behavior characteristics; S42. Call the dynamic prior model to update the prior distribution at the current time. The update method is based on the historical marine environmental parameters and squid biological activity parameters recorded in multiple consecutive time steps before the current time point. The historical parameters are assigned decreasing weights in chronological order. The dynamic prior data of squid resources corresponding to the current time point is calculated. The dynamic prior data of squid resources is used to express the empirical distribution trend of squid resources under the current environmental state; S43. The fusion feature representation and the squid resource dynamic prior data are used together in the Bayesian posterior inference process, and the posterior distribution of the squid resource distribution position and density prediction value of the squid resource at the current moment is estimated according to the Bayesian principle in combination with the likelihood information under the current feature conditions and the squid resource dynamic prior data; S44. Calculate the uncertainty quantification index corresponding to the current prediction result based on the posterior distribution of squid resources, the uncertainty index includes the variance of the predicted position coordinates in three spatial dimensions and the volatility variance of the squid density estimation result; S45. Output the dynamic estimation result of squid resources at the current moment, including the predicted position coordinates and density value, and simultaneously output the corresponding uncertainty quantitative index.

[0014] Optionally, S5 includes the following steps: S51. Dynamic estimation result of squid resources output at the current moment The corresponding uncertainty quantitative index Structuring: ; in, , , They represent the variance estimates of the three spatial dimension components of the predicted position under the Bayesian posterior distribution, Represents the variance of the squid density prediction at the current time step; The squid spatial position coordinates in the squid resource dynamic estimation result and the estimated squid density at time step t Construct a squid resource estimation data table based on time series and geographical area dimensions; S52. Introducing uncertainty information based on the squid resource estimation data table, establishing a confidence label for each prediction cell, and dividing the confidence levels; S53. Based on the sorted predicted squid spatial position coordinates and squid density estimation, combined with the sea area grid division rules, regional aggregation statistics of squid resources are performed, and the average squid density, density change rate and confidence level distribution per unit time in each sub-area are calculated to form regional-level spatial aggregation characteristics; S54. Construct a dynamic estimation report structure for squid resources. The report includes the following information fields: Time label: current prediction time step ; Location label: the spatial grid number of the predicted sea area; Prediction results: Squid spatial position coordinates , Squid density estimation ; Uncertainty quantification index: a set of indicators consisting of position coordinate variance and density variance ; Confidence level: high confidence / medium confidence / low confidence classification label; Spatial aggregation characteristics: regional average density value, regional density volatility index, local density gradient trend index Optionally, the credibility level classification rule is: High confidence area: If , marked as high confidence; Medium Trusted Area: If , marked as medium confidence; Low confidence zone: If , marked as low confidence; in, and Threshold the uncertainty interval.

[0015] The beneficial effects of the present invention are: (1) The present invention constructs a dual-channel structure of marine environment channel and squid behavior channel, extracts independent feature representations of temperature, salinity, and flow velocity physical parameters and squid distribution coordinates and density behavior parameters respectively, and performs feature cascading in the nested fusion layer to effectively capture the nonlinear response characteristics of environmental changes to squid aggregation behavior.

[0016] (2) The present invention designs a dynamic prior generation module, which performs weighted fusion of environmental data and behavioral parameters in the past K time steps based on the time decay function, generates a prior estimation distribution at time t, and performs Bayesian posterior inference with the current fusion features, so that the model can make estimates based on historical trends at each prediction time point, reflecting the inertia and delay characteristics of the marine ecosystem, and significantly enhancing the model's adaptability to complex background changes in the ocean.

[0017] (3) The present invention obtains the posterior distribution variance of the predicted squid position and density through the Bayesian model structure, and defines the confidence level label based on it to mark the credibility of each predicted grid unit. In addition, the thresholds δ1 and δ2 are set based on the variance aggregation index to achieve regional-level prediction credibility screening and aggregation statistical support, which significantly reduces the error prediction rate in the low-confidence prediction area and improves the prediction stability in the high-confidence area, providing more reliable support data for fishery vessel scheduling and fishing route planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 The present invention provides a flow chart of a method for estimating dynamic squid resources based on a Bayesian deep inversion network. DETAILED DESCRIPTION

[0019] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0020] refer to Figure 1 , a squid resource dynamic estimation method based on Bayesian deep inversion network, comprising the following steps: S1. Collecting a real-time multi-source data set including marine environment parameters and squid biological activity parameters, preprocessing the real-time multi-source data set to form a preprocessed multi-source data set; S2. Construct a Bayesian deep inversion network model; S3. Use the preprocessed multi-source data set to train the Bayesian deep inversion network model, use the dynamic prior update module to update the dynamic prior data in real time during the training process, and optimize the model parameters by maximizing the likelihood function and minimizing the prediction error to form a set of trained model parameters; S4. Input the multi-source data set collected in real time into the trained Bayesian deep inversion network model, and use the dynamic prior update module to update the prior data of the real-time multi-source data set to obtain the dynamic estimation results of squid resources and their uncertainty quantitative indicators; S5. Post-process the squid resource dynamic estimation results and uncertainty quantitative indicators to form a squid resource dynamic estimation report with temporal and spatial distribution characteristics; S6. Interface the squid resource dynamic estimation report with the fishery management system to conduct dynamic monitoring of squid resources and real-time decision-making.

[0021] In this implementation, S1 includes the following steps: S11. Collect real-time multi-source data including marine environmental parameters and squid biological activity parameters from ocean observation platforms, satellite remote sensing devices and underwater sensors to build a real-time multi-source data set : ; in, Indicates Original data records, Indicates the total number of real-time collected data. For the The collection timestamp of the data. For the The ocean environment parameter vectors recorded in ,salinity , chlorophyll concentration , current speed For the The squid biological activity parameter vectors recorded in the , respectively, represent the distribution location coordinates Biomass density index ; S12. Real-time multi-source datasets Perform time synchronization processing, uniformly map multi-source data from different devices and different time scales to a standard time series set, use a unified time step to time align all multi-source data, and construct a time-synchronized multi-source data set so that all parameters have a consistent time reference benchmark; S13. Perform spatial interpolation processing on the time-synchronized multi-source datasets. For areas with missing measurements in geographic space, interpolation estimation is performed based on the neighboring values ​​of the observed locations and their spatial positional relationships to fill in the missing spatial locations, thereby forming a spatially completed multi-source dataset, so that the marine environmental parameters and squid biological activity parameters have complete spatial coverage in the target area; S14. Fill missing values ​​in the spatially completed multi-source dataset. For the marine environmental parameters or squid biological activity parameter vectors that are still missing in non-spatial dimensions, use the sliding time window method to fill in the data based on the historical observations at the nearest moment. The filled data form the multi-source dataset after missing values ​​are processed. S15. Remove outliers from the multi-source data set after missing values ​​processing, calculate the historical mean and standard deviation of each parameter, and determine whether the current observation value of the multi-source data set after missing values ​​processing is within the preset reasonable range. If the deviation exceeds the set threshold, the record is marked as abnormal data and removed, and finally the pre-processed multi-source data set is formed. .

[0022] In this implementation, S2 includes the following steps: S21. Constructing a Bayesian deep inversion network model The Bayesian deep inversion network model includes an input layer, a dual-channel feature extraction structure, a fusion nesting layer, an output layer, and a dynamic prior model. The input layer receives the preprocessed multi-source data set. , used to study the dynamic distribution behavior characteristics of squid resources under different marine environmental conditions; S22. Construct a dual-channel feature extraction structure for the Bayesian depth inversion network model. The dual-channel feature extraction structures are the ocean environment channel Channel with Squid Behavior , used to simulate the response mechanism of squid to environmental stimuli; S23. Set a fusion nesting layer to perform feature-level connection operations on the feature vectors of the ocean environment channel and the squid behavior channel to obtain a fusion feature representation. ; S24. Set the squid resource estimation output of the output layer of the Bayesian deep inversion network model : ; in, To predict the spatial coordinates of the squid, To predict squid density estimates and dynamically estimate the spatial distribution and quantity status of squid resources; S25. Constructing a dynamic prior model , based on historical marine environmental parameters and squid biological activity parameters, a dynamic prior model with time weight is constructed: ; in, Represents the current moment generated by the dynamic prior model Dynamic prior data of squid resources, The history The ocean environment parameters and squid biological activity parameters at each moment, is the time decay weight, Extraction function for statistical features; S27. Dynamic prior data of squid resources and fusion feature representation Perform Bayesian posterior modeling and calculate the posterior output distribution of the dynamic estimation of squid resources: ; in, Indicates that in the fusion feature representation Given the conditions, the squid resource estimation output is The Bayesian posterior distribution of Represents the estimated output for a given squid resource When , the fusion feature representation The likelihood function of .

[0023] In this embodiment, in the marine environment channel The residual structure is constructed in the process to extract stable background change features: ; in, For the Layer environmental feature representation, which represents the implicit response characteristics of squid to marine environmental parameters, Indicates Layer environment feature representation, and are weights and biases, The Sigmoid function represents the attention weight, and f is the activation function; In the Squid Behavior Channel The attention mechanism is introduced to enhance the recognition ability of squid gathering area features: ; in, The Sigmoid function represents the attention weight, is the Hadamard product, For the Layer attention map parameters, For the The output feature vector of the squid behavior channel represents the spatial behavior characteristics of individual or group squids. For the Output feature vector of the layer squid behavior channel.

[0024] In this implementation, S3 includes the following steps: S31. Using preprocessed multi-source datasets Bayesian Deep Inversion Network Model Training is performed, and the training samples are composed of multiple input and output pairs, where each input vector is composed of the concatenation of marine environmental parameters and squid biological activity parameters, and the output label is composed of the corresponding squid real distribution position coordinates and density values; S32. During the training process, construct a Bayesian log-likelihood loss function including a dynamic prior constraint; S33. Before each training batch starts, the dynamic prior model is called to calculate the prior model based on the current training time step. The historical ocean environment parameters and squid biological activity parameters at each time point are used to update the squid resource dynamic prior data. The updated squid resource dynamic prior data is the squid resource dynamic prior data at the current time step, which is used to participate in the calculation of the posterior probability in the training loss, so that the training process continues to reflect the temporal evolution characteristics of the ocean environment and the dynamic changes of squid behavior. S34. In each training iteration, the Bayesian log-likelihood loss function is calculated using the training sample input of the current batch and the corresponding output label, and all trainable parameters in the Bayesian deep inversion network model are optimized by gradient descent based on the Bayesian log-likelihood loss function. The parameter update process uses the learning rate as the control factor, and the model weights are gradually corrected according to the current gradient direction; S35. Continue to perform training iterations until the Bayesian log-likelihood loss function becomes stable on the validation set or reaches the preset upper limit of the number of training rounds. After the training is completed, a set of trained model parameters is formed.

[0025] In this implementation, the Bayesian log-likelihood loss function consists of two parts: the first part is the log-likelihood term of the posterior distribution, which is used to evaluate the degree of matching of the model output results under the guidance of dynamic priors; the second part is the regularization term, which is used to constrain the amplitude of the network weight parameters to prevent the model from overfitting in marine multi-source data. The minimization goal of the Bayesian log-likelihood loss function is to maximize the credibility of the estimation results.

[0026] In this implementation, S4 includes the following steps: S41. The input vector in the real-time multi-source data set is sent to the input layer of the trained Bayesian deep inversion network model, and processed in sequence through the ocean environment channel, the squid behavior channel and the fusion nested layer to generate a fused feature representation, which comprehensively reflects the nonlinear coupling relationship between the current ocean environment changes and the squid individual behavior characteristics; S42. Call the dynamic prior model to update the prior distribution at the current time. The update method is based on the historical marine environmental parameters and squid biological activity parameters recorded in multiple consecutive time steps before the current time point. The historical parameters are assigned decreasing weights in chronological order. The dynamic prior data of squid resources corresponding to the current time point is calculated. The dynamic prior data of squid resources is used to express the empirical distribution trend of squid resources under the current environmental state; S43. The fusion feature representation and the squid resource dynamic prior data are used together in the Bayesian posterior inference process, and the posterior distribution of the squid resource distribution position and density prediction value of the squid resource at the current moment is estimated according to the Bayesian principle in combination with the likelihood information under the current feature conditions and the squid resource dynamic prior data; S44. Calculate the uncertainty quantification index corresponding to the current prediction result based on the posterior distribution of squid resources, the uncertainty index includes the variance of the predicted position coordinates in three spatial dimensions and the volatility variance of the squid density estimation result; S45. Output the dynamic estimation result of squid resources at the current moment, including the predicted position coordinates and density value, and simultaneously output the corresponding uncertainty quantitative index.

[0027] In this implementation, S5 includes the following steps: S51. Dynamic estimation result of squid resources output at the current moment The corresponding uncertainty quantitative index Structuring: ; in, , , They represent the variance estimates of the three spatial dimension components of the predicted position under the Bayesian posterior distribution, Represents the variance of the squid density prediction at the current time step; The squid spatial position coordinates in the squid resource dynamic estimation result and the estimated squid density at time step t Construct a squid resource estimation data table based on time series and geographical area dimensions; S52. Introducing uncertainty information based on the squid resource estimation data table, establishing a confidence label for each prediction cell, and dividing the confidence levels; S53. Based on the sorted predicted squid spatial position coordinates and squid density estimation, combined with the sea area grid division rules, regional aggregation statistics of squid resources are performed, and the average squid density, density change rate and confidence level distribution per unit time in each sub-area are calculated to form regional-level spatial aggregation characteristics; The realization process of spatial aggregation features at the regional level is based on the completed squid resource dynamic estimation results and uncertainty quantification indicators, which include the following implementation methods: Combined with the output squid space position coordinates Density estimation with squid , all prediction results are classified according to the three-dimensional grid spatial position of the sea area. The three-dimensional grid division rule of the sea area is divided into multiple spatial cells according to the fixed longitude and latitude and depth resolution preset, and each cell corresponds to a spatial index number.

[0028] Based on the multiple prediction points belonging to each spatial cell, the time dimension is organized according to the unit time window. In each time window, the squid density estimation of all prediction points in each grid cell is counted. And calculate its regional average density value to characterize the resource abundance characteristics of the region at the current time point.

[0029] Adjacent time steps were differentiated in the time window sequence to extract the squid density changes in the same spatial cell within a unit time interval, and then the density change rate of the cell was calculated to reflect the dynamic trend of squid resource changes in the area.

[0030] Combined with the uncertainty quantification index set , the confidence of multiple prediction points in each grid cell is aggregated and analyzed according to the variance weighting method of each prediction position or directly using the maximum variance value. The average uncertainty level of the area is calculated by normalizing and superimposing the uncertainty indicators. Based on this, the confidence level distribution is marked at the regional level, and the spatial distribution mapping of high confidence areas, medium confidence areas, and low confidence areas is formed.

[0031] Finally, the results are output as regional statistical features to constitute the spatial aggregation feature items of each spatial grid unit in the current time window, including: regional average squid density, regional squid density change rate, and regional confidence level distribution.

[0032] S54. Construct a dynamic estimation report structure for squid resources. The report includes the following information fields: Time label: current prediction time step ; Location label: the spatial grid number of the predicted sea area; Prediction results: Squid spatial position coordinates , Squid density estimation ; Uncertainty quantification index: a set of indicators consisting of position coordinate variance and density variance ; Confidence level: high confidence / medium confidence / low confidence classification label; Spatial aggregation characteristics: regional average density value, regional density volatility index, local density gradient trend index In this implementation, the credibility level classification rule is: High confidence area: If , marked as high confidence; Medium Trusted Area: If , marked as medium confidence; Low confidence zone: If , marked as low confidence; in, and Threshold the uncertainty interval.

[0033] Example 1: At 4:12 a.m. on October 3, 2024, in the sea area east of Island A, the duty personnel of the Marine Resources Information Center noticed an abnormal phenomenon in the monitoring system: the real-time observation data deployed on the "A Island-07" intelligent buoy platform showed that the temperature of the near-bottom seawater suddenly dropped from 22.4°C to 20.7°C, and the salinity rose from 32.8‰ to 33.5‰. At the same time, the current speed fluctuated violently for a short time to 1.93 m / s. This abnormal phenomenon lasted for about 18 minutes. The historical squid resource dynamic estimation report within 30 kilometers of the sea area was retrieved through the dispatching platform. It was found that in the past 72 hours, the estimated squid density in the area showed a stable fluctuation trend, with an average value of about 6.2kg / km². However, starting from 4:30 a.m., the underwater acoustic sensor deployed on the "A Island-07" buoy began to continuously detect high-frequency echo anomalies, which was initially judged to be squid aggregation.

[0034] Since the traditional estimation model (based on time series LSTM) usually requires more than 6 hours of historical data playback to make an estimation prediction, and the prediction does not have uncertainty differentiation, the on-duty personnel decided to activate the Bayesian deep inversion resource estimation system deployed on the "A Island-Resource Control Node 02" server of the present invention to perform a real-time dynamic inversion.

[0035] At 4:46 a.m., the system began to receive linkage data packets sent by the "A Island-07" buoy, the "B Island-03" submerged buoy, and the "C Island-02" remote sensing data processing terminal. The system quickly processed three types of data: A total of 8923 collection records were processed in this round of estimation; Each record contains timestamp, seawater temperature, salinity, chlorophyll concentration, current speed, and underwater acoustic estimated density and initial coordinate point; The processed data formed a complete time-synchronized sample set with a missing rate of only 1.8% and 273 outliers removed.

[0036] The system completed the real-time update of the dynamic prior model at 4:52 a.m.: it selected historical estimation records in the same sea area within the past 180 minutes (36 time points in total), calculated the prior density distribution by exponential decay weighting, and the Bayesian inversion model completed the grid-level resource inversion of the entire sea area in just 32 seconds after integrating the current input features with the dynamic prior.

[0037] At this time, the system generated a "Squid Resource Dynamic Inversion Forecast Report" in the background, which mentioned: Prediction time: 2024-10-03 04:52; Target area: 123.107°E, 29.744°N, depth -38.2 meters; Predicted squid density value: 9.73 kg / km²; Coordinate space prediction variance: (σx=5.3m, σy=6.1m, σz=3.8m); Density variance: 0.44 kg / km²; Confidence level: High confidence.

[0038] According to the automatic confidence differentiation strategy, the system highlighted the area of ​​approximately 19.4 square kilometers in the "southeast area of ​​the line connecting Island A and Island B" as a high-confidence prediction area, and recommended the deployment of operating vessels to the area to perform "fast exploration and slow enclosure" net operations. The report was immediately pushed to the municipal bureau dispatch end, and an operation suggestion notification was sent to the "32089" fishing boat.

[0039] The fishing boat departed from the port at 5:30 am and arrived at the target area at 7:20 am. The on-site trawling operation was carried out for 3 rounds, each round lasting 30 minutes. The final operation statistics are as follows: The first round of trawling: the total amount of squid caught was 714 kg, the net mouth covered an area of ​​7.1 km², and the measured density was 10.04 kg / km²; Second round of trawl: density measured 9.87 kg / km²; The third round of trawl: the density was measured at 10.23 kg / km²; Average measured density over 3 rounds: 10.05 kg / km²; The deviation from the predicted value of 9.73 kg / km² of the present invention is only 3.18%, and the confidence interval completely includes the measured value, verifying the high reliability.

[0040] For comparison, during the same period, the municipal bureau also used the still-running traditional LSTM resource estimation system to make predictions for the area, which returned a predicted density of 6.2 kg / km² with a position deviation of approximately 2.7 kilometers. It was not marked as a fishing hotspot. If the traditional system had been relied upon, the day's operations would have completely bypassed the high-density area, and the prediction effect would have been far inferior to that of the system of the present invention.

[0041] The system also automatically generated a predicted reliability distribution map, a regional heat overlay map, a volatility assessment map, and an operation suggestion layer during this operation, and archived them in the Municipal Bureau's "Resource Decision-making Support Platform."

[0042] In addition, a systematic evaluation was conducted on all forecast data for the A Island waters from September 30 to October 5:

[0043] The comparative data fully verified the significant advantages of the system of the present invention in terms of dynamic response speed, prediction space accuracy, density estimation error control, uncertainty modeling and visualization output. Especially in the context of complex environmental disturbances and rapid ocean changes, the system exhibits stable, reliable and practical characteristics.

[0044] The present invention constructs a dual-path structure of marine environment channel and squid behavior channel, respectively extracts independent feature representations of temperature, salinity, flow velocity physical parameters and squid distribution coordinates, density behavior parameters, and performs feature cascading in a nested fusion layer to effectively capture the nonlinear response characteristics of environmental changes to squid aggregation behavior.

[0045] The present invention designs a dynamic prior generation module, which performs weighted fusion of environmental data and behavioral parameters in the past K time steps based on the time decay function, generates a prior estimation distribution at time t, and performs Bayesian posterior inference with the current fusion features, so that the model can make estimates based on historical trends at each prediction time point, reflecting the inertia and delay characteristics of the marine ecosystem, and significantly enhancing the adaptability of the model to complex background changes in the ocean.

[0046] The present invention obtains the posterior distribution variance of the predicted squid position and density through the Bayesian model structure, and defines the confidence level label accordingly to mark the credibility of each predicted grid unit. In addition, the thresholds δ1 and δ2 are set based on the variance aggregation index to achieve regional-level prediction credibility screening and aggregation statistical support, which significantly reduces the error prediction rate in low-confidence prediction areas and improves the prediction stability in high-confidence areas, providing more reliable support data for fishery vessel scheduling and fishing route planning.

[0047] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for estimating squid resources dynamics based on a Bayesian deep inversion network, characterized in that: The steps include: S1. Collecting a real-time multi-source data set including marine environment parameters and squid biological activity parameters, preprocessing the real-time multi-source data set to form a preprocessed multi-source data set; S2. Construct a Bayesian deep inversion network model; S3. Use the preprocessed multi-source data set to train the Bayesian deep inversion network model, use the dynamic prior update module to update the dynamic prior data in real time during the training process, and optimize the model parameters by maximizing the likelihood function and minimizing the prediction error to form a set of trained model parameters; S4. Input the multi-source data set collected in real time into the trained Bayesian deep inversion network model, and use the dynamic prior update module to update the prior data of the real-time multi-source data set to obtain the dynamic estimation results of squid resources and their uncertainty quantitative indicators; S5. Post-process the squid resource dynamic estimation results and uncertainty quantitative indicators to form a squid resource dynamic estimation report with temporal and spatial distribution characteristics; S6. Interface the squid resource dynamic estimation report with the fishery management system to conduct dynamic monitoring of squid resources and real-time decision-making.

2. The method for estimating squid resources dynamics based on a Bayesian deep inversion network according to claim 1, characterized in that: The S1 comprises the following steps: S11. Collect real-time multi-source data including marine environmental parameters and squid biological activity parameters from ocean observation platforms, satellite remote sensing devices and underwater sensors to build a real-time multi-source data set : ； in, Indicates Original data records, Indicates the total number of real-time collected data. For the The collection timestamp of the data. For the The ocean environment parameter vectors recorded in ,salinity , chlorophyll concentration , current speed , For the The squid biological activity parameter vectors recorded in the 100 records represent the distribution location coordinates Biomass density index ; S12. Real-time multi-source datasets Perform time synchronization processing, uniformly map multi-source data from different devices and different time scales to a standard time series set, use a unified time step to time align all multi-source data, and construct a time-synchronized multi-source data set so that all parameters have a consistent time reference benchmark; S13. Perform spatial interpolation processing on the time-synchronized multi-source datasets. For areas with missing measurements in geographic space, interpolation estimation is performed based on the neighboring values ​​of the observed locations and their spatial positional relationships to fill in the missing spatial locations, thereby forming a spatially completed multi-source dataset, so that the marine environmental parameters and squid biological activity parameters have complete spatial coverage in the target area; S14. Fill missing values ​​in the spatially completed multi-source dataset. For the marine environmental parameters or squid biological activity parameter vectors that are still missing in non-spatial dimensions, use the sliding time window method to fill in the data based on the historical observations at the nearest moment. The filled data form the multi-source dataset after missing values ​​are processed. S15. Remove outliers from the multi-source data set after missing values ​​processing, calculate the historical mean and standard deviation of each parameter, and determine whether the current observation value of the multi-source data set after missing values ​​processing is within the preset reasonable range. If the deviation exceeds the set threshold, the record is marked as abnormal data and removed, and finally the pre-processed multi-source data set is formed. .

3. The method for dynamic estimation of squid resources based on Bayesian deep inversion network according to claim 2 is characterized in that: The S2 comprises the following steps: S21. Constructing a Bayesian deep inversion network model The Bayesian deep inversion network model includes an input layer, a dual-channel feature extraction structure, a fusion nesting layer, an output layer, and a dynamic prior model. The input layer receives the preprocessed multi-source data set. , used to study the dynamic distribution behavior characteristics of squid resources under different marine environmental conditions; S22. Construct a dual-channel feature extraction structure for the Bayesian depth inversion network model. The dual-channel feature extraction structures are the ocean environment channel Channel with Squid Behavior , used to simulate the response mechanism of squid to environmental stimuli; S23. Set a fusion nesting layer to perform feature-level connection operations on the feature vectors of the ocean environment channel and the squid behavior channel to obtain a fusion feature representation. ; S24. Set the squid resource estimation output of the output layer of the Bayesian deep inversion network model ; S25. Constructing a dynamic prior model , based on historical marine environmental parameters and squid biological activity parameters, a dynamic prior model with time weight is constructed: ； in, Represents the current moment generated by the dynamic prior model Dynamic prior data of squid resources, The history The ocean environment parameters and squid biological activity parameters at each moment, is the time decay weight, Extraction function for statistical features; S27. Dynamic prior data of squid resources and fusion feature representation Bayesian posterior modeling was performed to calculate the posterior output distribution of squid resource dynamics estimates.

4. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 3 is characterized in that: In the marine environment channel The residual structure is constructed in the process to extract stable background change features: ； in, For the Layer environmental feature representation, which represents the implicit response characteristics of squid to marine environmental parameters, Indicates Layer environment feature representation, and are weights and biases, The Sigmoid function represents the attention weight, and f is the activation function; In the Squid Behavior Channel The attention mechanism is introduced to enhance the recognition ability of squid gathering area features: ； in, The Sigmoid function represents the attention weight, is the Hadamard product, For the Layer attention map parameters, For the The output feature vector of the squid behavior channel represents the spatial behavior characteristics of individual or group squids. For the Output feature vector of the layer squid behavior channel.

5. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 3 is characterized in that: The S3 comprises the following steps: S31. Using preprocessed multi-source datasets Bayesian Deep Inversion Network Model Training is performed, and the training samples are composed of multiple input and output pairs, where each input vector is composed of the concatenation of marine environmental parameters and squid biological activity parameters, and the output label is composed of the corresponding squid real distribution position coordinates and density values; S32. During the training process, construct a Bayesian log-likelihood loss function including a dynamic prior constraint; S33. Before each training batch starts, the dynamic prior model is called to calculate the prior model based on the current training time step. The historical ocean environment parameters and squid biological activity parameters at each time point are used to update the squid resource dynamic prior data. The updated squid resource dynamic prior data is the squid resource dynamic prior data at the current time step, which is used to participate in the calculation of the posterior probability in the training loss, so that the training process continues to reflect the temporal evolution characteristics of the ocean environment and the dynamic changes of squid behavior. S34. In each training iteration, the Bayesian log-likelihood loss function is calculated using the training sample input of the current batch and the corresponding output label, and all trainable parameters in the Bayesian deep inversion network model are optimized by gradient descent based on the Bayesian log-likelihood loss function. The parameter update process uses the learning rate as the control factor, and the model weights are gradually corrected according to the current gradient direction; S35. Continue to perform training iterations until the Bayesian log-likelihood loss function becomes stable on the validation set or reaches the preset upper limit of the number of training rounds. After the training is completed, a set of trained model parameters is formed.

6. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 5 is characterized in that: The Bayesian log-likelihood loss function consists of two parts: the first part is the log-likelihood term of the posterior distribution, which is used to evaluate the degree of matching of the model output results under the guidance of dynamic priors; the second part is the regularization term, which is used to constrain the amplitude of the network weight parameters to prevent the model from overfitting in marine multi-source data. The minimization goal of the Bayesian log-likelihood loss function is to maximize the credibility of the estimation results.

7. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 5 is characterized in that: The S4 comprises the following steps: S41. The input vector in the real-time multi-source data set is sent to the input layer of the trained Bayesian deep inversion network model, and processed in sequence through the ocean environment channel, the squid behavior channel and the fusion nested layer to generate a fused feature representation, which comprehensively reflects the nonlinear coupling relationship between the current ocean environment changes and the squid individual behavior characteristics; S42. Call the dynamic prior model to update the prior distribution at the current time. The update method is based on the historical marine environmental parameters and squid biological activity parameters recorded in multiple consecutive time steps before the current time point. The historical parameters are assigned decreasing weights in chronological order. The dynamic prior data of squid resources corresponding to the current time point is calculated. The dynamic prior data of squid resources is used to express the empirical distribution trend of squid resources under the current environmental state; S43. The fusion feature representation and the squid resource dynamic prior data are used together in the Bayesian posterior inference process, and the posterior distribution of the squid resource distribution position and density prediction value of the squid resource at the current moment is estimated according to the Bayesian principle in combination with the likelihood information under the current feature conditions and the squid resource dynamic prior data; S44. Calculate the uncertainty quantification index corresponding to the current prediction result based on the posterior distribution of squid resources, the uncertainty index includes the variance of the predicted position coordinates in three spatial dimensions and the volatility variance of the squid density estimation result; S45. Output the dynamic estimation result of squid resources at the current moment, including the predicted position coordinates and density value, and simultaneously output the corresponding uncertainty quantitative index.

8. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 7 is characterized in that: The S5 comprises the following steps: S51. Dynamic estimation result of squid resources output at the current moment The corresponding uncertainty quantitative index Structuring: ； in, , , They represent the variance estimates of the three spatial dimension components of the predicted position under the Bayesian posterior distribution, Represents the variance of the squid density prediction at the current time step; The squid spatial position coordinates in the squid resource dynamic estimation result and the estimated squid density at time step t Construct a squid resource estimation data table based on time series and geographical area dimensions; S52. Introducing uncertainty information based on the squid resource estimation data table, establishing a confidence label for each prediction cell, and dividing the confidence levels; S53. Based on the sorted predicted squid spatial position coordinates and squid density estimation, combined with the sea area grid division rules, regional aggregation statistics of squid resources are performed, and the average squid density, density change rate and confidence level distribution per unit time in each sub-area are calculated to form regional-level spatial aggregation characteristics; S54. Construct a dynamic estimation report structure for squid resources. The report includes the following information fields: Time label: current prediction time step ; Location tag: spatial grid number of the predicted sea area; Prediction results: Squid spatial position coordinates , Squid density estimation ; Uncertainty quantification index: a set of indicators consisting of position coordinate variance and density variance ; Confidence level: high confidence / medium confidence / low confidence classification label; Spatial aggregation characteristics: regional average density value, regional density volatility index, local density gradient trend index.

9. The method for estimating squid resources dynamics based on Bayesian deep inversion network according to claim 8, characterized in that: The credibility level classification rules are as follows: High confidence area: If , marked as high confidence; Medium Trusted Area: If , marked as medium confidence; Low confidence zone: If , marked as low confidence; in, and Threshold the uncertainty interval.

Images