A multi-source marine environmental data revision method

Abstract

The application discloses a multi-source marine environment data revising method. The method is characterized in that: through preprocessing of original marine environment data, standard marine environment data is obtained; based on the standard marine environment data, a marine environment data prediction model is obtained; through the feature extraction capability of the marine environment data prediction model, the time sequence modeling capability of a bidirectional long short-term memory network, and the automatic focusing characteristic of an attention mechanism, the precision and reliability of data processing are improved; the marine environment data to be revised is input into the trained marine environment data prediction model, predicted marine environment data is obtained, the marine environment data to be revised is revised through the predicted marine environment data, and revised marine environment data is obtained, so that the precision and reliability of marine environment data revision are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of multi-source marine environment data revision, and in particular to a multi-source marine environment data revision method. BACKGROUND

[0002] With the progress of science and technology, the way of obtaining marine data is constantly diversified, including traditional field observation such as ship and buoy, as well as the application of satellite remote sensing and radar technology. However, the above methods have limitations: field observation is limited by geographical range and time interval; satellite remote sensing can provide continuous monitoring of a wide range, but is subject to weather conditions; radar technology can effectively monitor sea surface characteristics, but signal attenuation and reflection interference reduce data accuracy. In practical applications, in order to more comprehensively and accurately understand the marine environment, these multi-source data are often used comprehensively; however, due to various observation errors, instrument bias, environmental interference and uncertainty in the data processing process, marine data generally have quality problems, which directly affects the usability of the data and the accuracy of scientific research. In order to ensure the reliability and effectiveness of the data, it is necessary to revise these multi-source data.

[0003] However, the existing marine environment data revision methods mainly include system bias correction, numerical model assimilation, statistical regression analysis, etc., which have limitations in dealing with real-time, non-linear relationship and multi-source data fusion, resulting in low accuracy of the revised marine environment data.

[0004] Therefore, the existing marine environment data revision technology still has the technical problem of low revision accuracy. SUMMARY

[0005] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0006] The main purpose of the embodiments of the present disclosure is to provide a multi-source marine environment data revision method, which can improve the accuracy of the revised marine environment data.

[0007] The first aspect of the embodiments of the present application provides a multi-source marine environment data revision method, which comprises:

[0008] obtaining original marine environment data;

[0009] preprocessing the original marine environment data to obtain standard marine environment data;

[0010] inputting the standard marine environment data into a marine environment data prediction model to be trained to obtain a trained marine environment data prediction model;

[0011] input the marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data;

[0012] correct the marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data.

[0013] The embodiment of the application provides a multi-source marine environment data correction method, which comprises the following steps: preprocessing original marine environment data to obtain standard marine environment data; obtaining a trained marine environment data prediction model based on the standard marine environment data; inputting marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data; and correcting the marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data. In this way, the accuracy and reliability of marine environment data correction are effectively improved through the precise error calculation and adjustment of the model output.

[0014] In some embodiments of the application, the marine environment data prediction model to be trained comprises a feature extraction module, a time series modeling module and an attention module.

[0015] The method comprises the following steps:

[0016] The method comprises the following steps:

[0017] The method comprises the following steps:

[0018] The method comprises the following steps:

[0019] The method comprises the following steps:

[0020] The marine environment data prediction model to be trained is iteratively updated according to the loss function value until an iteration termination condition is met, so as to obtain the trained marine environment data prediction model.

[0021] In some embodiments of the present application, the feature extraction module comprises a first convolutional layer and a second convolutional layer connected in sequence, the first convolutional layer comprises 64 filters, a convolution kernel with a size of 1, and a first activation function, and the second convolutional layer comprises 64 filters, a convolution kernel with a size of 3, and the first activation function.

[0022] In some embodiments of the present application, the attention module comprises a similarity calculation module, an attention score module, and a weighted summation module.

[0023] The time series representation is input into the attention module, so that the attention module calculates the correlation of the time series to obtain a prediction value of the time series representation, which comprises:

[0024] The time series representation is input into the similarity calculation module to calculate the similarity of each key in the time series to obtain a similarity result of the time series.

[0025] The similarity result of the time series is input into the attention score module to obtain an attention score of the time series representation.

[0026] The attention score is input into the weighted summation module to obtain a prediction value of the time series representation.

[0027] In some embodiments of the present application, before the attention score is input into the weighted summation module to obtain a prediction value of the time series representation, the method further comprises:

[0028] The attention score is normalized to obtain a processed attention score.

[0029] The processed attention score is converted by the following formula to obtain an attention weight:

[0030]

[0031] wherein, t is a time step t , is an attention weight of a current time step t, is a current time step t processed attention score, T is a length of the time series;

[0032] The attention score is input into the weighted summation module to obtain a time series representation prediction value.

[0033] The attention weights are weighted summed by the following formula to obtain the time series representation prediction value:

[0034]

[0035] wherein, is a time series representation prediction value, is a hidden state of a time step t .

[0036] In some embodiments of the present application, the ocean environment data to be corrected is revised according to the predicted ocean environment data to obtain revised ocean environment data, including:

[0037] The difference between the predicted ocean environment data and the ocean environment data to be corrected is calculated;

[0038] The ocean environment data to be corrected is revised according to the difference by the following formula to obtain the revised ocean environment data:

[0039]

[0040]

[0041] wherein, is revised ocean environment data, is ocean environment data to be corrected, is predicted ocean environment data, is a difference.

[0042] In some embodiments of the present application, the loss function of the ocean environment data prediction model is:

[0043]

[0044] wherein, is a mean square error function, is a true value, is a prediction value, n is a sample number.

[0045] In a third aspect, the present application provides an electronic device, comprising at least one control processor and a memory connected to the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to perform the multi-source marine environmental data correction method.

[0046] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer executable instructions for causing a computer to perform the multi-source marine environmental data correction method.

[0047] It can be understood that the beneficial effects of the second aspect to the fourth aspect and the related technologies are the same as the beneficial effects of the first aspect and the related technologies, and the related description can be referred to in the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0048] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0049] Figure 1 is a flow diagram of a multi-source marine environmental data correction method provided by an embodiment of the present application;

[0050] FIG. 2(a) is a temperature map before correction of a multi-source marine environmental data correction method provided by an embodiment of the present application;

[0051] FIG. 2(b) is a temperature map after correction of a multi-source marine environmental data correction method provided by an embodiment of the present application;

[0052] Figure 3 is a structural diagram of a multi-source marine environmental data correction device provided by an embodiment of the present application;

[0053] Figure 4 is a structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0054] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, in which the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below are exemplary and are only used to explain the present application, and cannot be understood as a limitation of the present application.

[0055] In the description of the present application, if there is a description to first, second, etc. is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the sequence of indicated technical features.

[0056] In the description of the present application, it should be understood that the orientation description, such as the orientation or position relationship indicated by up, down, etc. is based on the orientation or position relationship shown in the drawings, only for the purpose of describing the present application and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0057] In the description of the present application, it should be noted that, unless otherwise explicitly limited, the words such as setting, installing, connecting, etc. should be broadly understood, and those skilled in the art can reasonably determine the specific meaning of the above words in the present application in combination with the specific content of the technical solution.

[0058] The ocean is of vital importance to human survival and development. Whether it is climate regulation, resource supply, or transportation and national security, it is impossible to do without a deep understanding of the marine environment. In marine scientific research and marine environmental monitoring, accurate marine environmental data is crucial. Accurate marine environmental data has important applications in the fields of marine forecasting, climate research, environmental protection, and marine resource management. With the continuous progress of science and technology, the means of obtaining marine environmental data is becoming increasingly abundant.

[0059] The conventional method of collecting marine environmental data is generally through in-situ observation equipment such as ships and buoys. This method can provide relatively direct and accurate local data, but is limited by the observation range and frequency. The advent of satellite remote sensing technology has greatly expanded the coverage of marine observation, enabling large-area, long-time series monitoring. However, satellite data is susceptible to cloud cover, atmospheric interference, and other factors during acquisition, resulting in uncertainty in data quality. Radar technology, as another important monitoring means, has unique advantages in monitoring marine surface features and dynamic changes, but also faces problems such as signal attenuation and reflection interference, affecting data accuracy. In practical applications, in order to more comprehensively and accurately understand the marine environment, it is often necessary to comprehensively use these multi-source data. However, due to various observation errors, instrument biases, environmental disturbances, and uncertainties in the data processing process, marine data generally has quality problems, directly affecting the usability of the data and the accuracy of scientific research. In order to ensure the reliability and effectiveness of the data, it is necessary to correct these multi-source data.

[0060] Currently, the correction techniques of marine environmental data mainly include systematic bias correction, numerical model assimilation, statistical regression analysis, etc. Systematic bias correction identifies and eliminates systematic errors by comparing historical data with reference standards; numerical model assimilation combines observation data with prediction models using numerical models to optimize prediction results; statistical regression analysis corrects data by establishing a statistical relationship between observation data and environmental variables. However, these methods have limitations in handling real-time, nonlinear relationships, and multi-source data fusion.

[0061] Therefore, with reference to Figure 1 The embodiment of the present application provides a multi-source marine environmental data correction method. The method is applied to a controller, which can be a server, an electronic device, a mobile terminal, etc. The method includes the following steps S101-S105:

[0062] Step S101, obtaining original marine environmental data.

[0063] In this step, the original marine environmental data is collected by multi-source marine environmental sensors (such as buoys, satellite remote sensing, ship observation, etc.). The marine environmental data includes but is not limited to seawater temperature, salinity, flow rate, wave height, wind speed, wind direction, and other marine environmental parameters.

[0064] Step S102, preprocessing the original marine environmental data to obtain standard marine environmental data.

[0065] In this step, the time format, spatial coordinate system, and measurement unit of the original marine environmental data are first adjusted for consistency. Then, data cleaning is performed through anomaly detection and elimination, missing value processing, etc. Anomaly detection and elimination mainly use statistical methods (box plot or standard deviation) and physical models to detect and eliminate obviously abnormal data points. Missing value processing mainly uses interpolation methods (linear interpolation) or historical data-based filling methods to handle missing values. Finally, the data is standardized to make different parameter data have the same dimension, and is normalized through the following formula to complete the preprocessing of the original marine environmental data:

[0066]

[0067] wherein, is the original data, is the mean of the data, is the standard deviation of the data.

[0068] Step S103, inputting the standard marine environmental data into a marine environmental data prediction model to be trained to obtain a trained marine environmental data prediction model.

[0069] In this step, the marine environment data prediction model to be trained is a multi-source marine environment data correction model based on a convolutional neural network, a bidirectional long short-term memory network and an attention mechanism, to fuse multi-source data from different sensors and platforms, and enhance the effectiveness and robustness of data analysis.

[0070] Step S104, input the marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data.

[0071] In this step, after obtaining the marine environment data to be corrected, first, the multi-level and local features in the input data are extracted using a convolutional neural network (CNN) to capture spatial features; then, the time series data is processed by a bidirectional long short-term memory network (BiLSTM) to comprehensively model the dynamic changes of the marine environment; finally, an attention mechanism is introduced to enable the model to automatically identify and focus on key time steps and features, improving the accuracy and interpretability of the prediction.

[0072] Step S105, correct the marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data.

[0073] In this step, the prediction result of the trained marine environment data model and the marine environment data to be corrected are used to calculate the correction value, and the marine environment data to be corrected is corrected by the correction value to quantify the difference between the model prediction and the actual, thereby improving the accuracy of the data.

[0074] Unlike the prior art, the embodiments of the present application preprocess the original marine environment data to obtain standard marine environment data, obtain a trained marine environment data prediction model based on the standard marine environment data, use the feature extraction capability of the marine environment data prediction model, the time series modeling capability of the bidirectional long short-term memory network, and the automatic focusing property of the attention mechanism to improve the accuracy and reliability of data processing, input the marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data, correct the marine environment data to be corrected by the predicted marine environment data to obtain corrected marine environment data, and thus effectively improve the accuracy and reliability of marine environment data correction through the accurate error calculation and adjustment of the model output.

[0075] The specific implementation of each of the above steps is described below.

[0076] Before the step S103 of inputting the standard marine environment data into the marine environment data prediction model to be trained to obtain the trained marine environment data prediction model, the method further includes the following steps S201 to S205.

[0077] Step S201, input the standard marine environment data into the feature extraction module, so that the feature extraction module extracts spatial features of the standard marine environment data.

[0078] Step S202, input the spatial features into the time series modeling module, so that the time series modeling module models the past and future time of the spatial features to obtain a time series representation.

[0079] Step S203, input the time series representation into the attention module, so that the attention module calculates the correlation of the time series to obtain a predicted value of the time series representation.

[0080] Step S204, determine the loss function value of the marine environment data prediction model to be trained according to the predicted value of the time series representation and the standard marine environment data.

[0081] Step S205, iteratively update the marine environment data prediction model to be trained according to the loss function value until the iteration termination condition is met, and obtain the trained marine environment data prediction model.

[0082] In this embodiment, the marine environment data prediction model to be trained includes a feature extraction module, a time series modeling module and an attention module.

[0083] In this embodiment, a convolutional neural network is used as the feature extraction module, which extracts local features of the input data through a convolutional layer, and reduces the spatial dimension of the features through a pooling layer. In this embodiment, the convolutional neural network consists of two one-dimensional convolutional layers (i.e. a first convolutional layer and a second convolutional layer), each followed by a one-dimensional max-pooling layer and a Dropout layer.

[0084] Specifically, the first convolutional layer includes 64 filters, a convolution kernel of size 1 and a first activation function, the second convolutional layer includes 64 filters, a convolution kernel of size 3 and the first activation function, and a one-dimensional pooling layer is connected after the first convolutional layer, which performs pooling with a window size of 2. The first activation function is a ReLU activation function.

[0085] In this embodiment, different convolution kernel sizes in the convolutional layer are used to capture different scales of temporal information in the marine environment data. The first convolutional layer has a smaller convolution kernel, which can effectively capture finer features, while the second convolutional layer has a larger convolution kernel, which can capture more extensive temporal patterns. The one-dimensional max-pooling layer reduces the number of parameters, reduces the risk of overfitting, while maintaining the representativeness of the captured features. The Dropout layer randomly discards the output of neurons to enhance the generalization of the model.

[0086] In the embodiment, the above convolutional neural network is adopted as the feature extraction module to extract spatial features of the standard marine environment data.

[0087] In the embodiment, a bidirectional long short-term memory network (Bi-LSTM) is adopted as the time series modeling module. The traditional recurrent neural network (RNN) is designed to process sequential data by passing hidden states across time steps to capture temporal dependencies. The long short-term memory network (LSTM) is an enhanced RNN architecture that alleviates the gradient problem in sequence processing by using forget gates, input and output gates. The forget gate discards unnecessary old information, the input gate records relevant new information, and the output gate determines how the current memory state is reflected in the output. These gates enable LSTMs to effectively retain long-term information and effectively solve prediction sequence problems.

[0088] Specifically, the LSTM network structure consists of the following four steps: updating the forget gate, the input gate, the cell state, and the output gate. The specific steps of the LSTM network structure are as follows:

[0089] Update the forget gate:

[0090]

[0091] Update the input gate:

[0092]

[0093] Update the cell state:

[0094]

[0095] Update the output gate:

[0096]

[0097]

[0098] wherein, , , and are the forget gate, the input gate, the cell state, and the output gate at time t , and are the corresponding weight coefficients and biases, is the hidden state at the previous time , is the input at time t , is a second activation function (Sigmoid activation function), Time hidden state.

[0099] In the embodiment, the limitation of the LSTM unit is that it only transmits information in a single direction, while the Bi-LSTM contains two independent LSTM layers, one processing the forward sequence (from the beginning to the end of the sequence) and the other processing the reverse sequence (from the end to the beginning of the sequence), while considering both forward and backward information, making it able to capture richer contextual information. In this embodiment, the Bi-LSTM is used as a time series modeling module to model the past and future time of the spatial features extracted by the convolutional neural network, to obtain the corresponding time series representation.

[0100] In the embodiment, the Bi-LSTM network is used to fully utilize past and future information and capture the dynamic changes of time series to improve the modeling ability of time dependence.

[0101] In the embodiment, a self-attention mechanism is introduced as an attention module to enhance the generalization ability of the marine environment data prediction model. The model can dynamically focus on important parts while processing sequences through the attention module. Specifically, the correlation of the time series is calculated through the attention module to obtain the predicted value of the time series representation. The loss function value of the marine environment data prediction model to be trained is determined according to the predicted value of the time series representation and the standard marine environment data. The marine environment data prediction model to be trained is iteratively updated according to the loss function value until the iteration termination condition is met, and the trained marine environment data prediction model is obtained.

[0102] In the embodiment, first, the multi-level features extracted by CNN and the time series modeling ability of BiLSTM effectively capture the spatial patterns and dynamic changes of marine environment data, improving the processing accuracy. Second, the introduction of the attention mechanism enables the model to automatically identify key features and time steps, enhancing the reliability and interpretability of the results to help researchers understand the main factors affecting the data. In addition, the model can integrate data from different sources, improving its adaptability to heterogeneous data and making the results more comprehensive and accurate. Finally, the self-learning and online adjustment capabilities of the model enable it to optimize over time and adapt to new data patterns, thus exhibiting stronger stability and effectiveness in long-term applications, meeting the practical needs of marine environment monitoring and research.

[0103] In some embodiments, the loss function of the marine environment data prediction model is the mean squared error function (MSE), which can be represented by the following formula:

[0104]

[0105] wherein, is a true value, is a true value, is a predicted value, n is a sample number.

[0106] The inputting the time series representation into the attention module in step S203 to make the attention module calculate the correlation of the time series to obtain the predicted value of the time series representation includes the following steps S301 to S303.

[0107] Step S301, inputting the time series representation into the similarity calculation module to calculate the similarity of each key in the time series to obtain the similarity result of the time series.

[0108] Step S302, inputting the similarity result of the time series into the attention score module to obtain the attention score of the time series representation.

[0109] Step S303, inputting the attention score into the weighted summation module to obtain the predicted value of the time series representation.

[0110] In the embodiment, the attention module includes a similarity calculation module, an attention score module and a weighted summation module.

[0111] In the embodiment, in the similarity calculation module, the output of the Bi-LSTM network is connected through a fully connected layer, that is, the time series representation of the output of the Bi-LSTM network is input into the fully connected layer to obtain the similarity result of the time series. And the attention score is calculated in the attention score module by the following formula:

[0112]

[0113] wherein, represents the attention score, , represents the weight and bias of the attention mechanism, represents the input of the attention mechanism.

[0114] In the embodiment, the calculated attention score is input into the weighted summation module, and the attention scores are weighted and summed to obtain the predicted value of the time series representation, that is, the predicted value output by the marine environment data prediction model.

[0115] Wherein, before the inputting the attention score into the weighted summation module in step S303 to obtain the predicted value of the time series representation, the method further includes the following steps S401 to S402.

[0116] Step S401, normalizing the attention score to obtain the processed attention score.

[0117] Step S402, the processed attention score is converted to obtain an attention weight by the following formula:

[0118]

[0119] wherein, t is a time step t , is an attention weight of a current time step t, is a current time step t processed attention score, T is a length of a time sequence.

[0120] The step S303 of inputting the attention score into the weighted summation module to obtain the predicted value represented by the time sequence further includes the following step S403.

[0121] Step S403, the attention weight is weighted summed by the following formula to obtain the predicted value represented by the time sequence:

[0122]

[0123] wherein, is a predicted value represented by a time sequence, is a hidden state of a time step t .

[0124] In the embodiment, the model can automatically identify and focus on key features and time steps through the attention module, which enhances the interpretability and accuracy of the prediction result.

[0125] The step S105 of correcting the to-be-corrected marine environment data according to the predicted marine environment data to obtain the corrected marine environment data includes steps S501 to S502.

[0126] Step S501, calculate the difference between the predicted marine environment data and the to-be-corrected marine environment data.

[0127] Step S502, correct the to-be-corrected marine environment data according to the difference to obtain the corrected marine environment data by the following formula:

[0128]

[0129]

[0130] wherein, is the corrected marine environment data, is the to-be-corrected marine environment data, is the predicted marine environment data, is the difference.

[0131] In the present embodiment, the main purpose of data correction is to adjust the original measured data using the results of model prediction to improve the accuracy of the data. Before data correction, the predicted values generated by the model and the actual measured values are collected; the actual measured values refer to the original marine environmental data (such as temperature, salinity, etc.) obtained by sensors or other measuring devices, i.e., the marine environmental data to be corrected; the predicted values are the predicted values of the data at the same time point generated by the marine environmental data prediction model; the correction values are calculated by the predicted values output by the marine environmental data prediction model to quantify the difference between the model prediction and the actual value.

[0132] In the present embodiment, the error between the model output and the actual data is calculated by the following formula:

[0133]

[0134] Error represents the difference between the predicted value and the actual measured value, a positive value indicates that the model underestimates the actual value, and a negative value indicates that the model overestimates the actual value; according to the calculated prediction error , the correction of the actual measured value. In the present embodiment, the correction value is equal to the error e .

[0135]

[0136] The original measured data is corrected using the calculated correction value, and the corrected data is obtained by the following formula:

[0137]

[0138] is directly represented by the predicted value as:

[0139]

[0140] In the present embodiment, by properly adjusting the measured values, the corrected marine environmental data can more accurately reflect the state of the marine environment.

[0141] In some embodiments, after generating the corrected data, the accuracy of the correction is quantitatively evaluated by the mean square error function (MSE).

[0142]

[0143] wherein, is the th real value, is the th corrected value, The total number of samples.

[0144] As shown in FIG. 2(a) and FIG. 2(b), for the temperature data in the marine environment data, the temperature correction before and after the use of the technical solution provided by the embodiment is obtained. FIG. 2(a) is the temperature data before correction, and FIG. 2(b) is the temperature data after correction.

[0145] In the embodiment, by preprocessing the original marine environment data, standard marine environment data is obtained, and a trained marine environment data prediction model is obtained based on the standard marine environment data. Through the feature extraction capability of the marine environment data prediction model, the time series modeling capability of the bidirectional long short-term memory network, and the automatic focusing characteristics of the attention mechanism, the accuracy and reliability of data processing are improved. The marine environment data to be corrected is input into the trained marine environment data prediction model to obtain predicted marine environment data. The marine environment data to be corrected is corrected by the predicted marine environment data to obtain corrected marine environment data. In this way, the accuracy and reliability of marine environment data correction are effectively improved through accurate error calculation and adjustment of model output.

[0146] As shown in FIG. 2(a) and FIG. 2(b), for the temperature data in the marine environment data, the temperature correction before and after the use of the technical solution provided by the embodiment is obtained. FIG. 2(a) is the temperature data before correction, and FIG. 2(b) is the temperature data after correction. Figure 3 Some embodiments of the present application provide a multi-source marine environment data correction device, as shown in the device includes an acquisition module 1100, a processing module 1200, a training module 1300, a prediction module 1400, and a correction module 1500. Specifically:

[0147] The acquisition module 1100 is configured to acquire original marine environment data.

[0148] The processing module 1200 is configured to preprocess the original marine environment data to obtain standard marine environment data.

[0149] The training module 1300 is configured to input the standard marine environment data into a marine environment data prediction model to be trained to obtain a trained marine environment data prediction model.

[0150] The prediction module 1400 is configured to input marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data.

[0151] The correction module 1500 is configured to correct the marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data.

[0152] It should be noted that the multi-source marine environment data correction device provided by the embodiment and the multi-source marine environment data correction method described above are based on the same inventive concept, and therefore the related content of the multi-source marine environment data correction method described above is also applicable to the content of the multi-source marine environment data correction device, and therefore, the details are not repeated here.

[0153] As Figure 4 The embodiment of the present application also provides an electronic device, and the electronic device comprises:

[0154] at least one hydrogen fuel cell;

[0155] at least one memory;

[0156] at least one processor;

[0157] at least one program;

[0158] The program is stored in the memory, and the processor executes the at least one program to implement the multi-source marine environment data correction method described above.

[0159] The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a vehicle-mounted computer, etc.

[0160] The electronic device of the embodiment of the present application will be described in detail below.

[0161] The processor 1600 can be implemented in the form of a general central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the embodiments of the present disclosure.

[0162] The memory 1700 can be implemented in the form of a read only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 1700 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present disclosure are implemented by software or firmware, the related program codes are saved in the memory 1700 and are called and executed by the processor 1600 to implement the multi-source marine environment data correction method.

[0163] The input / output interface 1800 is used to realize information input and output.

[0164] The communication interface 1900 is configured to realize the communication interaction between the device and other devices, and can realize the communication through a wired manner (for example, a USB, a network cable, or the like) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, or the like).

[0165] The bus 2000 is configured to transmit information between various components (for example, the processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900) of the device.

[0166] The processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900 are connected to each other through the bus 2000 to realize the communication connection between the device.

[0167] The disclosure also provides a storage medium, which is a computer readable storage medium, and stores computer executable instructions for causing a computer to execute the above-mentioned multi-source marine environment data correction method.

[0168] The memory is a non-transitory computer readable storage medium, and can be used to store a non-transitory software program and a non-transitory computer executable program. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, for example, at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and the remote memory can be connected to the processor through a network. Examples of the network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0169] The embodiments described in the disclosure are used to more clearly illustrate the technical solutions of the disclosure, and do not constitute a limitation on the technical solutions provided by the disclosure. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the disclosure are also applicable to similar technical problems.

[0170] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the disclosure, and can include more or fewer steps than the figures, or combine certain steps or different steps.

[0171] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, that is, can be located in one place, or can be distributed on multiple network units. According to actual needs, part or all of the modules can be selected to realize the purposes of the embodiments.

[0172] Those skilled in the art can understand that all or some of the steps in the method disclosed above, the functional modules / units in the system and the device can be implemented by software, firmware, hardware or a proper combination thereof.

[0173] The terms "first", "second", "third", "fourth" and the like in the description of the application and in the claims hereof, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so termed is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of orderly or chronological permutations. Moreover, the terms "comprise", "have" and any variations thereof are intended to cover a non-exclusive inclusion, for example, a process, method, article, or apparatus that comprises a list of steps or units can not necessarily be limited to those steps or units, but can include additional steps or units not expressly listed or inherent to such process, method, article, or apparatus.

[0174] It should be understood that, in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0175] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0176] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Part or all of the units may be selected according to actual needs to achieve the purpose of the embodiment.

[0177] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0178] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes multiple instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0179] The above is a specific description of the preferred implementation of the embodiments of the present application, but the embodiments of the present application are not limited to the above implementation. Those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the embodiments of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the embodiments of the present application.

[0180] The embodiments of the present application have been described in detail above in combination with the drawings, but the present application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the purpose of the present application.

Claims

1. A multi-source marine environmental data correction method, characterized in that, The method comprises: obtaining original marine environment data; preprocessing the original marine environment data to obtain standard marine environment data; inputting the standard marine environment data into a marine environment data prediction model to be trained to obtain a trained marine environment data prediction model; inputting marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data; correcting the marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data; the marine environment data prediction model to be trained comprises a feature extraction module, a time series modeling module and an attention module; the method comprises: inputting the standard marine environment data into the feature extraction module to enable the feature extraction module to extract spatial features of the standard marine environment data to obtain spatial features of the standard marine environment data; inputting the spatial features into the time series modeling module to enable the time series modeling module to model the spatial features in the past and the future to obtain time series representations; inputting the time series representations into the attention module to enable the attention module to calculate the correlation of the time series to obtain predicted values of the time series representations; determining a loss function value of the marine environment data prediction model to be trained according to the predicted values of the time series representations and the standard marine environment data; iteratively updating the marine environment data prediction model to be trained according to the loss function value until an iteration termination condition is met to obtain the trained marine environment data prediction model; the attention module comprises a similarity calculation module, an attention score module and a weighted summation module; the method comprises: inputting the time series representations into the similarity calculation module to calculate the similarity of each key in the time series to obtain a similarity result of the time series; inputting the similarity result of the time series into the attention score module to obtain an attention score of the time series representations; inputting the attention score into the weighted summation module to obtain the predicted values of the time series representations; the similarity calculation module connects the output of the Bi-LSTM network through a fully connected layer, and inputs the time series representations of the output of the Bi-LSTM network into the fully connected layer to obtain the similarity result of the time series.

2. The multi-source ocean environment data correction method of claim 1, wherein, the feature extraction module comprises a first convolutional layer and a second convolutional layer connected in sequence, the first convolutional layer comprises 64 filters, a convolution kernel with a size of 1 and a first activation function, and the second convolutional layer comprises 64 filters, a convolution kernel with a size of 3 and the first activation function.

3. The multi-source ocean environment data correction method of claim 1, wherein, Before the attention score is input into the weighted summation module to obtain the predicted value of the time series representation, the method further comprises: normalizing the attention score to obtain a processed attention score; the processed attention score is converted to obtain an attention weight by the following formula: wherein, t is the time step t , is the attention weight for the current time step t, is the current time step t processed attention score, T is the length of the time series; the attention score is input into the weighted summation module to obtain the predicted value of the time series representation, comprising: the attention weight is weighted and summed by the following formula to obtain the predicted value of the time series representation: wherein, is a prediction value represented as a time series, is a hidden state at time step t .

4. The multi-source ocean environment data correction method of claim 1, wherein, the predicted marine environment data is corrected according to the marine environment data to be corrected to obtain corrected marine environment data, comprising: calculating the difference between the predicted marine environment data and the marine environment data to be corrected; the marine environment data to be corrected is corrected according to the difference to obtain the corrected marine environment data by the following formula: wherein, is the revised marine environmental data, is the marine environmental data to be revised, is the predicted marine environmental data, is the difference.

5. The multi-source ocean environment data correction method of claim 1, wherein, the loss function of the marine environment data prediction model is: wherein, is a mean square error function, is a true value, is a predicted value, n is a number of samples.

6. A multi-source marine environment data correction device, characterized by, the device comprises: an acquisition module for acquiring original marine environment data; a processing module for preprocessing the original marine environment data to obtain standard marine environment data; a training module for inputting the standard marine environment data into a marine environment data prediction model to be trained to obtain a trained marine environment data prediction model; a prediction module for inputting marine environment data to be corrected into the trained marine environment data prediction model to obtain predicted marine environment data; a correction module for correcting marine environment data to be corrected according to the predicted marine environment data to obtain corrected marine environment data; the marine environment data prediction model to be trained comprises a feature extraction module, a time series modeling module and an attention module; the standard marine environment data is input into the marine environment data prediction model to be trained to obtain the trained marine environment data prediction model, comprising: the standard marine environment data is input into the feature extraction module to enable the feature extraction module to extract spatial features of the standard marine environment data to obtain spatial features of the standard marine environment data; the spatial features are input into the time series modeling module to enable the time series modeling module to model the past and future time of the spatial features to obtain a time series representation; the time series representation is input into the attention module to enable the attention module to calculate the correlation of the time series to obtain a predicted value of the time series representation; the loss function value of the marine environment data prediction model to be trained is determined according to the predicted value of the time series representation and the standard marine environment data; the marine environment data prediction model to be trained is iteratively updated according to the loss function value until the iteration termination condition is met to obtain the trained marine environment data prediction model; the attention module comprises a similarity calculation module, an attention score module and a weighted summation module; The time series representation is input into an attention module, so that the attention module calculates the correlation of the time series to obtain a predicted value of the time series representation, including: The time series representation is input into the similarity calculation module to calculate the similarity of each key in the time series to obtain a similarity result of the time series; The similarity result of the time series is input into the attention score module to obtain an attention score of the time series representation; The attention score is input into the weighted summation module to obtain a predicted value of the time series representation; The similarity calculation module connects the output of the Bi-LSTM network through a full connection layer, and inputs the time series representation of the output of the Bi-LSTM network into the full connection layer to obtain a similarity result of the time series.

7. An electronic device, comprising: The memory is connected in communication with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to perform the multi-source marine environmental data correction method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer readable storage medium stores computer executable instructions for causing a computer to perform the multi-source marine environmental data correction method of any one of claims 1 to 5. The computer readable storage medium stores computer executable instructions for causing a computer to perform the multi-source marine environmental data correction method of any one of claims 1 to 5.

Images