47 results about "Data separation" patented technology

A system for interactive searches based on user queries data and a plurality of Large Language Models (LLMs) including a processor of a Human-Machine Interface (HMI) server node configured to host a network of LLMs and at least one machine learning module (ML) and connected to at least one user-entity node over a network and a memory on which are stored machine-readable instructions that when executed by the processor, cause the processor to: receive a search request input data from the at least one user-entity node; evaluate, by a first dedicated LLM, relevance of the search request input data by discerning between primary and secondary information; responsive to evaluation by the first dedicated LLM, derive classifying features from the primary and secondary information and generate a feature vector based on the classifying features; ingest the feature vector into the ML module configured to extract additional search parameters from a predictive search model based on historical search data associated with the at least one user-entity node; dissect, by a second dedicated LLM, the search request input data and the additional search parameters to separate the data into qualitative and quantitative criteria elements based on the primary and the secondary information; transform, by a third dedicated LLM, the quantitative criteria elements into structured queries for a database; process, by a fourth dedicated LLM, the qualitative criteria elements by searching through a semi-structured data repository; and synthesize, by the fifth LLM, processed search findings into a succinct human-language summary.

The invention belongs to the field of tunnel engineering, and relates to a tunnel defect detection data spatial resolution enhancement method and device based on a Vine Copula dependent structure and a conditional random field, a computer readable medium and a computer program product. The method comprises the following steps: firstly, separating an overall trend and random fluctuation from original tunnel defect monitoring data; then establishing a Vine Copula multivariable dependence model to represent a statistical correlation relationship among the plurality of defect indexes; and then generating defect distribution data with high spatial resolution by using a conditional random field interpolation simulation technology in combination with a spatial autocorrelation analysis result. According to the invention, while the accuracy of the existing measuring point data is ensured, the spatial correlation structure and multivariable joint distribution characteristics of the defect data are maintained, and the precision of spatial interpolation prediction is improved.

The invention discloses a public institution energy consumption missing data estimation method based on an improved BPNN. The method comprises the steps that public institution energy consumption multi-feature data are acquired and preprocessed; performing feature screening; integrity check is carried out, and missing data and complete data are separated; constructing a missing data estimation model based on a BP neural network according to the complete data, performing network structure optimization by adopting a particle swarm optimization algorithm, and optimizing an initial weight threshold by adopting an enhanced chaos sparrow search algorithm; training the missing data estimation model; and inputting the missing data into a trained missing data estimation model, and estimating the missing value of the number of energy users. According to the method, the BP neural network architecture is constructed by fusing the particle swarm and the chaos sparrow collaborative optimization mechanism, reasonable estimation of the missing value of the number of energy users of the public institutions is realized, the limitation of a traditional optimization method is broken through, and the method has remarkable advantages in the aspects of parameter space search efficiency and global optimization ability.

The invention provides a bacterial colony heterogeneous data separation method and device, equipment and a storage medium. Relates to the technical field of microbial colony heterogeneous data separation. The method comprises the following steps: performing hierarchical clustering on microbe single-cell Raman spectrum data to identify outliers; based on the outliers, performing de-noising processing on the single-cell Raman spectrum data of the microorganisms to obtain de-noised data; performing spectral clustering based on the de-noised data to obtain a spectral clustering result; wherein the spectral clustering result is used for identifying or marking bacterial colony cells in different growth periods; and evaluating the spectral clustering result by using a contour coefficient and a CH index to determine an optimal clustering number, and performing spectral clustering on the de-noised data based on optimal clustering data to obtain a final clustering result. According to the method, outlier detection, pruning optimization, spectral clustering and quantitative evaluation are combined, so that the precision and stability of microbe heterogeneity analysis are remarkably improved.

The invention discloses a power grid voltage long-time-sequence dynamic prediction method and system. The method comprises the following steps: acquiring historical voltage effective value data and cleaning the historical voltage effective value data; separating the data into a trend component and a fluctuation component by using a signal decomposition algorithm; extracting overall and global time sequence features and local mutation features through a double-flow network; fusing the features by using an adaptive gating unit, inputting the features into a decoder, and outputting a voltage predicted value and a fluctuation interval; and generating a stability evaluation index based on the fluctuation interval. According to the method, through multi-scale feature decoupling and physical constraint loss function optimization, long-time evolution and instantaneous disturbance of the voltage can be accurately captured, the problems that a traditional model lags in sudden change working condition prediction and lacks physical consistency are solved, and a reliable basis is provided for power grid safety scheduling.

The invention discloses a vehicle braking distance prediction method and device, equipment and a storage medium, and relates to the technical field of vehicle driving safety, and the method comprises the steps: obtaining a road surface texture data set and a friction coefficient data set, inputting the road surface texture data set to a three-dimensional road surface reconstruction model, and carrying out data separation, data noise reduction and data reconstruction. And inputting the friction coefficient data set and the reconstructed pavement texture data into an anti-skid performance prediction model to obtain a prediction result of the pavement friction coefficient, and based on the reconstructed pavement texture data and the predicted pavement friction coefficient, establishing a three-dimensional tire water skiing model and carrying out fluid-solid coupling simulation to obtain a tire-road adhesion characteristic curve of the wet and slippery pavement. And inputting the tire road adhesion characteristic curve into the vehicle braking simulation model, and calculating the vehicle braking distance on the wet and slippery road surface through dynamic simulation, so that the problem that the actual tire road adhesion characteristic change is ignored in a traditional braking distance formula is solved, and high-precision prediction of the braking distance in a complex scene in a rainy day is realized.

The application discloses a power grid voltage long-time sequence dynamic prediction method and system, and the method comprises the following steps: obtaining historical voltage effective value data and cleaning; using a signal decomposition algorithm to separate the data into a trend component and a fluctuation component; extracting global long-time sequence features and local mutation features through a double-flow network; using an adaptive gating unit to fuse the features and inputting the decoder to output voltage prediction values and fluctuation intervals; and generating a stability evaluation index based on the fluctuation interval. Through multi-scale feature decoupling and physical constraint loss function optimization, the application can accurately capture the long-time evolution and instantaneous disturbance of the voltage, solve the problems of lagging prediction of traditional models for mutation working conditions and lack of physical consistency, and provide a reliable basis for power grid safety scheduling.

The invention discloses a DAS-VSP four-wave-field separation method and system based on deep learning, and relates to the field of data processing, and the method specifically comprises the steps: obtaining VSP data, and separating the VSP data into multiple types of wave fields; superposing the multiple types of wave fields to generate a mixed wave field, and generating a training set based on the mixed wave field and the multiple types of wave fields; constructing an initial wave field separation model based on a U-Net network, and performing iterative training on the initial wave field separation model by using the training set to generate a final wave field separation model; and obtaining to-be-processed VSP data, inputting the to-be-processed VSP data into the wave field separation model, and outputting prediction results of various wave fields. According to the method, high-precision separation can be guaranteed, and meanwhile, rapid and efficient wave field treatment is achieved.

The application provides a pipeline system-based pre-breakage leakage automatic analysis method and system. The pipeline system-based pre-breakage leakage automatic analysis method can realize accurate identification of the safety state of all positions of a pipeline, constructs a whole-process automatic pre-breakage leakage analysis system from three-dimensional model input to criterion determination, effectively solves the technical pain points of data separation, multiple manual intervention, redundant calculation, low determination efficiency and insufficient accuracy in traditional analysis, and greatly improves the automation, intelligent level and result reliability of pipeline pre-breakage leakage analysis.

An embodiment of the present application provides a data processing method and device for Kafka storage and computing separation. By constructing a multi-layer data processing architecture, including four levels: network communication, task distribution, processing scheduling, and storage. Through the hierarchical design of the write-ahead log area and the main storage area, hot and cold data separation is achieved. Combining the replica manager, coordinator, and load predictor to realize the intelligent scheduling of system resources. The adaptive compression algorithm and verification mechanism are adopted to ensure the data transmission efficiency and integrity. A message buffer is created to store hot data, and flexible access to hot and cold data is realized based on the message position. This method effectively solves the deficiencies of traditional technologies in aspects such as storage and computing coupling, storage efficiency, and load balancing, and significantly improves the performance and scalability of the message queue system.

A system for interactive searches based on user queries data and a plurality of Large Language Models (LLMs) including a processor of a Human-Machine Interface (HMI) server node configured to host a network of LLMs and at least one machine learning module (ML) and connected to at least one user-entity node over a network and a memory on which are stored machine-readable instructions that when executed by the processor, cause the processor to: receive a search request input data from the at least one user-entity node; evaluate, by a first dedicated LLM, relevance of the search request input data by discerning between primary and secondary information; responsive to evaluation by the first dedicated LLM, derive classifying features from the primary and secondary information and generate a feature vector based on the classifying features; ingest the feature vector into the ML module configured to extract additional search parameters from a predictive search model based on historical search data associated with the at least one user-entity node; dissect, by a second dedicated LLM, the search request input data and the additional search parameters to separate the data into qualitative and quantitative criteria elements based on the primary and the secondary information; transform, by a third dedicated LLM, the quantitative criteria elements into structured queries for a database; process, by a fourth dedicated LLM, the qualitative criteria elements by searching through a semi-structured data repository; and synthesize, by the fifth LLM, processed search findings into a succinct human-language summary.

A method for separating load effects on a dedicated track bridge comprises the following specific steps: 1. Identifying the train passing phase; 2. Using a nonlinear model to separate train load and temperature load effect data; 3. Using an adaptive algorithm with a memory effect to remove temperature load effect data; 4. Using the multi-scale singular value decomposition (MS-SVD) method to process monitoring data; 5. Using a joint distribution model based on probability theory to optimally estimate the separation of temperature and train load effect data; and 6. Describing the temporal dynamics of train load and temperature load using a state transition matrix. The present invention improves the accuracy of monitoring data: Through adaptive data separation technology, the present invention effectively distinguishes between train load effects and temperature load effects in bridge health monitoring data, avoiding interference from slowly changing factors such as degradation effects, and significantly improving the accuracy of bridge structure health assessment results.

The invention discloses a storage chip aging prediction method and device based on topological sorting and a medium, and relates to the technical field of chip aging tests.The method comprises the steps that an initial static topological graph is constructed based on a predefined internal dependency relationship of a storage chip; collecting performance monitoring data of the storage chip, performing time sequence relevance identification, and generating a dynamic dependency edge; fusing the dynamic dependent edge with the initial static topological graph to generate a mixed topological graph; decomposing the mixed topological graph through a graph community division algorithm to generate a plurality of sub-graph communities; performing topological sorting on each sub-graph community to generate a functional component test sequence of each sub-graph community; and respectively executing an aging test task on each sub-graph community and the corresponding functional component test sequence through a plurality of distributed test nodes to generate a plurality of pieces of performance parameter attenuation data. According to the method, the performance monitoring data and the static topology are fused, the hybrid topological graph of the chip is constructed, and the problem of relevance deficiency caused by data separation is solved.

The invention relates to a house type image size measurement method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a to-be-identified house type image; performing target detection on the to-be-identified house type image to obtain a target detection result; performing data separation according to the target detection result to obtain a data separation result; according to a data separation result, carrying out independent scale value calculation to obtain scale value data; and performing size batch calculation on the to-be-detected furniture of the target detection result according to the scale value data to obtain a size measurement result. By adopting the method, high accuracy and high robustness of house type image size measurement can be ensured by detecting and fusing information of a plurality of measuring scales, and full-automatic and high-efficiency house type image size measurement is realized.

To reduce a storage capacity required for computation of a convolutional neural network while suppressing a decline in reliability of a computation result.SOLUTION: An information processing apparatus includes: acquisition means that acquires target data from data for inference input in the information processing apparatus; and computation means that executes convolution computation using computation data including the target data acquired by the acquisition means and margin data different from the target data that is required to obtain computation result data in a predetermined size, and outputs computation result data. According to a preset data acquisition condition, the acquisition means acquires first data, which is part of the margin data, from a data group existing around the target data in the data for inference, and does not acquire second data, which is the margin data except the first data, from the data group. A coefficient of the convolution computation is determined by learning to which the data acquisition condition is reflected.SELECTED DRAWING: Figure 19

The present application relates to the technical field of heat-conducting material detection, and discloses a heat-conducting gel water resistance detection method and system based on a sensor. The method comprises receiving real-time monitoring data of a heat-conducting gel in a water resistance test environment collected by a spectrum sensor and a thermal imaging sensor; dividing samples into multiple independent detection sub-regions, comparing the real-time monitoring data with historical database reference data to separate stable components and variable components; using pattern classification technology to identify unexpected change patterns, using relationship mining technology to find causal correlations and co-occurrence characteristics, and obtaining key influence factors; taking the key influence factors as input, combining with similar material standard water resistance performance data to construct a reference performance curve, determining a potential water resistance risk level according to the curve deviation degree and an intelligent evaluation framework; and using a probability inference model to predict water resistance changes within a preset time window and generating a detection result report.

The invention discloses a wind turbine generator yaw error prediction control method and system based on deep learning, and the method comprises the steps: firstly processing multi-dimensional original sensor data through a self-adaptive denoising module, separating out a pure physical signal, and obtaining an instantaneous difference signal; furthermore, the difference signal is modeled in combination with the operation context of the unit, and the expected noise characteristic under the current working condition is predicted, so that the attributive noise level estimation is generated, and the normal operation fluctuation and the abnormal interference are effectively distinguished. On the basis, de-noising input and noise level estimation are cooperatively input into a Bayesian prediction model, prediction yaw error distribution containing a mean value sequence and a variance sequence is output, and explicit quantization of prediction result uncertainty is achieved. Finally, control risk assessment is performed based on the distribution, and adaptive control parameters are dynamically synthesized to optimize yaw action, so that the accuracy and safety of a control decision in an uncertain environment are ensured while data noise interference is suppressed.

The present invention provides a data separation method and a separation system, wherein the data separation method comprises the following steps: obtaining a full-cycle processing curve during workpiece processing, wherein the full-cycle processing curve comprises a plurality of single-cycle processing curves; selecting one of the single-cycle processing curves in the full-cycle processing curve as a template curve, and the other single-cycle processing curves as processing curves to be matched; determining a standard sample curve in the template curve; judging whether the standard sample curve matches each processing curve to be matched; judging whether to update the standard sample curve based on the matching results between n adjacent processing curves to be matched that match the standard sample curve and the standard sample curve; and separating the template curve before the update and all processing curves to be matched that match the standard sample curve in the template curve.

Computer-implemented method for generating a process model (601) for use in the analysis of the operation of a process (300) which can be operated in a number of different process stages as defined by a stage variable associated with the process (300), comprising: Acquiring training data from the process (300) during the operation of the process (300), wherein the training data include a value for each of a set of process parameters, a value for the stage variable, and a value of an outcome variable associated with each of a variety of different process measurement times; Dividing the training data into time fractions of the data using a computer processing device (102, 13) to produce a set of time fractional data for each time fraction of the data, wherein each set of time fractional data includes a value for each of the set of process parameters, a value for the stage variable and a value for the result variable; Storing sets of time-proportional data in a computer memory; using a computer processing device (102, 13) determine a set of process stage means from the training data, wherein the set of process stage means includes a stage variable mean for each of the process stages and one or more process parameter means for each of the process stages; Storing the sets of process stage resources in a computer memory; using a computer processing device (102, 13) determining a set of time-part means for each of the time parts of the data using the stored process stage means, wherein each set of time-part means includes a time-part mean for each of the process parameters; using a computer processing device (102, 13) developing a set of deviations from the mean for each time fraction of the data, wherein the set of deviations from the mean for a given time fraction of the data for each process parameter within the given time fraction of the data includes the use of the process parameter value of the given time fraction of the data and the time fraction mean for the process parameter for the given time fraction of the data to develop the deviation from the mean for the process parameter for the given time fraction of the data; and using a computer processing device (102, 13) generating a process model (601) using the sets of deviations from the mean for the time components of the data and the result variable values ​​for the time components of the data.

The invention relates to the technical field of environmental monitoring, in particular to a real-time marine environmental element monitoring system, which comprises an acquisition unit, a construction unit, a processing unit and a separation unit, and is characterized in that the acquisition unit adapts acquisition frequency according to marine depth, combines threshold triggering and external instructions, and acquires environmental data, node parameters and target signals in real time; the construction unit simulates an interference signal refraction propagation path through ray acoustics based on the data, extracts signal features, and fuses multi-dimensional data to train an interference prediction model; the processing unit inputs related data into a model and outputs a precise interference result; and the separation unit is combined with the interference result, and the interference signal is stripped from the mixed signal through frequency domain conversion, dynamic filtering, time domain reconstruction and verification. According to the method, accurate prediction and separation of interference signals are realized, the real-time performance, accuracy and reliability of marine environment element monitoring are remarkably improved, and powerful technical support is provided for marine resource development.

The application discloses an intelligent data access and integration system based on a dynamic expansion architecture, comprising: a data source automatic identification module, which automatically detects and identifies various data sources and their characteristics; a data adaptation module, which adapts according to the characteristics of the data sources; an intelligent data quality management module, which predicts and automatically corrects data quality problems based on a self-learning mechanism; a data preprocessing module, which is used for cleaning and converting original data; a data integration module, which adjusts the data processing flow according to real-time load through a dynamic expansion architecture and a load balancing mechanism; a data storage module, which supports various database types and realizes cold and hot data separation storage; an API management module, which develops and maintains data access interfaces; and a monitoring and optimization module, which monitors system performance and data flow in real time and guarantees stable and efficient operation of the system. The application improves data access efficiency, quality and utilization of system resources and ensures system stability.

This invention discloses a method and system for isolating microgrids using a grid-connected energy storage converter, belonging to the field of microgrid state identification technology. The method includes: collecting and cleaning multi-source monitoring data of the microgrid; using a disturbance baseline separation algorithm based on dynamic time window integration and threshold filtering to separate the data into steady-state and disturbance components; inputting the two components into a pre-trained state identification deep learning model, which is processed sequentially through a disturbance feature decoupling module, a dual-channel spatiotemporal feature fusion module, and a spatiotemporal feature collaborative extraction module to extract spatial feature vectors containing spatiotemporal correlations; and finally, calculating the state probability distribution using an adaptive state classification head based on Mahalanobis distance to determine the operating state. This invention significantly improves the identification accuracy and stability under noisy, transient, and transitional states by effectively separating disturbances, decoupling features, and collaboratively modeling spatiotemporal information.

The application discloses a gravity data separation method, device and equipment for a metal mining area and a medium, and relates to the technical field of gravity anomaly data processing. The method comprises the following steps: measuring gravity data of each measuring point in a target area, and preprocessing to obtain Bouguer gravity anomaly data of each measuring point in the target area; calculating the radial logarithmic power spectrum of the Bouguer gravity anomaly data, dividing the radial logarithmic power spectrum into multiple characteristic frequency bands, and extracting the Bouguer gravity anomaly data of each characteristic frequency band; and performing physical property inversion calculation based on the Bouguer gravity anomaly data of each characteristic frequency band to obtain the underground space density distribution result of the target area. In the constraint iteration process, the inversion result is approximated to the prior information, the gravity potential field inversion result of the metal mining area with clear prior information is strictly constrained in the minimum interval of the known prior information, and therefore the accuracy of the underground ore body position inversion is improved.

The present invention discloses a dynamic monitoring method and platform for fire safety risks, which relate to the field of risk monitoring technology. The method includes: dividing the target spatial area and outputting first and second spatial areas; performing data separation monitoring on the first and second areas and outputting first and second monitoring data sets; constructing a risk assessment model, including a first and a second assessment model; the first assessment model receives the first data set for assessment and outputs a first risk indicator; the second assessment model receives the second data set for assessment and outputs a second risk indicator; and providing risk reminders for the target area based on the risk indicators, so as to solve the technical problem that traditional fire safety risk monitoring methods cannot realize regional division and data separation monitoring, resulting in insufficient monitoring accuracy, and achieve the technical effect of accurately identifying fire safety risks in the area by combining separate assessment models to analyze different types of data respectively, thereby improving monitoring accuracy and the accuracy of risk assessment.

The application belongs to the cross field of artificial intelligence and physiological signal and medical image processing fields, and provides a brain electrical signal data signal analysis method based on multi-modal deep learning. The purpose is to solve the problem that the existing scheme still needs to set the extracted signal features manually and cannot completely cover all effective features of west syndrome. The main scheme includes extracting original brain electrical signal data and original muscle electrical signal data to obtain binary data; data separation obtains multi-channel time-frequency domain data and divides training set data, validation set data and test set data; the neural network model of deep learning is set, and the best model parameter is obtained through training; then the data is input for classification, the abnormal brain electrical signal data and muscle electrical signal data in each time period are obtained, and the annotation is saved as an annotation file.

The invention relates to the technical field of heat-conducting material detection, and discloses a sensor-based heat-conducting gel water resistance detection method and a sensor-based heat-conducting gel water resistance detection system. The method comprises the following steps: receiving real-time monitoring data, acquired by a spectrum sensor and a thermal imaging sensor, of the heat-conducting gel in a water resistance test environment; dividing a sample into a plurality of independent detection sub-regions, comparing real-time monitoring data with historical database reference data, and separating stable components from variable components; identifying an unexpected change mode by using a mode classification technology, and carrying out causal association and co-occurrence feature discovery on the unexpected change mode by using a relation mining technology to obtain a key influence factor; constructing a reference performance curve by taking the key influence factor as input and combining standard water resistance data of similar materials, and determining a potential water resistance risk level according to a curve deviation degree and an intelligent evaluation framework; and predicting the water resistance change in a preset time window by using the probability inference model, and generating a detection result report.

This application discloses a vehicle braking distance prediction method, apparatus, device, and storage medium, relating to the field of vehicle driving safety technology. The method includes: acquiring a road surface texture dataset and a friction coefficient dataset; inputting the road surface texture dataset into a three-dimensional road surface reconstruction model; performing data separation, data denoising, and data reconstruction; inputting the friction coefficient dataset and the reconstructed road surface texture data into an anti-skid performance prediction model to obtain a predicted result for the road surface friction coefficient; based on the reconstructed road surface texture data and the predicted road surface friction coefficient, establishing a three-dimensional tire hydroplaning model and performing fluid-structure interaction simulation to obtain the tire-road adhesion characteristic curve of the wet road surface; inputting the tire-road adhesion characteristic curve into a vehicle braking simulation model; and calculating the vehicle braking distance on the wet road surface through dynamic simulation. This solves the problem that traditional braking distance formulas ignore actual tire-road adhesion characteristic variations, achieving high-precision prediction of braking distance in complex rainy scenarios.

The invention provides a model construction method, an industrial decision prediction method and device and electronic equipment, and relates to the technical field of communication. The method comprises the following steps: constructing an industrial field knowledge graph; the industrial field knowledge graph defines industrial field rules according to standardized symbols; constructing a feature extractor; the feature extractor is used for performing data processing and feature extraction on the industrial data; performing joint optimization on the industrial field knowledge graph and the feature extractor through differentiable logic to obtain an industrial model; the industrial model is used for outputting an industrial decision prediction result corresponding to the industrial data. According to the scheme of the invention, the problems of rule and data separation and lack of interpretability of an output decision prediction result in a current industrial large model can be solved.

The invention discloses a gravity data separation method, device and equipment for a metal mine area and a medium, and relates to the technical field of gravity anomaly data processing.The method comprises the steps that gravity data of each measurement point in a target area is obtained through measurement, and Bouguer gravity anomaly data of each measurement point in the target area is obtained through preprocessing; calculating a radial logarithmic power spectrum of the Bouguer gravity anomaly data, dividing the radial logarithmic power spectrum into a plurality of characteristic frequency bands, and respectively extracting the Bouguer gravity anomaly data of each characteristic frequency band; and performing physical property inversion calculation based on the Bouguer gravity anomaly data of each characteristic frequency band to obtain an underground space density distribution result of the target area. According to the method, the inversion result approaches to the prior information in the constraint iteration process, and the gravity potential field inversion result, with the clear prior information, of the metal mine area is strictly constrained in the minimum interval with the known prior information, so that the accuracy of underground ore body position inversion is improved.

This application relates to the field of navigation planning technology, and particularly to a method and system for rapid navigation path planning in stochastic flow fields. The method includes: based on flow field ensemble forecast data, separating the deterministic and uncertain components of the stochastic flow field, constructing a median flow field and a perturbation flow field set for the stochastic flow field; using the median flow field to characterize the deterministic component of the stochastic flow field, and using the perturbation flow field set to characterize the uncertain component of the stochastic flow field; proposing a conservative adjustment strategy for navigation speed based on the median flow field and the perturbation flow field set; under the condition of conservative navigation speed, using an improved level set equation to evolve the reachable frontier clockwise from the navigation starting point until the reachable frontier reaches the navigation target point and stops evolving; after reaching the target point, performing a counter-clockwise backtracking of the reachable frontier to calculate the time-optimal trajectory under the conservative navigation speed adjustment strategy, and using this time-optimal trajectory as the navigation planning path in the stochastic flow field.