201 results about "Error correction model" patented technology

The invention provides a VDSL ultra-low time delay communication method and system, and relates to the technical field of communication, and the method comprises the steps: encrypting a clock synchronization channel through a quantum key distribution protocol, dynamically dividing micro time slot resources in an orthogonal frequency division multiplexing symbol period, and generating a dynamically adjusted micro time slot resource distribution result; calculating an optimal phase offset matrix of the metasurface intelligent reflecting surface through a depth deterministic strategy gradient algorithm to obtain an optimized electromagnetic wave propagation path; generating a global optimization check matrix by aggregating the locally trained lightweight error correction model gradient of each node to obtain a compensated data stream; and constructing a causal graph model dynamic pruning high-entropy path to minimize causal entropy, through multi-agent reinforcement learning, taking time delay-energy efficiency as a game target to decide an optimal modulation order and a subcarrier switching strategy, and obtaining an optimized stable communication link. According to the invention, high-reliability and low-delay communication basic support is provided for high-precision intelligent manufacturing.

The invention discloses a self-adaptive InSAR-GNSS high-precision three-dimensional deformation resolving method with regularization introduced twice, and the method comprises the steps: obtaining InSAR deformation data and GNSS three-dimensional deformation data of a to-be-monitored region, and building an InSAR error correction model considering spatial correlation based on the InSAR deformation data and the GNSS three-dimensional deformation data; performing system error correction on the InSAR deformation data by using the corrected error correction model to obtain preliminary correction data; grouping the preliminary correction data according to data precision characteristics and determining an initial weight; iteratively calculating until unit weight variances of each group of observation data are approximately equal to obtain a final weight; and obtaining a high-precision earth surface three-dimensional deformation rate based on the final weight. According to the method, Tikhonov regularization is introduced twice, the weight is dynamically adjusted in combination with the IGGIII equivalent weight function, and various error influences are effectively reduced.

The invention provides an optical-electric hybrid ultra-low time delay communication method and system, and relates to the technical field of communication, and the method comprises the steps: outputting a global synchronous clock signal with clock jitter less than 1ppm, dynamically dividing micro time slot resources in an orthogonal frequency division multiplexing symbol period, and generating a dynamically adjusted micro time slot resource distribution result; calculating an optimal phase offset matrix of the metasurface intelligent reflecting surface through a depth deterministic strategy gradient algorithm to obtain an optimized electromagnetic wave propagation path; generating a global optimization check matrix by aggregating the locally trained lightweight error correction model gradient of each node to obtain a compensated data stream; and constructing a causal graph model dynamic pruning high-entropy path to minimize causal entropy, through multi-agent reinforcement learning, taking time delay-energy efficiency as a game target to decide an optimal modulation order and a subcarrier switching strategy, and obtaining an optimized stable communication link. According to the invention, high-reliability and low-delay communication basic support is provided for high-precision intelligent manufacturing.

The invention belongs to the technical field of ocean surface flow velocity remote sensing measurement, and particularly relates to a Luoga No.2 SAR (Synthetic Aperture Radar) flow velocity correction method based on Bragg orbit velocity, which comprises the following steps of: verifying the physical correctness of an orbit velocity mechanism relative to a phase velocity mechanism; under a DopRIM framework, a physically consistent sea wave error correction model is established; calibrating parameters of the sea wave error correction model; applying the calibrated sea wave error correction model to along-orbit interference SAR observation data of the Luoma No. 2 satellite, and calculating the radial surface flow velocity and sea wave error of each resolution unit; comparing an output result with a matched reference truth value to evaluate the effectiveness of the sea wave error correction model; comparing the sea wave error correction model with a KaDOP model, and respectively counting and analyzing simulated sea wave errors under different methods; the sea wave error simulation result is visually displayed, and the improvement effect is quantitatively evaluated through in-situ observation data and a KaDOP simulation result.

The invention discloses an expressway vehicle centimeter-level trajectory tracking system based on 5G-U Beidou fusion, and relates to the technical field of 5G-Beidou traffic trajectory tracking, and the system comprises a multi-source heterogeneous data collection module which collects Beidou positioning, 5G vehicle state and visual data; the multi-frequency multi-constellation fusion positioning module is used for improving the algorithm and improving the ionosphere correction precision; the dynamic environment perception and error correction module is used for constructing a dynamic electronic fence and optimizing an error correction model; the centimeter-level track generation and visualization module is used for improving an interpolation algorithm and smoothing a track; and the cloud collaboration and edge computing module adopts federal learning to realize model collaboration optimization, and also relates to a plurality of innovation modules such as safety guarantee and intelligent service. The system integrates 5G-U and Beidou technologies, realizes centimeter-level high-precision positioning, can dynamically sense the environment and correct errors, improves the data transmission processing efficiency and safety, integrates semantic enhancement, intelligent service and visualization functions, and assists the intelligent development of expressways.

The invention belongs to the technical field of power distribution networks, and discloses a load prediction and early warning method and system for a transformer area containing distributed resources, and the method comprises the steps: combining a two-dimensional time sequence data sample set of a high-risk transformer area with a transformer area feature operation data set according to the transformer area and a timestamp, and generating a prediction model training sample data set; building a lightweight gradient boosting tree as a main prediction model, inputting a training sample data set for training, and optimizing hyper-parameters of the main prediction model by adopting a Bayesian optimization algorithm; and establishing a residual error correction model based on local weighted Gaussian process regression, superposing a load prediction result of a prediction day of the main prediction model of the to-be-predicted transformer area with a residual error correction value of a prediction day of the residual error correction model to obtain a final load prediction result, and outputting transformer area weight / overload early warning information. According to the method, the LGBM is adopted as the main prediction model for load prediction, the residual error correction model is adopted for residual error correction, and the robustness and adaptability of the model are improved.

The invention discloses a low-drift laser scanning control method and system based on dynamic error compensation, relates to the technical field of laser scanning control, and aims to solve the drift problem in the laser scanning process and improve the scanning precision. The system comprises a system initialization and calibration module, a data acquisition and storage module, an error modeling module, a route planning and correction module and a real-time compensation and closed-loop optimization module, wherein the error modeling module and the route planning and correction module are key modules. The error modeling module comprises a key route screening unit and a data preprocessing unit, and is responsible for constructing a space-time joint error correction model; the route planning and correction module comprises an initial route generation unit, an error prediction unit and the like, and dynamic correction of a pre-planned route is achieved. According to the system, real-time compensation and closed-loop optimization are achieved through cooperation of all the modules in combination with a dynamic error compensation algorithm, scanning drift is effectively reduced, low-drift and high-precision laser scanning is guaranteed, and meanwhile follow-up scanning precision is optimized.

The invention relates to the field of battery thermal management, in particular to an AUV lithium ion battery thermal state prediction method. Comprising the following steps: constructing an electrothermal coupling reduced-order thermal model, and generating initial temperature estimation with physical consistency; a physical guidance space-time dynamic graph convolutional network PG-STDGCN is constructed as an error correction model, the model constructs a static and dynamic fused adjacency matrix by embedding physical priori such as a battery topological structure and circuit characteristics into dynamic graph learning, and a correction value of initial temperature estimation is output; and adding the initial temperature estimation and the correction value to obtain a final battery thermal state prediction result. According to the method, organic fusion from physical modeling to data-driven correction is realized, interpretability, precision and adaptability are considered under the dynamic working condition of the AUV, and the battery pack-level multi-cell temperature prediction performance is remarkably improved.

The invention discloses an electric energy metering error correction method and system for a charging facility under multiple working conditions. The method comprises the following steps: screening features and constructing cross features according to correlation coefficients and mutual information of various historical working condition parameter data; establishing an error correction model according to the cross characteristics and the error parameter data, and updating and optimizing the error correction model through data collected in real time in a plurality of set fusion time recording periods and historical working condition parameter data; measuring the electric energy metering value in real time, predicting an error correction value through the updated and optimized error correction, and correcting the measured electric energy metering value according to the predicted error correction value; and setting a monitoring period, and calculating a working condition health index value according to the predicted error correction value in the period to judge whether the working condition is abnormal. According to the invention, through real-time error correction, model dynamic updating and working condition health monitoring, the electric energy metering precision of the charging facility under a complex working condition and the stability of system operation are improved.

The invention provides a self-error-correction retrieval enhancement agent system. Firstly, error detection is carried out on RAG generation content through a trained retrieval enhancement error correction model, then error features are input into an agent model to carry out automatic planning of a solution, and finally a closed-loop process of error correction is completed through an execution engine. In the aspect of the implementation method, the scheme constructs a three-stage data mining normal form including error response sampling, key feature labeling and error type labeling, and the error recognition precision of the error correction model is effectively improved. Furthermore, a'generation-detection-correction-evaluation 'closed-loop optimization loop is formed, and the effect that the RAG capability is autonomously enhanced by utilizing the evaluation-guided proxy workflow under the condition that the RAG system does not need manual intervention is achieved.

The invention relates to a real-time process monitoring method based on data analysis, and the method comprises the following steps: S1, building a three-dimensional component priority evaluation model based on a component operation scene type, a real-time resource occupancy rate and a data dependence degree, and dynamically matching an acquisition strategy; s2, a cleaning rule is adapted according to a data source, a feature extraction dimension is adjusted in combination with process dynamic features, and an improved time sequence decomposition algorithm is used for separating data trends, fluctuations and abnormal residual errors; s3, constructing an exclusive baseline sub-model by using an online learning algorithm according to a scene label, and establishing a scene switching mechanism; s4, in combination with component interaction anomaly features, an anomaly level is judged through a mixed detection model; s5, on the basis of exception processing and user feedback, constructing an error correction model optimization parameter; and S6, generating a report containing an abnormal propagation path, and triggering hierarchical collaborative response of the associated component. The invention aims to solve the problems of single acquisition dimension, no scene adaptability in preprocessing, incomplete abnormal detection and the like in the existing monitoring technology.

The invention provides an intelligent tailing dam displacement prediction and early warning method, and belongs to the technical field of safety prediction and early warning, and the method comprises the steps: obtaining online monitoring historical data of a tailing dam, and carrying out the preprocessing of the data; constructing a GRU main prediction model to predict main prediction displacement; obtaining an error sequence through the measured data and the main prediction displacement; decomposing the error sequence into a trend term error sequence and a noise term error sequence through PSO-VMD-TOPSIS, and respectively constructing a trend term error correction model and a noise term error correction model based on GRU; an error correction value and a standard deviation of a trend item and a noise item are obtained through an MC-dropout technology; a dynamic weight calculation mechanism is constructed, a final displacement prediction value and a confidence interval are obtained in combination with the main prediction displacement, and whether an alarm is given or not is judged; according to the invention, high-precision real-time prediction, uncertainty quantification and real-time early warning of the displacement of the tailing dam are realized, and intelligent decision support is provided for safety state evaluation and disaster early warning of the tailing dam.

The invention discloses a power transmission line foundation pit three-dimensional laser radar equipment error self-correction control method, and relates to the technical field of power transmission line construction, and the method comprises the following steps: firstly, defining the distance from three-dimensional laser radar equipment to a target point, a horizontal included angle and a vertical included angle based on a Cartesian coordinate system; determining an initial coordinate of the target point; secondly, selecting a reference control point, removing noise and gross error points, correcting offset in the vertical direction, and finely processing point cloud data; thirdly, defining errors in the X-axis direction, the Y-axis direction and the Z-axis direction, and establishing an error correction model; and finally, based on the observation value and an error correction model, carrying out self-correction on the target coordinate, and carrying out correction through an environment feature self-adaptive error weight distribution mechanism. According to the method, closed-loop iterative optimization is performed through a multi-level data verification mechanism, the measurement precision is ensured, the operation efficiency is improved, the environmental adaptability is enhanced, and the method is suitable for power transmission line foundation pit division operation under complex terrains.

The invention discloses a language disease error correction model fusion method based on a weighted voting method, and relates to the field of model fusion technology and natural language processing, and the method comprises the following steps: 1, obtaining language disease error correction results of an original text in different models; taking the error correction suggestion with the highest weighted score of the corresponding model as the final error correction suggestion of the error point; summarizing to obtain a first-stage fusion result; in the second stage, language disease detection results of the original text in different models are obtained, and the language disease detection results and the fusion result in the first stage are used for screening; and obtaining an intersection fusion result and a union fusion result. And the user can obtain an error correction result meeting the requirement based on the fusion result. According to the language disease error correction model fusion method based on the weighted voting method, the advantages of all the models are integrated through a reasonable fusion strategy, and the error correction effect of the error correction model on the Chinese text is improved.

The invention discloses a VerilogA code automatic error correction method based on a classification error library and a large language model, and the method comprises the steps: constructing a classification system of the error types of VerilogA codes, starting from an MTJ device, carrying out the statistics of 84 common errors in the VerilogA codes of MTJ based on the system, and developing an automatic training sample generation system according to the 84 common errors; the method comprises the following steps: selecting a DeepSeek-R1-7B model as a basic model through a benchmark test, and carrying out fine tuning training by utilizing a classified error data set generated by scripting to form a special error correction model for the field. The processing process comprises the steps that VerilogA codes to be corrected are received, error types and reasons are positioned through a trained model, correction codes are generated, and verification of eight types of typical circuits is passed. An error classification-model training-simulation verification system is established, verification and error correction of VerilogA codes are achieved, and time and resources in the development process are saved.

The invention discloses a power grid safety margin prediction and risk assessment method and system based on a double-layer LSTM-XGBoost model, and the method comprises the steps: collecting the operation data of a power grid in different scales, and constructing a day degree data set and a time degree data set; performing standardization processing on the day degree data set and the time degree data set, and remodeling the day degree data set and the time degree data set into an LSTM input format to obtain a day degree input sequence and a time degree input sequence; respectively training a day degree prediction model and a time degree prediction model based on the double-layer LSTM network, and outputting a day degree safety margin prediction value and a time degree safety margin prediction value; an XGBoost residual error correction model is trained, the residual error of the day degree and hour degree prediction model is corrected, and a combined prediction result is generated; and comparing the hour prediction value of the next 24 hours with the day prediction reference value, counting the times that the hour prediction value is lower than the day prediction reference value, and calculating the real-time risk rate. Efficient decision support is provided for power system dispatching, and the safety and stability of a power grid are effectively improved.

The invention provides a laser ceilometer system error compensation method based on regression analysis, and belongs to the technical field of laser ceilometers. A measurement range is divided into four height intervals by constructing a layered height interval step regression equation set, and an independent nonlinear regression equation is established; designing a double-layer game optimization framework to realize collaborative optimization of global error minimization and local fitting precision maximization, executing historical data preprocessing and data set division, and implementing a double-layer game model collaborative optimization algorithm to determine a regression equation coefficient and a neural network parameter; a self-adaptive cloud height error correction model based on a Transform architecture is constructed to realize real-time error compensation, the optimal performance of the model is kept through sliding time window monitoring and automatic retraining, and the technical problem that the measurement precision of the laser ceilometer is affected by atmospheric environment parameters and equipment parameter changes, and consequently system errors are remarkable is solved.

The invention relates to the technical field of flywheel energy storage, and discloses an SOC estimation method for a flywheel energy storage array system. According to the method, running parameters such as flywheel rotating speed and bus current are synchronously collected through a multi-source sensing unit, a dynamic parameter observation model is input to estimate dynamic characteristic parameters, and the model dynamically adjusts observation weight based on historical data. And inputting the dynamic characteristic parameters into a multi-model fusion algorithm to generate a fused state feature vector, wherein the algorithm distributes fusion coefficients according to spatial relevance of each flywheel unit. And correcting the state feature vector deviation by using an error correction model to obtain an SOC estimation result, and storing the SOC estimation result. The method also relates to the steps of model construction, data acquisition and processing, algorithm optimization and the like. The method can improve the SOC estimation precision, is suitable for complex working conditions, guarantees the data integrity and safety, and is of great significance to the development of the flywheel energy storage technology.

The invention provides a short-term power load prediction method and device, electronic equipment and a medium, and the method comprises the steps: obtaining and preprocessing historical load data and influence factors, carrying out the initial prediction through a bidirectional LSTM model fusing an attention mechanism, recognizing a current stage through a load mode state classifier, and carrying out the prediction of a short-term power load. The three types of error correction models are a stable period, a transition period and an abnormal period, and then adapters are selected from the three types of error correction models which are independently trained, namely, a lifting tree model with low correction strength is adopted in the stable period, a medium correction model supporting incremental learning is adopted in the transition period, and a high correction strength model supporting online learning is adopted in the abnormal period; and performing targeted error compensation on the initial prediction value, and finally outputting an optimized load prediction result. According to the method, through a staged adaptive correction mechanism, the power load prediction precision and the anti-interference capability are remarkably improved, and the performance is better especially in a load mode sudden change or abnormal scene.

The invention relates to the technical field of assembly, and discloses an assembly method for a prefabricated equipment foundation of a transformer substation, which comprises the following steps: acquiring an initial position and a rotation angle of each prefabricated part as initial data of an assembly process, the initial data comprising a spatial position and a rotation angle of the part; calculating an assembly precision error of each prefabricated part based on the initial data, wherein the assembly precision error comprises a position error and a rotation error; and a multi-objective optimization algorithm is adopted, the position and the rotation angle, needing to be adjusted, of each component in the assembly process are optimized based on the calculated assembly precision errors, an error correction model is established, and the optimization algorithm comprises a genetic algorithm and a particle swarm optimization algorithm. According to the method, the Kalman filtering algorithm and the real-time assembly error correction model are combined, the position and posture of each component can be dynamically optimized in real time, the assembly precision is improved, error accumulation is reduced, and the overall assembly efficiency is improved.

The invention provides a gearbox variable working condition mechanical loss torque efficiency correction test method based on a neural network. The method comprises the following steps: S1, determining a to-be-tested gearbox and basic parameters; s2, building a gearbox variable working condition test system; s3, acquiring basic operation data of the gearbox; s4, calculating theoretical mechanical loss torque and efficiency of the gear box; s5, the actually-measured mechanical loss torque and efficiency of the gearbox are collected, and an error data set is generated; s6, constructing and training an error correction model fusing the back propagation neural network and support vector regression; s7, performing mechanical loss torque and efficiency correction based on the fusion model; and S8, verification and feedback optimization of a correction result. According to the method, on the basis of all-working-condition data, the defect of a single model is overcome through double-algorithm fusion, precise correction of the mechanical loss torque and efficiency of the gearbox under the variable working conditions is achieved, and reliable data support is provided for gearbox design optimization, working condition matching and energy efficiency evaluation.

The invention provides a UWB positioning method and system based on multi-unmanned aerial vehicle cooperation, and the method comprises the steps: carrying out the initial configuration of unmanned aerial vehicles carrying a UWB base station according to preset flight control parameters, constructing a communication link between the unmanned aerial vehicles, and generating a node configuration set. An error correction model is constructed by using UWB ranging data between known and unknown nodes, the position of the unknown node is solved, and a three-dimensional air positioning base station coordinate set of all unmanned aerial vehicle nodes is generated. And performing time reference alignment on all the unmanned aerial vehicles, and constructing a time difference positioning model based on the coordinate set. By activating the mobile terminal, the mobile terminal is controlled to broadcast UWB signals to the multiple unmanned aerial vehicles, signal propagation duration is obtained and input into the time difference model, and space coordinates of the time difference model are calculated. According to the scheme, multi-unmanned aerial vehicle air ad hoc network positioning is supported, dependence of fixed anchor points is avoided, dynamic deployment, rapid networking and air positioning in a rescue scene can be achieved, and the accuracy of positioning and navigation in a complex rescue scene can be effectively improved.

The invention discloses a redundant biaxial rotation inertial navigation state monitoring method based on an error correction model, which is particularly suitable for a platform equipped with multiple sets of inertial navigation systems with indexing mechanisms. By correcting a speed error equation, the influence of a specific force item is eliminated, and the problem of inaccurate calculation of a speed error model in a dynamic environment is solved. Meanwhile, a combined state Kalman filter under a speed error correction model combining navigation information of the two inertial navigation systems is constructed, and the relative attitude, the relative speed and the relative position between the two systems are used as observation values. A residual error normalization strong tracking filtering technology is adopted, and gyroscopic drift and accelerometer zero offset of the two sets of systems are estimated and monitored on line. And designing a monitoring algorithm based on error parameters of online estimation, and evaluating the state of the inertial device. In the whole process, external reference information is not needed, and inertial device state monitoring can be realized only by depending on navigation data of the two inertial navigation systems.

The invention discloses a thermal power generating unit coordination control data and mechanism hybrid modeling method and device, computer equipment and a computer readable storage medium, and relates to the technical field of thermal energy engineering and power system modeling and control. Establishing a high-precision dynamic mechanism model of the working medium flow and heat transfer process of the boiler-steam turbine system; then, constructing an error correction model fusing the long-short-term memory network, the convolutional neural network and residual connection; and finally, fusing the mechanism model and the error correction model through parallel computing to form a dynamic hybrid model with physical interpretability and high precision. Through the hybrid modeling strategy, the prediction accuracy of key parameters such as the main steam pressure, the separator outlet steam enthalpy value and the unit load in the wide load range, especially under the dry-state operation working condition is remarkably improved, and a reliable model basis is provided for designing an advanced unit coordination control system.

The invention discloses an on-chip calibration device parasitic parameter determination and traceability method based on an inversion algorithm, and the method comprises the steps: determining eight error terms through TRL calibration which employs three calibration standard components: a straight-through calibration standard component, a reflection calibration standard component and a transmission line calibration standard component; the method comprises the following steps of: acquiring a relation between error terms in TRL calibration and SOLT calibration according to a signal flow graph, acquiring an expression for extracting a true value of a reflection coefficient of a calibration sheet through inversion according to a 12-item error correction model algorithm, substituting scattering parameters acquired by testing into an error term analytical solution of TRL calibration, solving a TRL calibration error term, and calculating the true value of the reflection coefficient of the calibration sheet according to the error term analytical solution of the TRL calibration. The reflection coefficient truth values of the open circuit and short circuit of the calibration piece and the load piece are extracted; according to the method, the extraction process of the parasitic parameters of the calibration piece is simplified, error sources introduced in the extraction process of the parasitic parameters of the calibration piece are reduced, and the extracted parasitic parameters of the calibration piece have a good effect.

The invention discloses a method for forecasting flood in a drainage basin influenced by strong human activities based on a hybrid model. The method comprises the following steps: firstly, constructing a distributed hydrological model combining sub-drainage basin division and a Muskingum method, and performing calibration; using a hydrological model simulation error and multi-step watershed average rainfall based on river channel propagation time as characteristics, and using a Bayesian optimized LSTM (Long Short Term Memory) to construct an error correction model; and finally coupling the two models to realize real-time forecasting. Verification of the Lanxi river basin shows that compared with a traditional model, the Nash efficiency coefficient of the mixed model is remarkably improved, the flood peak relative error is smaller than or equal to 8%, the peak current time difference is smaller than or equal to 2 hours, precision is stable within the 12-hour forecast period, a long series of continuous historical data is not needed to serve as support, and only data in the session flood period are needed; the method is suitable for hour-scale flood forecasting of the watershed lack of data and strong human activity, an efficient and reliable technical means is provided for flood control and disaster reduction, and the method has wide engineering application prospects.

The invention relates to the technical field of smart power grids, in particular to a distributed photovoltaic-oriented smart power grid load prediction method and system, and the method comprises the steps: collecting the total load data, distributed photovoltaic power generation power data and meteorological data of a target region and an adjacent region, and carrying out the preprocessing; extracting spatial correlation characteristics and time sequence evolution characteristics of the photovoltaic power generation power data, and fusing to generate a low-dimension spatial-temporal characteristic matrix; splicing the spatial-temporal feature matrix with the total load data and the meteorological data to form a multi-modal fusion feature vector, and inputting the multi-modal fusion feature vector into a machine learning prediction model to generate an initial load prediction result; and introducing real-time updated ultra-short-term weather forecast data, carrying out dynamic correction on an initial prediction result through a linear regression error correction model, and outputting a final load prediction value. According to the method, the space-time precision and the real-time adaptive capacity of load prediction in a distributed photovoltaic scene can be remarkably improved, and the method has good practicability and engineering generalizability.

The invention discloses a satellite precipitation two-stage error correction method based on an actual measurement site, and the method comprises the steps: extracting drainage basin DEM data, obtaining a mask file of a drainage basin, and extracting the satellite precipitation of the drainage basin; performing inverse distance weighted interpolation on the rainfall data of the actual measurement site to a corresponding resolution to obtain grid data of the actual measurement site; error correction data are obtained through a deep learning model or dynamic quantile mapping; calculating a residual error between the rainfall data after error correction and grid rainfall of the actual measurement site, and learning a nonlinear relationship between the rainfall data and the residual error by using a gradient lifting regression tree model; and adding the precipitation residual error of the nonlinear residual error correction model to the error-corrected satellite precipitation error to obtain a residual error-corrected precipitation product. According to the method, the satellite precipitation product is corrected based on the actual measurement site data, the high-precision precipitation product is obtained, point-to-surface conversion of the precipitation data can be realized, and data support is provided for input of a refined hydrological model.

The invention relates to a hydropower cluster power generation power prediction method based on a residual hybrid model, and the method comprises the following steps: firstly, collecting the historical power data of a hydropower cluster, and carrying out the periodic coding, and forming a time feature set; historical power data is subjected to lagging processing, a lagging power feature set is constructed, the lagging power feature set and a time feature set are combined into a comprehensive feature set, and then the comprehensive feature set is divided into a training set and a test set. Training a basic prediction model by using the comprehensive feature set to obtain basic prediction results of the training set and the test set; and calculating a residual error based on a prediction result of the training set, constructing a residual error data training set, and training a residual error correction model according to the residual error data training set. And finally, performing residual error prediction on the test set through the residual error correction model, and adding a residual error prediction result and a basic prediction result to obtain a final prediction value. According to the method, the prediction precision and robustness are effectively improved through residual hybrid modeling.

The invention provides a text error analysis method and device based on a large model and a medium, and belongs to the technical field of natural language processing. The text error condition analysis method based on the large model comprises the following steps: constructing a training data set covering multiple fields and multiple error types, inputting the training data set into a large language model, and supervising and training the large language model; according to a multi-task collaborative error correction mode, a cue word optimization mode and a multi-error correction model fusion mode, an error correction strategy of the large language model output text data is designed; and screening the text data by using an error correction strategy in combination with the multi-error correction model to obtain a plurality of candidate correction results, and selecting an optimal correction result from the plurality of candidate correction results according to a voting mechanism. The problem that in the prior art, a document error correction mode is insufficient in error correction capability, including the problems of poor error correction accuracy and low error correction efficiency can be solved.