190results about How to "Improve predictive performance" patented technology

The invention discloses an agricultural load prediction method and system based on multivariate time sequence decoupling multi-modal learning, and belongs to the technical field of agricultural load prediction. Comprising the steps of collecting historical agricultural load and meteorological data; decomposing historical agricultural load and meteorological data by using multivariate variational mode decomposition to obtain cycle, trend and residual mode components; and respectively constructing a time convolutional neural network, a bidirectional gating cycle unit and a support vector regression model for the decomposed period, trend and residual modal component data set, and fully mining feature information of each mode after decomposition, thereby realizing accurate prediction of agricultural load. According to the method, the potential nonlinear space-time coupling relationship between the agricultural load and the meteorological factor is captured, the prediction effect in a seasonal periodic fluctuation scene of the agricultural load and a long-term trend and agricultural load abnormal scene is improved, the agricultural load prediction precision is improved, and a support is provided for reliable and stable operation of a power grid.

The invention provides a hydrological flow long sequence prediction method and system of an improved state space model. The method comprises the steps of collecting multi-source data of a drainage basin to be predicted; constructing a time sequence sample pair by the processed multi-source data through a sliding window method, wherein the time sequence sample pair comprises an input sequence and a target sequence; a HydroMama model is constructed according to the time sequence sample pair, a HydroMama prediction model is trained, and an optimal hyper-parameter combination is searched for; based on the trained HydroMama model, traffic prediction and result restoration are realized by adopting an autoregression mechanism; the calculation efficiency is high, and the continuous state equation discretization of the state space model is utilized, so that the long-distance meteorological-hydrological hysteresis effect which is difficult to capture by a traditional cycle model can be captured; according to the method, skewed distribution of hydrological data is processed through logarithmic transformation, and an anti-overfitting objective function is combined, so that the prediction performance of the model on unseen data is remarkably improved, and the practical application value is high.

The invention relates to a load prediction method TSR-MAEN based on decomposition representation and an attention encoder network. Firstly, the algorithm provides a trend-seasonal decomposition representation method based on comparative learning so as to enhance the decoupling performance of a model on a trend component and a seasonal component of a load sequence and the representation capability of decomposition characteristics. Secondly, in order to deeply mine and decompose time sequence information contained in a feature hidden space and dependency between features, the algorithm proposes a mixed attention encoder network; through segmented embedding, a trend-season dynamic projection mixed self-attention mechanism and a multi-head mixed output mechanism, trend and season multi-dimensional features after load sequence decomposition are effectively utilized to express and enhance the long-range dependence information extraction capability of the model. Experiments show that the TSR-MAEN algorithm provided by the invention has excellent performance in an electrical load prediction task of oil and gas development, and the prediction error is reduced by 8.51% compared with that of an optimal baseline algorithm.

The invention discloses a large-range landslide mass displacement early warning method based on GNSS and radar data fusion, and the method comprises the steps: obtaining GNSS monitoring sequence data in a landslide region based on the position information of a reference station and a GNSS monitoring station in a GNSS monitoring system, and further obtaining the displacement related information of the landslide region; collecting data through a satellite InSAR, and generating deformation field data of a landslide area by comparing radar images at different time points; performing space-time alignment on the obtained data, and performing two types of data fusion S by adopting an adaptive Kalman filtering method after alignment; decomposing time series data of the fused landslide mass displacement by using wavelet transform, and extracting multi-scale features of a time domain and a frequency domain; the proposed features are combined with historical data of landslide mass displacement, a landslide mass displacement trend prediction model using an LSTM method added with a memory decline factor is input, and future trend prediction of landslide displacement is carried out; setting a multi-level threshold value, judging the change trend of the landslide mass displacement according to the displacement rate, the acceleration and the prediction result, and generating a multi-level early warning signal.

The invention discloses a cross-temperature health state assessment method and system for an underwater robot battery, and relates to the technical field of battery health state assessment, and the method comprises the steps: S1, collecting data in the operation process of an underwater robot; s2, an environment temperature profile is constructed, and working condition types are identified; s3, constructing a multi-dimensional health feature vector; s4, constructing an LSSVM and SVR regression model, and optimizing model hyper-parameters; s5, calculating characteristic stability and comprehensive complexity indexes, and adaptively selecting a model; s6, identifying the working condition category of the charge-discharge cycle, extracting and standardizing the features, calling the adaptive model, iteratively outputting SOH estimated values, and splicing to form an SOH evolution curve; according to the cross-temperature health state evaluation method and system for the underwater robot battery provided by the invention, the problems of unstable cross-temperature performance and insufficient generalization ability in the prior art are solved.

The invention relates to a multi-task face sensing method and device, electronic equipment, a storage medium and a computer program product. The multi-task face sensing method comprises the steps that a to-be-detected face picture is acquired; a to-be-detected face picture is input into the multi-task face perception model, face features and positioning reference points are extracted from the face picture through the multi-task face perception model, and the positioning reference points are used for executing a positioning task about face key point detection; semantic reference points are calculated based on the positioning reference points through a multi-task face perception model, and the semantic reference points are used for executing semantic tasks about face state detection; and executing a positioning task and a semantic task based on the face features, the positioning reference points and the semantic reference points through a multi-task face perception model. Therefore, by explicitly utilizing the physical representation of the face key points, effective supervision for the semantic reference points is realized, the guiding effect of the face key points on the face perception task is enhanced, and the prediction effect of the face perception model can be improved.

This invention relates to the field of urban flooding prediction technology, and discloses a method, system, and medium for constructing a training dataset for urban flooding prediction. The method includes: acquiring historical rainfall data, performing event-based segmentation, and extracting multi-dimensional features of each rainfall event; calculating the distance matrix between any two rainfall events in each dimension of features; weighting and fusion of the distance matrices to obtain the comprehensive distance between any two rainfall events; then selecting typical rainfall events through cluster analysis to generate a representative rainfall event set; simulating each rainfall event in the representative rainfall event set using a hydrological and hydrodynamic mechanism model to generate flooding depth distribution labels; and pairing the multi-dimensional features of each rainfall event with the flooding depth distribution labels to generate a training dataset. This invention improves the accuracy of urban flooding prediction by constructing a training dataset closely related to physical processes and capable of enhancing the predictive performance of machine learning models.

This invention discloses a method, system, device, and storage medium for predicting oil pipe corrosion rate based on a PCA-PSO-SVR hybrid model, specifically including the following steps: S1 Collecting corrosion detection data and operating condition parameters of the oil pipe to form a dataset; S2 Preprocessing the dataset; S3 Using the PCA model to perform dimensionality reduction on the dataset and extracting the main features affecting the corrosion rate; S4 Initializing the parameters of the PSO model; S5 Optimizing the parameters of the SVR model using the PSO model; S6 Constructing a corrosion rate prediction model based on the optimized SVR model; S7 Inputting the preprocessed dataset into the corrosion rate prediction model to obtain the prediction result and complete the prediction.

The invention relates to the technical field of energy scheduling of an energy storage system, and provides a partition control energy scheduling method and system for an energy storage system in a holographic data scene, which realize synchronous acquisition and fusion of an internal battery state and external environment data to form a complete system operation portrait. And further, a multi-parameter SOE deviation dynamic model including charging and discharging behaviors, capacity attenuation trend and thermal influence is established, so that the system can sense energy deviation of each partitioned energy storage unit in real time. On the basis, a partition cooperative control decision model with the minimum SOE deviation as a target function is constructed, multiple constraints such as the operation capacity of the energy storage converter, the power grid requirement and the safety limitation are integrated, the model is solved through a Lagrange multiplier method, and initial power distribution of all partitions is obtained. And finally, correcting an SOE prediction error by means of high-frequency real-time input of holographic data, and adjusting charging and discharging power distribution in combination with SOE dynamic deviation to realize refined optimization of a partition scheduling scheme.

The invention provides a method and system for collecting working state data of a manufacturing execution system, and the method comprises the steps: obtaining an equipment working data set, a personnel working video set and a workpiece information set, and respectively extracting a corresponding equipment working feature sequence set and a corresponding workpiece feature sequence set, performing illumination compensation, foreground segmentation and time window splitting on the personnel operation video set, and performing multi-dimensional attitude decoupling, action semantic recognition and production rhythm modeling on the personnel operation video set after the time window splitting to obtain a personnel action feature sequence set; performing dynamic time warping, beat alignment and feature binding based on the production task order to obtain a multi-modal production feature sequence set, and performing collaborative feature enhancement and dimension reduction compression to obtain a multi-modal working time sequence feature sequence set; and by taking the production task order as an edge index, constructing a dynamic hypergraph model of the multi-modal working time sequence feature sequence set, and carrying out man-machine collaborative analysis on the dynamic hypergraph model to obtain a working state data sequence.

The application provides a method for training a blood pressure prediction model based on meta learning, which comprises the following steps: acquiring a training set, dividing the training set into a first training set, a second training set and a third training set; pre-training a blood pressure prediction model using the first training set to obtain a pre-trained blood pressure prediction model; initializing an initial meta learner using the parameters of the pre-trained blood pressure prediction model; training the initial meta learner using a plurality of training tasks in the second training set based on a meta learning algorithm to obtain a target meta learner; initializing the target meta learner for each patient in the third training set respectively to obtain an initial personalized blood pressure prediction model corresponding to each patient, and training the initial personalized blood pressure prediction model corresponding to each patient using the training data of the patient in the third training set to obtain a personalized blood pressure prediction model corresponding to the patient.

The invention relates to the field of carbon emission prediction, in particular to a subway station construction stage carbon emission prediction method and system. Measuring and calculating carbon emission of the subway integrated station at different stages according to a carbon emission coefficient method, and performing data preprocessing to obtain carbon emission monitoring data; performing feature and target variable separation on the carbon emission monitoring data to obtain a feature matrix and a target variable vector; constructing a carbon emission prediction model to perform feature processing on the carbon emission monitoring data to obtain carbon emission feature data; the method comprises the following steps: constructing a carbon emission prediction integrated model based on LightGBM and XGBoost, identifying carbon emission characteristic data through the carbon emission prediction integrated model, and optimizing parameters in the LightGBM model and the XGBoost model by using grid search to obtain a carbon emission prediction amount in a subway station construction stage.

The invention relates to a residual life prediction method based on a deep residual shrinkage TCN. The method comprises the following steps: firstly, performing adaptive noise full-set empirical mode decomposition on vibration data of a vibration part, extracting multi-domain statistical characteristics, constructing a health index based on monotonicity, tendency and predictability weighted fusion, and dividing the health index into a health stage and a degradation stage by using a 3 sigma method; secondly, constructing a depth residual shrinkage TCN, and inputting the health index in the step 1 into the depth residual shrinkage TCN to obtain an intermediate feature of life prediction; and finally, key information in the intermediate features is further screened in combination with a multi-scale attention mechanism, and a final prediction result of the remaining service life is obtained. Experiments prove that the method can effectively improve the prediction precision of the residual life of the vibration component, and compared with the current advanced method, the method has the advantages that the root mean square error is at least reduced by 31%, and the mean absolute error is at least reduced by 24%.

The application relates to the technical field of electric digital data processing, and particularly discloses a big data information collection and processing method and system. The method comprises the following steps: firstly, the running state of an internet data source end is traced back and extracted in a specified window, the source health degree is calculated, and whether the data is reliable is determined according to the source health degree, so that the abnormal source can be timely de-weighted or isolated, abnormal dragging of account closing is avoided, and a basis is provided for subsequent parameter self-adaptation; subsequently, the output data set is called by a data service port according to the trusted state, sensitive fields are subjected to big data encryption processing, link opening idempotent production is written to eliminate repetition caused by retries and stabilize the order, so that an effective data set is generated, the exposure of the plaintext is reduced, repeated writing is eliminated, the caliber and throughput are compatible, and finally, the effective data set is subjected to adaptive data decryption at a data calculation port, minimum-range decryption is completed only in the required period and necessary fields, and processable indexes are output, so that big data information collection and processing are completed.

The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system, and belongs to the technical field of soft measurement. The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system, and belongs to the technical field of soft measurement. The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system, and belongs to the technical field of soft measurement. The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system, and belongs to the technical field of soft measurement. The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system, and belongs to the technical field of soft measurement. The application discloses a sensor fault robust multi-output soft measurement method and system based on adversarial learning, relates to a sensor fault robust multi-output soft measurement method and system,

The invention relates to the technical field of index prediction in the social economic field, and discloses a social economic index set prediction method, which comprises the following steps of collecting multi-source data related to social economy, the multi-source data comprises medical business data of a medical institution, medical insurance data of a medical insurance department and social economic environment data of an external data source; and cleaning and standardizing the collected multi-source data to remove noise, missing values and abnormal values in the data, and unifying the data format and dimension. Data are obtained through multiple channels, the data basis of social and economic index analysis is enriched, multi-aspect conditions are presented, macroscopic information is reflected, understanding of phenomena and trends is enhanced, the accuracy and scientificity of index prediction are improved, and more valuable data support is provided for decision making. The problems of limited data sources and insufficient data mastering in the previous social economic index research are solved.

The application provides an end-to-end business process bottleneck prediction and simulation method based on process mining, comprising: identifying the time proportion of implicit links such as manual review in the path segment which are not recorded by the system through analyzing the timestamp deviation value set, determining the silent activity proportion distribution and the service time expansion range of adjacent nodes; identifying the nodes in the adjacent nodes whose service time expansion range exceeds the preset expansion threshold, obtaining the corrected node service time set by stripping the silent activity time consumption component of each node in the event log; according to the adjusted bottleneck judgment threshold, classifying the time expansion characteristics of each node in the bottleneck candidate node set by using the support vector machine algorithm, and identifying high-risk silent bottleneck nodes; simulating the high-risk silent bottleneck nodes, obtaining path simulation data in the simulation platform, calculating the coverage rate improvement range by using the logistic regression algorithm, and predicting the bottleneck coverage rate.

The application relates to the technical field of hydrogeological monitoring, in particular to a hydrogeological dynamic monitoring and analysis system and method based on big data; specifically, hydrogeological data is collected in real time, data denoising, space-time calibration and error correction are carried out in combination with edge computing, noise and drift error are eliminated by adopting variational mode decomposition, geological model constraint and wavelet transform, and the storage and transmission efficiency is optimized by increment encoding and Huffman encoding, and a quadratic form check code is generated to ensure data reliability; a space-time deep learning model and a self-attention mechanism are fused to predict a hydrological evolution trend, adaptive anomaly detection is realized by combining DBSCAN clustering and prediction error analysis; water resource allocation strategies are optimized through reinforcement learning, risk levels are evaluated in layers, and management schemes are generated, and finally, monitoring data, abnormal scores and decision results can be visually displayed. The application integrates multi-source data fusion, space-time modeling and intelligent decision-making, and improves hydrological monitoring precision and response efficiency.

The application relates to an FPGA prototype verification method and system of a smart meter chip, and belongs to the field of integrated circuit testing and electric power metering. The method aims to solve the problems of ASIC design transplantation to an FPGA platform, large difference between a verification environment and a real power grid, and lack of a real-time closed-loop verification mechanism. A technical scheme efficiently transplants the design to the FPGA by optimizing ASIC front-end synthesis constraints and adopting a netlist-level IP replacement strategy; a hardware-in-the-loop system is constructed, which comprises a real power grid signal source, a separation device simulation front end, an FPGA verification board, a high-precision standard electric energy meter and a host computer. The system can generate a real power grid excitation and realize synchronous collection of prototype output and the standard meter, real-time error analysis and dynamic adjustment of parameters, thereby forming a closed-loop verification. The application significantly improves code transplantation efficiency, provides a highly consistent real power grid verification environment, and greatly improves verification iteration and defect positioning speed.

The application discloses a new energy station electric quantity prediction method and system based on a global-local double-layer prediction architecture, and aims at solving problems of insufficient monthly electric quantity prediction data of new energy stations, large medium and long term weather prediction errors, and low prediction precision of traditional experience methods, adopts a global-local double-layer model architecture, a XGBoost global prediction model is constructed by using all station data in the upper layer, common rules such as seasons, weather, installed capacity, shutdown plan and the like are mined, the global model output is taken as an enhanced feature, a XGBoost local model is trained in combination with single station data in the lower layer, and individual characteristics such as station equipment efficiency and operation and maintenance level are adapted, and a holiday adjustment factor is introduced to post-process and correct the prediction result. The application effectively solves the small sample modeling problem commonly existing in monthly electric quantity prediction, and significantly improves the monthly electric quantity prediction accuracy of wind power stations and photovoltaic stations.

The invention relates to the technical field of process control, and discloses a digital twinning control method and system for coring of a special optical component. The method comprises the steps that a digital twin model is constructed through a multi-source sensor, and a machining geometric reference is determined; distortion features are identified, residual stress distribution is analyzed, and classification identifiers are generated; a dynamic simulation prediction technology is adopted to optimize a machining path based on distortion causes, and an adjustment scheme is formed; quantizing the parameter weight and adjusting the stress release regulation and control parameter to generate an optimized execution scheme; processing dynamic evolution is tracked, the distortion cause attenuation trend is analyzed, and geometric residual error distribution is obtained; if the error does not reach the standard, clamping parameters are adjusted, the model is updated, and a final control strategy is determined; and executing the strategy and integrating multi-source data, and outputting a final processing control result. According to the invention, the precision of special optical component coring processing is improved.

The application provides a method for identifying the inflammatory activity grade of ulcerative colitis based on dynamic graph multi-instance learning, which comprises data collection, data preprocessing, dynamic graph multi-instance learning model construction for inflammation activity diagnosis and model evaluation. The data collection mainly comprises collecting pathological images of ulcerative colitis patients, digitizing through a digital scanner, and excluding a part of pathological images containing problems such as blur, discoloration and abnormal staining, and finally determining the grading label of each pathological image by senior gastrointestinal pathology experts. The data preprocessing mainly comprises processing the digital pathological images, converting the digital pathological images into image blocks, then using a visual basic model to extract features from each image block, and finally constructing a dynamic graph multi-instance learning model for training. The application promotes the exploration of the internal relationship of the image blocks input into the WSI, and improves the prediction effect of the model.

This invention belongs to the field of power system data analysis and power consumption forecasting technology, specifically involving a method and system for predicting the power consumption curve of proxy power purchasers in extreme scenarios. The method first constructs a power consumption forecasting model. The model uses an encoder to extract features from multi-source meteorological data of extreme scenarios through supervised contrastive learning, obtaining meteorological feature vectors for the extreme scenarios. A decoder is then used to predict the total daily power consumption based on the extracted meteorological feature vectors. Multi-source meteorological data for the forecast day is acquired and input into the trained power consumption forecasting model to predict the total daily power consumption for the forecast day. Based on the total daily power consumption for the forecast day, the time-of-use power consumption curve for the forecast day is obtained. This invention achieves high-precision total daily power consumption forecasting under extreme scenarios by efficiently extracting meteorological data features through an encoder and achieving rapid and accurate adaptation with a small number of samples through a decoder.

The application relates to a kind of detection method, device, system and storage medium of LED lamp bead, comprising the following steps, the electrode of the LED lamp bead is tested, and the initial current value of the LED lamp bead is obtained;Based on the initial current value, the LED lamp bead is applied to step voltage, and voltage-current curve is obtained;The LED lamp bead corresponding to the rated operating point in the voltage-current curve is detected by optical detector, and the light spot image is obtained;Based on the light spot image, the LED lamp bead is tested by reverse voltage, and the reverse leakage current value is obtained;Based on the voltage-current curve, light spot image and reverse leakage current value, the quality of the LED lamp bead is evaluated, and the quality evaluation result of the LED lamp bead is obtained, which solves the technical problem that the traditional detection method is mainly concentrated on simple on-off test or single parameter measurement, and it is difficult to comprehensively reflect the comprehensive performance of LED lamp bead.

The invention discloses a scheduling cost-oriented multi-representation fusion time sequence prediction method, and belongs to the technical field of time sequence data processing. The method aims at solving the technical problems that an existing prediction model is insufficient in extreme value prediction, and the optimization target is disjointed from the actual scheduling cost. The method is technically characterized by comprising the following steps of: converting numeric time series data into time series image representation, and performing fusion modeling on the numeric representation and the image representation to capture a local form and a global trend of the data; and a scheduling cost perception loss function is adopted for model training, and the loss function can apply asymmetric cost penalty to under-prediction and over-prediction. According to the method, the prediction precision of sudden change areas such as peak values is improved through multi-representation fusion, and the prediction result more meeting the actual scheduling requirement is generated through the loss function guide model aligned with the service cost, so that the comprehensive scheduling cost and the service default risk are effectively reduced.

The application provides a multimodal information fusion model and method for DTA prediction, the model comprising a drug molecule structure information encoder, a target structure information encoder, a multimodal balance module and a drug target fusion module; the drug molecule structure information encoder uses a Transformer model to encode drug string modal information and uses a GIN model to extract drug graph modal information features; the target structure information encoder uses a Transformer model to encode target string modal information and uses a GCN model to extract drug graph modal information features; the multimodal balance module uses a contrast learning method to balance and integrate drug string and graph modal information and balance and integrate target string and graph modal information; and the drug target fusion module connects the two modal features of the drug and target obtained by the multimodal balance module and is used for DTA prediction.

The application belongs to the field of municipal engineering rainwater real-time measurement, and provides a rainfall picture rainfall intensity calculation method based on raindrop extraction-rainfall intensity calculation, comprising: obtaining rainfall video and decomposing it into rainfall images according to frames; constructing a target function of image raindrop extraction, and extracting raindrop images by using an ADMM algorithm; modifying a CNN model and training the same; and predicting rainfall intensity. The application can prevent random noise in the image from being extracted together, so that the extracted raindrop image is more accurate. The image raindrop extraction algorithm proposed in the application has higher calculation efficiency, can efficiently extract raindrops in the image, and can realize real-time calculation of the algorithm. The application can effectively improve the generalization ability of the CNN model, especially in complex scenes such as night with insufficient light. Rainfall events with large rainfall usually occur at night, and the prediction ability in the night scene has very important significance for the prevention and control of urban waterlogging.

The invention discloses an IT asset management system and method based on data analysis, and relates to the technical field of business information management and intelligent decision, and the system comprises a data fusion module, a digital accompanying shadow construction module, a deduction analysis module, a decision generation module and a dynamic coupling relation digital accompanying shadow calibration module. According to the method, a dynamic digital accompanying shadow containing a physical environment and business logic dual coupling relationship is constructed, and simulation deduction of operation and maintenance operation is carried out on the dynamic digital accompanying shadow, so that fundamental transformation from passive response type operation and maintenance to active prediction type risk management is realized. The method can reveal and quantify hidden cascading risks generated due to physical proximity or business dependence, so that an operation and maintenance team can predict potential chain influences on the whole production system before executing any operation, most secondary fault risks are eliminated in the bud state on the premise that actual production safety is not affected, and the production efficiency is improved. And the stability and the operation toughness of the system are enhanced.

This invention provides an enzyme turnover rate prediction method based on a dual-route hybrid expert mechanism, belonging to the field of bioinformatics technology. It solves the problems of low data utilization, insufficient information mining, and poor robustness in existing prediction methods due to missing values ​​and temperatures. The technical solution includes the following steps: S1: Constructing a unified standard enzyme turnover rate dataset; S2: Extracting multimodal embedding features; S3: Constructing intra- / inter-modal encoders; S4: Combining modalities into an expert hybrid module; S5: Designing an attention fusion mechanism. This invention can achieve high-precision prediction under complex in vitro environmental conditions.