Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

53 results about "Quantile regression model" patented technology

Quantile Regression. Ordinary least squares regression models the relationship between one or more covariates X and the conditional mean of the response variable Y given X=x. Quantile regression extends the regression model to conditional quantiles of the response variable, such as the 90th percentile.

A power system distribution substation load data abnormity detection and repair method

The invention relates to a power system distribution transformer load data abnormity detection and repair method, and belongs to the technical field of power system analysis and control. The method comprises a coarse detection stage and a micro detection stage of the abnormal condition of the distribution transformer load data: in the coarse detection stage, missing data is complemented by using alow-rank matrix technology, and an obvious abnormal value is preliminarily repaired; In the micro-detection stage, a random forest quantile regression model is used for constructing the influence relation of factors such as historical loads, weather and day types on the to-be-detected loads, and local abnormal data are finely adjusted. According to the method, diversified data and historical loaddata can be fully utilized to detect and repair abnormal data in the distribution transformer load in the power system, a reliable data basis is provided for power load prediction and power system operation, the power load prediction precision is effectively improved, and the operation cost of the power system is reduced.
Owner:江苏云上电力科技有限公司 +1

Wind electricity power probability density predicting method based on genetic algorithm and support vector quantile regression

The invention discloses a wind electricity power probability density predicting method based on a genetic algorithm and support vector quantile regression. The method is characterized by comprising the following steps: 1, acquiring data of output power of a wind electric field and carrying out data cleaning; a, carrying out normalization processing on sample data, and selecting data of training sets and testing sets; 3, establishing a support vector quantile regression model; 4, optimizing support vector quantile regression parameters by using the genetic algorithm; and 5, establishing a wind electricity power probability density predicting model to obtain a final wind electricity power predicting result. By the genetic algorithm, global searching can be implemented for optimization, wind electricity power predicting precision is improved, nondeterminacy of wind electricity power can be quantified, and basis is provided for safe and stable running of wind electricity connection.
Owner:HEFEI UNIV OF TECH

Quantile probabilistic short-term power load prediction integration method

The invention relates to a quantile probabilistic short-term power load prediction integration method, and belongs to the technical field of power system analysis. The method comprises the steps of: dividing historical load data into two parts, wherein the first part is used for training a single quantile probabilistic prediction model; and the second part is used for determining the weights of multiple prediction methods, so that load prediction is integrated; performing bootstrap sampling on the first part of data, so that multiple new training data sets are obtained; for each training dataset, training three regression models including a neural network quantile regression model, a random forest quantile regression model and a gradual gradient regression tree quantile regression model;and, establishing an optimization model by taking the quantile loss minimization as a target function on the second part of data set, and determining the weights of all kinds of quantile regression models, so that a quantile probabilistic integration load prediction model is finally obtained. By means of the method in the invention, on the basis of all kinds of single prediction models, the probabilistic load prediction precision is further improved; and the running cost of the power system is easily reduced.
Owner:TSINGHUA UNIV

Method for power load condition density prediction

ActiveCN104217258AAccurately describe the law of changePowerfulForecastingModel selectionAlgorithm
A method for power load condition density prediction comprises the steps of model building, model solution, model selection and condition density prediction. The method has the advantages that a power load neural network quantile regression model is built, the advantages of a neural network model and a quantile regression model are combined, the variation law of the power load can be depicted accurately, and the method shows powerful functions. A standard gradient optimization algorithm of the power load neural network quantile regression model is given; and on the premise that model estimation accuracy is not affected, the model computation speed is improved. An AIC criterion of the neural network quantile regression model selection of the power load is established, and the problem that the model is too complex and is subject to overfitting is effectively solved. The power load condition density prediction method is established on the basis of the neural network quantile regression, not only is the model prediction accuracy remarkably improved, but also whole probability density prediction results of the power load are obtained, and the method can provide more useful information and facilitate scientific decision-making.
Owner:STATE GRID CORP OF CHINA +1

Power load probability density prediction method based on fuzzy support vector quantile regression

The invention discloses a power load probability density prediction method based on fuzzy support vector quantile regression. According to the method, maximum day load data and average temperature data before a prediction day are collected; a train set and a test set are established through adoption of history data; lagrangian multipliers and support vector subscripts of a fuzzy support vector quantile regression model are obtained through utilization of the train set; the fuzzy support vector quantile regression prediction model is established according to obtained model parameter values; the test set is substituted into the model to obtain prediction values; and probability density prediction of the maximum day load is realized through utilization of the obtained prediction values under different quantiles and through application of kernel density estimation. According to the method, prediction errors can be effectively reduced, the power load prediction precision is improved, the good prediction effect is obtained, and the relatively reliable basis is provided for a power system scheduling department to adjust power consumption plans and optimize generator set contribution.
Owner:HEFEI UNIV OF TECH

Nuclear extreme learning machine quantile regression-based wind power interval prediction method

The invention discloses a nuclear extreme learning machine quantile regression-based wind power interval prediction method. The method comprises the steps that a wind power plant output power and windspeed data is collected; the data is processed simply, and unreasonable data is deleted; a nuclear extreme learning machine quantile regression model is built; by means of a particle swarm algorithm,nuclear extreme learning machine quantile regression parameters are optimized, and a regression module is determined; test data is put, and a wind power prediction interval is obtained. Accordingly,the nuclear extreme learning machine quantile regression principle and a nuclear extreme learning machine model are effectively combined, and the optimal model parameter is obtained by conducting search and optimization through the particle swarm algorithm, uncertain information in wind power can be effectively grasped, then a better prediction result is obtained, and the basis can be provided forsafe and stable running of wind power integration.
Owner:中电投东北新能源发展有限公司

Genetic support vector machine photovoltaic power interval prediction method based on quantile regression

ActiveCN108921339AImprove forecast accuracyMake up for improper selection leading to over-learningForecastingCharacter and pattern recognitionSupport vector machineWeather factor
The invention discloses a genetic support vector machine photovoltaic power interval prediction method based on quantile regression. The method comprises the steps of extracting a solar radiation value, a temperature value and photovoltaic power of historical data to obtain a data sample, and performing normalization preprocessing; optimizing parameters of a support vector machine through a genetic algorithm to overcome the fluctuation and randomness of photovoltaic power generation, building a prediction model, and obtaining high-precision photovoltaic deterministic predictive power; and by analyzing a photovoltaic power prediction error of the prediction model, determining a quantile regression variable, and building a corresponding quantile regression model according to uncertain weather factors, so that photovoltaic power interval prediction is achieved. According to the method, photovoltaic power prediction error distribution does not need to be assumed; accurate photovoltaic power interval prediction ranges under different confidence degrees are obtained; richer information is provided for dispatching decision and operation risk assessment of an electric power system; and thetechnical problem of a non-ideal photovoltaic power interval prediction result is solved.
Owner:NANJING INST OF TECH

Regional wind power prediction method and system based on space-time quantile regression

ActiveCN110648014ASolve the problem of choosing explanatory variablesReduce the impact of safe and stable operationClimate change adaptationForecastingNumerical weather predictionAlgorithm
The invention provides a regional wind power prediction method and system based on space-time quantile regression. The method comprises the following steps: collecting the operation and numerical weather prediction data of a plurality of wind power plants in a preset time period, converting the collected data into a feature map, and building a training set, a verification set and a test set; establishing a space-time quantile regression model, and training and optimizing the model by utilizing the training set, the training set, the verification set and the test set; acquiring operation data and environment data of each wind power plant in real time, and predicting regional wind power generation in a certain time period in the future according to the optimized space-time quantile regression model. According to the invention, short-term non-parameterized probability prediction is carried out on regional wind power through the space-time quantile regression model; the selection problem of explanatory variables in regional wind power prediction with large input information is solved, the prediction accuracy and reliability are greatly improved, and a specific solution is provided forregional wind power generation probability prediction with big data.
Owner:SHANDONG UNIV +3

Generation method and system based on wind power/photovoltaic classic scene set

ActiveCN110807554AThe prediction error probability density function is accurateSave time and costForecastingNormal densityAlgorithm
The invention provides a generation method and system based on a wind power / photovoltaic classic scene set. The generation method comprises the steps of S100, performing point prediction on wind power / photovoltaic output; s200, combining the wind power / photovoltaic point prediction data and the actual measurement data to perform normalization processing on the point prediction error, and dividingthe point prediction error into a training set and a test set; s300, constructing an error neural network quantile regression model by utilizing the error training set data to obtain corresponding parameters; s400, testing the neural network quantile regression model of the trained error by using the test set, and deriving the probability density distribution of the error; s500, carrying out Latinhypercube sampling on the error probability density function at different moments, performing processing to obtain a wind power / photovoltaic output sample, and carrying out Cholesky decomposition onthe sample to obtain a plurality of initial scenes; and S600, performing preliminary reduction on the initial scene, and then performing scene reduction by adopting a backward reduction method to obtain a classic scene set.
Owner:HEFEI UNIV OF TECH

Flight running time prediction method based on reliability

The invention discloses a flight running time prediction method based on reliability, and belongs to the technical field of flight running time prediction. The method aims to solve the technical problem that flight running time under different reliability is difficult to accurately predict in the prior art. The method comprises the following steps: establishing a meteorological message and flightoperation basic database; based on the meteorological message and the flight operation basic database, at least two time periods are extracted from one day to serve as delay time windows respectively;establishing a quantile regression model of each delay time window based on the meteorological message and the flight operation basic database; and predicting flight operation time under different reliabilities according to flight planned departure time and the quantile regression model.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

QRNN-based reasonable range estimation method for low-voltage transformer area line loss rate

The invention discloses a QRNN-based reasonable range estimation method for low-voltage transformer area line loss rate in the technical field of line loss rates. The method aims to solve the technical problem that in the prior art, the line loss rate evaluation basis cannot reflect the difference of a transformer area in structure, power supply range and load level, and comprises the steps of collecting transformer area operation data based on an existing power utilization management system, and screening out transformer area line loss rate influence factors; correlation analysis is carried out on the line loss rate of the transformer area, and line loss rate key factors influencing the line loss rate are extracted; calculating a reasonable range of the line loss rate of the low-voltage transformer area based on a neural network quantile regression model; and according to the calculated reasonable range of the line loss rate of the specific transformer area, diagnosing whether the transformer area is an abnormal transformer area, and adjusting the transformer area. Key factors influencing the line loss rate are extracted on the basis of actual operation data of an existing power utilization information system, an existing marketing system, an existing PMS system and the like, the reasonable range of the line loss rate can be given for a specific transformer area, and a reliable basis is provided for diagnosing a transformer area with abnormal line loss.
Owner:JIANGSU FRONTIER ELECTRIC TECH +1

Listed company performance comprehensive evaluation method

InactiveCN104217296AAccurately describe the changing law of performance levelPowerfulResourcesNerve networkTemporal resolution
The invention discloses a listed company performance comprehensive evaluation method. The listed company performance comprehensive evaluation method comprises steps such as model establishment, model resolution, model selection, condition density prediction, randomness evaluation, and the like. The effects of the listed company performance comprehensive evaluation method are that a neural network quantile regression model of the listed company performance comprehensive evaluation is established, the priorities in the two aspects of the neural network model and the quantile regression model are combined, a variation rule of the performance level of the listed company can be depicted exactly, and powerful functions are presented. As a standard gradient optimization algorithm of the neural network quantile regression model for the performance comprehensive evaluation of the listed company is given, the calculation speed of the model is improved under the prediction of not affecting the model estimation accuracy. As an AIC standard selected by the neural network quantile regression model of the listed company performance comprehensive evaluation is established, a suitable model structure can be selected, and the model is prevented from being too complex and falling in an over-fitting awkward situation efficiently.
Owner:STATE GRID CORP OF CHINA +1

Energy demand condition density prediction method

ActiveCN104217105ASimplify Modeling ComplexitySpecial data processing applicationsAlgorithmNonlinear structure
The invention relates to an energy demand condition density prediction method. The method comprises the following steps of establishing a support vector quantile regression module; establishing a support vector weighing quantile regression module for energy demand; estimating the parameters of the models; predicting the condition density, and the like. The method has the beneficial effects that by combining the advantages of non-linear processing capability of a support vector machine and complete description capability of quantile regression on the condition distribution feature, the support vector quantile regression module for predicting the energy demand is established; on one hand, the non-linear structure of an energy system in a low-dimension space is mapped into a high-dimension space by the support vector machine, and is converted into a linear structure, so the complexity of modeling is reduced; on the other hand, the change rule of the whole condition distribution of energy demand is depicted by the quantile regression, and more available information is provided; a non-parameter kernel density estimation technology is adopted to establish the energy demand condition density prediction method, and the complete prediction of whole condition distribution feature of energy demand is realized.
Owner:STATE GRID CORP OF CHINA +1

Quantile regression-based wind power fluctuation interval analysis method

The invention discloses a quantile regression-based wind power fluctuation interval analysis method. The method comprises the steps of obtaining output power data of a wind power plant firstly; determining the quantiles of the output power data of the wind power plant; establishing a regression function and a regression model for each quantile by adopting a support vector machine; next, solving the regression model of each quantile by adopting a prime-dual interior point method, and calculating the quantile in the next moment; S6: obtaining the actually measured value of the wind power data atthe next moment; and finally, returning to the repeated cycle to obtain the output power data fluctuation interval of the wind power plant. The quantile regression-based wind power fluctuation interval analysis method provided by the invention can realize the adaptive selection of the regression functions through the support vector machine without making any assumed conditions on the random disturbance terms, so as to determine the quantile regression models; next, the models are solved by the prime-dual interior point method with infeasible initial points, so that wind power fluctuation interval analysis for the further moments is realized; and by virtue of the method, the complete wind power fluctuation interval analysis result can be obtained, and the newest change condition also can be reflected in real time.
Owner:TSINGHUA UNIV +3

Cluster photovoltaic power probability prediction method and system, medium and electronic equipment

The invention provides a cluster photovoltaic power probability prediction method and system, a medium and electronic equipment, and the method comprises the steps: collecting the historical data of all photovoltaic stations, and carrying out the normalization of the collected historical data; respectively extracting representative features from the input data of the single photovoltaic stations by using an improved convolutional neural network-quantile regression model, and comprehensively extracting correlation features between the regional photovoltaic stations; the improved convolutional neural network-quantile regression model outputting a quantile prediction result of the regional photovoltaic power generation power according to the extracted correlation characteristics between the regional photovoltaic stations. According to the invention, the structure of the convolutional neural network is improved; the improved convolutional neural network firstly carries out feature extraction on each photovoltaic field station in the region, and then carries out correlation feature extraction on the photovoltaic field stations in the whole region, so that the precision of cluster photovoltaic power probability prediction is greatly improved, and the calculation cost is reduced.
Owner:SHANDONG UNIV

Electric quantity prediction system and prediction method thereof

The invention relates to an electric quantity prediction system and a prediction method thereof, and belongs to the field of electric quantity prediction. Because the electric quantity measurement model is improved and the quantile regression model is introduced, the valuable electric quantity reference interval can be predicted more accurately, the interference of data noise is effectively reduced, the characteristics of the electric quantity are further extracted from time and space by combining the advantages of the long and short-term memory network and the convolutional neural network, and the attention mechanism is used to update the weight and optimize the feature extraction effect, and the quantile regression model is used to realize probability prediction, so that a credible electric quantity prediction reference range can be provided for researchers on the basis of ensuring the accuracy of point prediction.
Owner:SHANGHAI UNIVERSITY OF ELECTRIC POWER

Surface temperature prediction method considering spatial dependency and heterogeneity

The invention belongs to the field of heat island effect, and relates to a surface temperature prediction method considering spatial dependency and heterogeneity. The data mainly comprise surface temperature data. The influence mechanism of the influence factors on the surface temperature is mined by adopting the spatial quantile regression model, and then the surface temperature is predicted. According to the method, the heterogeneity influence of influence factors on the surface temperature is considered while the spatial dependency of the surface temperature is considered, the surface temperature is predicted with higher precision, and theoretical support is provided for efficient treatment of the heat island effect.
Owner:TIANJIN UNIV

Short-term photovoltaic output probability prediction method based on simplest gated neural network

InactiveCN112465251AReduce input dimensionalityAvoid heavy prediction calculationsForecastingNeural architecturesData setAlgorithm
The invention relates to a short-term photovoltaic output probability prediction method based on a simplest gated neural network, and the method comprises the following steps: 1), carrying out the normalization of original data containing a plurality of to-be-selected weather variables, and carrying out the reduction of the dimension of the original data through employing a maximum information coefficient MIC; 2) dividing the reduced feature data set into a training data set and a test data set, and respectively dividing the training data set and the test data set into four weather type data of sunny days, cloudy days, cloudy days and rainy days by adopting a K-means algorithm; 3) constructing a neural network quantile regression model and performing training by adopting the training dataset; and 4) performing prediction by adopting the trained neural network quantile regression model to obtain quantiles under various conditions, and obtaining an approximately complete probability density function through kernel density estimation. Compared with the prior art, the invention has the advantages that the prediction reliability and precision are improved, the prediction interval is narrower, the coverage rate is higher, and the method is simple and rapid.
Owner:SHANGHAI UNIVERSITY OF ELECTRIC POWER

Landslide displacement prediction method and device and storage medium

The invention provides a landslide displacement prediction method and device, and a storage medium. The landslide displacement prediction method comprises the following steps: obtaining historical monitoring data of landslide deformation induction factors and landslide displacement in a preset time as samples; establishing a neural network quantile regression model to obtain landslide displacementprediction results y of the m neural network quantile regression base learners; utilizing a kernel density estimation method to obtain a probability distribution function of a landslide displacementprediction result; and taking the probability distribution function as a weight, and obtaining a final combined prediction value of the landslide displacement through weighted average. The technical scheme provided by the invention has the beneficial effects that by utilizing conditional quantiles, the probability density function of each landslide displacement is obtained through a kernel densityestimation method, the final combined prediction value of the landslide displacement is obtained through weighted average, multiple prediction models can be subjected to weighted combination, the combined prediction model can eliminate large deviation generated by a single prediction model, and the reliability and precision of landslide displacement prediction are remarkably improved.
Owner:CHINA UNIV OF GEOSCIENCES (WUHAN)

Photoelectric probability density prediction method based on B-spline quantile regression

The invention discloses a photoelectric probability density prediction method based on B-spline quantile regression, and the method comprises the steps: 1, collecting photoelectric data, carrying outthe normalization of the photoelectric data, and dividing historical photoelectric data into a training set and a test set; 2, constructing a B-spline quantile model, and calculating parameters of theB-spline quantile regression model by utilizing the training set data; and 3, substituting the test set data into the B spline quantile model to obtain predicted values under different quantiles, andrealizing photoelectric probability density prediction by applying kernel density estimation. The prediction precision of photovoltaic power generation can be improved, and the uncertainty of a prediction result is comprehensively measured, so that a reliable basis is provided for safely and stably connecting the photovoltaic power generation into a power grid.
Owner:HEFEI UNIV OF TECH

Hotel yield prediction method, system and device and storage medium

The invention provides a hotel yield prediction method, system and device and a storage medium, and the method comprises the steps: obtaining the yield data of a plurality of hotels, wherein the yielddata comprises historical yield data, hotel attribute feature data and current reservation progress data; constructing a quantile loss function based on the quantile regression model; constructing aninitial deep learning network model; based on the quantile loss function and the yield data of the plurality of hotels, training the initial deep learning network model, and respectively obtaining atarget network model corresponding to each hotel; predicting the yields of the plurality of hotels in a future preset time period based on the target network model, and outputting the yield value of each hotel in the preset time period; according to the invention, the yield of multiple hotels can be predicted at the same time, and the hotel yield prediction accuracy is improved.
Owner:CTRIP COMP TECH SHANGHAI

Wind power non-parametric probability interval ultra-short-term prediction method

ActiveCN110598929AInterval score excellentImprove forecast confidenceForecastingNeural architecturesLearning machineGuideline
The invention discloses a wind power non-parametric probability interval ultra-short-term prediction method. The method is based on adaptive LASSO and an extreme learning machine. Firstly, nonlinear quantile regression is carried out on a wind power sequence to obtain self-adaptive adjustment parameters; an optimal quantile regression model output coefficient is calculated based on an extreme learning machine by utilizing quantile regression based on adaptive LASSO and an improved Bayesian information criterion; and finally, a wind power time sequence is input to obtain an ultra-short-term prediction value. According to the quantile regression prediction model constructed through the method, the interval score is obviously superior to that of a traditional prediction model based on quantile regression, the prediction precision and interval width comprehensive indexes are good, and the reliability of wind power prediction is greatly improved.
Owner:HOHAI UNIV +3

Housing source value parameter estimation method and apparatus

InactiveCN109544410AValuation results are objective and accurateProduct appraisalCell EvaluationEstimation methods
The present application relates to the field of machine learning and neural networks for processing housing source data of a community, in particular to a housing source value parameter estimation method and apparatus, a computer device and a storage medium. The method comprises the following steps of: acquiring each cell index of a target cell; inputting it into the preset cell similarity model to obtain the cell evaluation index of the target cell; searching a similar cell corresponding to the target cell according to the cell evaluation index of the target cell in the preset cell set; obtaining the average value parameters of the housing source of the similar district, estimating the average value parameters of the housing source of the target district, obtaining the average value parameters of the housing source of the target district, obtaining the evaluation parameters of the target housing source, and obtaining the estimated value parameters of the target housing source throughthe preset semi-parametric quantile regression model. This application can estimate the value parameters of the target house without referring to the transaction data of the house of the same house type as the target house source and without requiring the valuation of the valuer, and the evaluation result is more objective and accurate.
Owner:平安直通咨询有限公司

Data analysis method based on quantile logistic regression

The invention discloses a data analysis method based on quantile logistic regression, and relates to a data analysis method based on quantile logistic regression. When a traditional logistic regression model is used as a classifier, the given information is very limited, and the prediction accuracy of the model is very poor. The method comprises the steps of 1, data cleaning and preprocessing; 2,quantile logistic regression modeling is carried out; 3, parameters of the quantile logistic regression model is solved, and the parameters are solved to minimize the sum of weighted error absolute values; and 4 an unconstrained quantile regression objective function and a constrained quantile regression objective function are calculated. Researchers can research different groups more carefully, and more information can be obtained through data. Meanwhile, the quantile regression model is not sensitive to abnormal values, the prediction effect of the model is not affected by a small number ofabnormal values, and the robustness of the model is greatly improved.
Owner:深圳索信达数据技术有限公司

Method for recommending heat island retarding measures according to heat island intensity grade

The invention belongs to the field of heat island effect and urban planning management, and relates to a recommendation method for formulating retarding measures according to heat island intensity grades. The data mainly comprise air temperature data or surface temperature data and influence factor data. According to the method, a spatial quantile regression model considering spatial dependency and heterogeneity is adopted, the difference influence of urban elements on the heat island effect is mined, and differential retarding measures according to different heat island intensities are provided according to a regression result. Therefore, targeted and efficient slowing-down measures are selected in different regions according to the intensity of the heat islands in the regions instead of adopting a unified policy with other regions. By adopting the method, unnecessary waste of manpower, material resources and financial resources is reduced, resource elements are scientifically configured, and the utilization efficiency of resources is improved.
Owner:TIANJIN UNIV

Short-term building load probability density prediction method

The invention relates to a short-term building load probability density prediction method, which comprises the following steps of: 1) screening an original exogenous variable by adopting an orthogonalmaximum correlation coefficient by taking a maximum correlation coefficient and minimum redundancy as principles to obtain a selected exogenous variable; 2) for the selected exogenous variable, introducing a binary risk indication variable to improve the accuracy of peak load prediction; 3) constructing a convolution gating quantile regression model, taking the screened exogenous variables and binary risk indication variables as input together, and taking building load prediction values under different quantiles as output for training; and 4) predicting by adopting the trained constructed convolution gating quantile regression model to obtain predicted values under different quantiles, and obtaining a probability density distribution function from the predicted values under the differentquantiles through a fitted kernel function. Compared with the prior art, the method has the advantages of improving the peak time prediction accuracy, improving the prediction precision, fully extracting features and the like.
Owner:SHANGHAI UNIVERSITY OF ELECTRIC POWER

Disturbance voltage maximum amplitude quantification method and device of in-situ intelligent measurement equipment

The invention discloses a disturbance voltage maximum amplitude quantification method and device of in-situ intelligent measurement equipment. The disturbance voltage maximum amplitude quantification method of the local intelligent measurement equipment comprises the following steps: extracting an inter-fracture breakdown voltage and an nth micropulse amplitude of a port disturbance voltage of nth breakdown of a VFTO in the same disconnecting switch operation process; forming an event pair by the extracted breakdown voltage between the fractures and the micro-pulse amplitude so as to form a sample set of a quantile regression model; and training a quantile regression model based on the sample set to generate a corresponding quantile regression equation, the generated quantile regression equation being capable of quantitatively evaluating the port disturbance voltage amplitude of the in-situ intelligent measurement device under the operation of the isolation switch. Therefore, the disturbance level of the on-site intelligent measurement equipment port of the specific GIS test loop under the operation of the isolating switch can be quantitatively evaluated, and a targeted immunity test assessment scheme is formulated.
Owner:CHINA ELECTRIC POWER RES INST +1

Method and device for predicting demand quantity of resource supply chain and storage medium

The embodiment of the invention discloses a method and device for predicting the demand of a resource supply chain, and a storage medium, and the method comprises the steps: carrying out the preprocessing and feature extraction of a data source, and obtaining training data, supply data of resources based on different data distribution, and supply data of resources with different data qualities; training and predicting the hierarchical model to obtain a prediction result, wherein the prediction result comprises distribution of a plurality of prediction values of the to-be-supplied resource; outputting a prediction result to a quantile regression model; and performing regression analysis on a prediction result by utilizing a quantile regression model in a prediction stage to obtain probability distribution of resources with different data distributions under different quantile combinations and probability distribution of resources with different data qualities under different quantile combinations. According to the scheme, the future commodity supply demand can be comprehensively estimated, and the commodity stockout and waste phenomena are greatly reduced.
Owner:SF TECH +1

A qrnn-based method for estimating the reasonable range of line loss rate in low-voltage station area

The invention discloses a QRNN-based method for estimating the reasonable range of the line loss rate of the low-voltage station area in the field of line loss rate technology, aiming to solve the problem that the evaluation basis of the line loss rate in the prior art cannot reflect the structure, power supply range, and load level of the station area Based on the existing power management system to collect operating data of the station area, the factors affecting the line loss rate of the station area are screened out; the correlation analysis is carried out on the line loss rate of the station area, and the line loss rate that affects the line loss rate is extracted The key factor; based on the neural network quantile regression model, the reasonable range of the line loss rate of the low-voltage station area is calculated; according to the calculated reasonable range of the line loss rate of the specific station area, it is diagnosed whether it is an abnormal station area, and it is adjusted. The method of the present invention is based on the actual operation data of the existing electricity information system, marketing system, PMS system, etc., extracts the key factors affecting the line loss rate, and can give a reasonable range of the line loss rate for a specific station area, for It provides a reliable basis for diagnosing the abnormal line loss area.
Owner:JIANGSU FRONTIER ELECTRIC TECH +1
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products