66results about How to "Achieving Precise Forecasting" patented technology

The invention discloses a high-temperature creep deformation prediction and creep damage analysis method for a high-chrome steel component. The method comprises the following steps of: establishing a theoretical model, determining material parameters, designing a numerical integration algorithm, carrying out the secondary development of finite element software, carrying out the creep damage analysis of the component and the like. Compared with an existing technical scheme, the method disclosed by the invention is characterized in that a great improvement is carried out, the accurate prediction of the creep behavior and the creep damage of the high-chrome steel component can be realized, and therefore, the method has an important application value in the fields of the security design and the residual life evaluation of high-temperature high-pressure components in a supercritical generator set.

The invention discloses an online lithium ion battery residual life predicting method based on relevance vector regression, belongs to the technical field of lithium ion battery life prediction, and solves the problem that the residual life of the existing lithium ion battery is predicted by an offline method with low precision. The method comprises the following steps: firstly selecting original samples, performing phase-space reconstruction to construct a training sample set; initializing the model parameters of RVM (relevance vector machine); performing RVM training to obtain a RVM prediction model; comparing the obtained prediction value with ynew, if yes, the constructed novel training set WS equal to WSUINS, retraining RVM, and updating the RVM prediction model; otherwise, keeping the RVM prediction model stable; performing recurrence prediction until the prediction value is smaller than the invalid threshold value U, and finishing the online prediction of the residual life of the predicted lithium ion battery. The method is suitable for prediction of the lithium ion battery residual life.

The invention discloses an iron steel electricity consumption forecasting method and device. The forecasting method comprises the following steps: acquiring iron steel historical data of an area to be forecasted within a preset period of time to serve as sample data, wherein the iron steel historical data comprises electricity consumption index and iron steel electricity consumption; performing dimensionless treatment on the sample data to obtain normalized electricity consumption index and normalized iron steel electricity consumption; adopting the normalized electricity consumption index as an input variable, and the normalized iron steel electricity consumption as an output variable to realize network training so as to build a nerve network model; adopting the normalized electricity consumption index as the input variable and the normalized iron steel electricity consumption as the output variable to build a support vector regression model; combining the nerve network model and the support vector regression model to obtain a combined model, and using the combined model to forecast the iron steel electricity consumption. According to the invention, the demand on the forecasting precision of the iron steel electricity consumption is satisfied to the utmost extent, and reference is provided for economical and reasonable planning of a power grid.

The invention discloses an automatic control system and an optimal trajectory planning method of an intelligent excavator. The system comprises a data acquiring and executing module, a network, a processor and a storage device; and the data acquiring and executing module comprises a 3D scanner, a power sensor, a proximity sensor, a displacement sensor, an angle sensor, a torque sensor, an information device, a controller, the processor, the storage device and a wireless transceiver. According to the automatic control system and the optimal trajectory planning method of the intelligent excavator, because the 3D scanner is adopted to determine the contour of the surface of a to-be-excavated material, the coordinate matrix of the contour is obtained through the processor, and therefore accurate modeling is performed on a material pile, and accurate prediction on the surface load of the excavated complex material pile is achieved. According to the automatic control system and the optimal trajectory planning method of the intelligent excavator, because the information device and the processor are used for optimizing control parameters in the excavating process of the excavator, the purposes that the excavating load is minimum, and the excavating energy consumption is lowest are achieved. According to the automatic control system and the optimal trajectory planning method of the intelligent excavator, the accuracy of a load predicting algorithm is improved, and accurate prediction on the excavating load is achieved.

The invention provides a wind power prediction method and system, and the method comprises the following step: (A), obtaining a historical data set which comprises the wind speed and wind power, and carrying out the preprocessing of the historical data set; (B), training an artificial neural network prediction model through the historical data set; (C), predicting the wind power based on the trained artificial neural network prediction model and the inputted wind speed data. According to the invention, the method employs the historical data comprising the wind speed and the wind power, achieves a wind power prediction model through a method combining a genetic algorithm and a back propagation neural network, achieves the precise prediction of the wind speed data, and improves the prediction efficiency of the wind power.

The invention relates to an environmental space robot heaven-earth teleoperation system with the complex structure. The system comprises a software system and a hardware system, wherein the software system comprises an on-satellite system, a ground measurement and control system, and a ground instruction generation and visual simulation system; and the hardware system comprises a six-degree-of-freedom air bearing table, a five-degree-of-freedom air bearing table, a multi-degree-of-freedom flexible operation mechanical arm, a binocular camera, a laser measuring sensor, a racemization and capturing tool, an Omega-7 operating handle and driving pedal, and an upper and lower controller. According to the system, a virtual reality technology is adopted, the dynamic three-dimensional environmentbetween the space target and the mechanical arm is dynamically and virtually restored in real time, information such as whether the mechanical arm intersects with the surrounding environment or not iscalculated, switching of different viewing angles of the dynamic three-dimensional environment can be achieved through a keyboard input mode, and a visual platform is provided for teleoperation control of the mechanical arm.

The present invention discloses a GPR modeling method based on kernel slow feature analysis and time delay estimation, which is applied to chemical processes with time delay and non-linearity. The method is characterized in that: by using fuzzy curve analysis, the delay information in industrial data is fully tapped to obtain the optimal delay in the data, and the reconstruction of the modeling data is carried out; the kernel slow feature analysis method is further used to carry out nonlinear feature extraction on the reconstructed data; and finally, based on the extracted features, a Gaussian process regression model is established to realize accurate prediction of the key variables, so as to improve product quality and reduce production costs.

The invention relates to a statistical forecast method and an apparatus for the urban heat island strength, which considers the influence of at least three weather factors such as horizontal total wind speed, wind direction and air temperature to the measuring result of the urban heat island strength, applies the historical observation data of the city to be researched for establishing a data analysis process apparatus, and obtains the numerical value forecast result of the specific weather factors such as horizontal total wind speed, wind direction and air temperature of the city to be researched by a numerical value mode so as to correctly perform the statistical forecast for the urban heat island strength and provide a certain technology support for the weather study and atmosphere science.

The invention discloses an air conditioner load short-term prediction method and system based on a sparrow optimization algorithm. Historical data of six factors including cold load, outdoor temperature, wet bulb temperature, relative humidity, solar radiation intensity and outdoor wind speed at different moments are selected as input variables; and a grey correlation degree analysis method improved by an entropy weight method is used for analyzing the weighted correlation degree between the input variables and the output variable of the air conditioner cooling load at the current moment, the input variables with the weighted correlation lower than 0.02 are removed, and the remaining variables are reserved. An SVM is established according to the number of the reserved input variables and the air conditioner cooling load at the current moment; then, the optimal hyper-parameter of the SVM is optimized by using a sparrow algorithm to obtain an SSA-SVM prediction model; and finally, load prediction is carried out on the SSA-SVM prediction model to obtain a prediction value. According to the method, the defect that the SVM depends on artificial experience to obtain the optimal hyper-parameter is overcome, the prediction precision of the SVM is improved, and the problem that the energy consumption is too high due to large prediction deviation of the cooling load of the air conditioner is solved.

The invention discloses a power battery health state and residual life prediction method based on big data, and the method comprises the steps: 1, obtaining a large amount of real-time driving condition data from an electric vehicle, and carrying out the data cleaning and filling; 2, extracting four working condition characteristic values including temperature, speed, current and mileage value by processing the discharge data to express equivalent cycle index increase; 3, processing charging data to obtain an accurate battery average capacity curve; 4, based on a battery charging and discharging characteristic experiment, obtaining a relation between SOH and equivalent cycle times, and establishing a battery health state evaluation model based on driving condition-charging calculation; and 5, constructing an integrated neural network machine learning model by taking the characteristic working condition as input and the difference of the equivalent cycle times as output, thereby realizing accurate estimation and prediction of the SOH of the battery.

The invention discloses a building operation energy consumption prediction method, which comprises the steps of preprocessing operation energy consumption data of an office building, performing feature analysis, and obtaining a time sequence of the operation energy consumption of the office building in a form of kilogram standard coal in a unified dimension; carrying out phase-space reconstructionon the time sequence of office building operation energy consumption by using a C-C correlation estimation method; judging whether the reconstructed office building operation energy consumption timesequence has chaotic characteristics or not by utilizing the maximum lyapunov index; performing short-term energy consumption prediction on the office building operation energy consumption time sequence with the chaotic characteristics by utilizing a Chaos-SVR neural network; dividing an error interval of a prediction result by using a mean-variance method, and constructing a Markov probability transfer matrix; carrying out Markov chain error correction on the Markov probability transfer matrix, and predicting the operation energy consumption of the office building according to the predictionvalue. According to the method, the office building operation energy consumption prediction precision is remarkably improved, and a decision basis is provided for optimized operation and energy-savingmanagement of the office building.

The invention discloses a method for predicting economic output through power consumption data, and relates to the technical field of power load prediction. At present, only a linear model is appliedfor quantitative analysis, and the accuracy is low. The method comprises the following steps: carrying out correlation inspection according to data between electricity consumption of each department and industrial development, and obtaining industrial development data related to the electricity consumption; inputting the obtained power consumption data into a first prediction model to obtain a first prediction value corresponding to industrial development; inputting the obtained power consumption data into a second prediction model to obtain a second prediction value corresponding to industrial development; judging a difference value between the first prediction value and the second prediction value, and correcting the first prediction model and the second prediction model when the difference value is greater than a threshold value; and when the difference value is smaller than the threshold value, taking the average value of the first predicted value and the second predicted value asa predicted value corresponding to industry development. According to the technical scheme, the nonlinear regression model is adopted, the two prediction models are comprehensively monitored, and theprecision and the accuracy are effectively improved.

The invention discloses a weighted Gaussian model soft sensor modeling method with time delay estimation, belonging to the field of complex industrial process modeling and soft measurement. The methodcomprises the steps of estimating process delay parameters by using a sliding grey relational degree algorithm and extracting process delay information; when a new sample arrives, reconstructing themodel sample based on the time delay parameters estimated in the offline phase; establishing a weighted Gaussian model to construct the joint probability density function of input and output variablesby the weight relative to the training sample; at last, employing that condition distribution function to estimate the value of the output variable in real time to predict the key variable accurately. In an intuitive and effective way and with the low computational complexity, the time delay information of variables extracted from the process history database is used for the soft measurement modeling data reconstruction, the actual causal correspondence between input and output is corrected, the interference of process random noise is effectively solved, and more accurate prediction results are obtained. Therefore, the product quality is improved and the production cost is reduced.

The invention belongs to the technical field of power transmission and distribution monitoring, and particularly relates to a power distribution network overhead line icing risk index prediction method. The method comprises the following steps that 1 a time sequence of an overhead line icing risk index evolution system is established; 2 a phase space of an overhead line icing risk index non-linear dynamic system is reconstructed; 3 a phase point, at a next moment, of the phase space is calculated; 4 the overhead line icing risk index prediction value is calculated. According to the power distribution network overhead line icing risk index prediction method, prediction is accurate; by establishing a power distribution network arrester non-lightning flashover risk index prediction model suitable for engineering practical application, accurate prediction of the arrester non-lightning flashover risk index is effectively realized; for the regions, where power distribution network natural weather conditions are severe and changeable, such as coastal regions and the regions along rivers, a complete and reliable prediction system can be realized, and the influences of severe environments are avoided.

The invention relates to an intelligent ammeter error online estimation method based on edge calculation. The comprises the following steps of step S1, acquiring the number information and the electric quantity information of an intelligent electric energy meter by the intelligent ammeter through a mobile terminal, establishing an ammeter data set, and transmitting the ammeter data set to an intelligent air switch for storage; step S2, enabling the intelligent air switch to acquire real-time data of each intelligent ammeter through the mobile terminal and calculating an ammeter error coefficient; step S3, enabling the intelligent air switch to store the calculated error coefficients of all the ammeters in a memory chip; and step S4, calling the data stored in the intelligent air switch byan intelligent ammeter error on-line analysis system, and storing the data in a database; and step S5, analyzing the database in real time by an intelligent ammeter error analysis terminal, and givingan alarm to remind the worker to check at the field if discovering that the continuous K-times of error analysis results of the same ammeter are displayed as tolerance. Through the method disclosed by the invention, the accuracy of the error prediction is greatly improved, an operation of calculating the electric energy error from the terminal is realized, and the precise prediction is realized.

The invention relates to a method for predicting performance indexes of a drafting link in a carbon fiber precursor production process, in particular to a method for predicting performance indexes of a drafting link in a carbon fiber precursor production process based on a least square support vector machine (LS-SVM) optimized by a particle swarm optimization (PSO) algorithm improved by a power law. The method comprises the following steps of selecting a six-stage draft ratio of the drafting link in the carbon fiber precursor production process as feature information, performing linear function normalization, and establishing an input sample dataset; determining main performance indexes, including linear density, precursor strength and breaking elongation rate, influencing carbon fiber quality, performing logarithmic function normalization, and establishing an output sample dataset; and building an LS-SVM model according to the input and output sample datasets, adopting a Gauss radial basis function (RBF) as a kernel function of the LS-SVM, and selecting an optimal penalty factor C and a kernel function parameter sigma by using PSO. The PSO process is improved according to the power law, so that the optimization speed can be greatly increased and accurate prediction is realized.

The invention relates to a time sequence running sample generation method of a distribution network in a garden. The time sequence running sample generation method comprises the following steps of S1,respectively performing cleaning energy main body output generation, energy storage main body output generation and load main body output generation; S2, generating running data of the distribution network in the garden; and S3, outputting a time sequence running sample of the distribution network in the garden, wherein the time sequence running sample of the distribution network in the garden comprises output powers of a wind power generation unit and a photovoltaic power generation unit, an output power of an energy storage unit, basic power demand quantity of a load main body, power demandquantity of a flexible load unit in the load main body and the running data of the distribution network in the garden. Compared with the prior art, the time sequence running sample generation methodhas the advantages that the analogue simulation on real-time running condition of the distribution network in the garden is achieved, scientific and reasonable practical sample data is provided for planning design, running control, transformation and investment of the distribution network in the garden, and the sample data volume deficiency or insufficiency in big data application relevant research in a traditional distribution network in the garden is made up.

The invention discloses a variable transmission ratio rack gear shaping force prediction method. The method comprises the following steps of S1, establishing a cutter gear tooth model, discretizing the cutter gear tooth model, and meanwhile, establishing a machined rack model in a workpiece coordinate system; S2, calculating a polar angle and a polar diameter from the discrete point of the rack tothe center of slotting cutters; S3, the discrete points are screened out to serve as cut points, and the number of all the screened discrete points is the size a of the cutting area; S4, judging thevertical position relation between the discrete points of all the slotting cutters and the machined surface of the machined rack; S5, screening out the discrete points at the same position on the machined surface of the machined rack, and the number of all the screened discrete points is the size b of the contact length; and S6, substituting the cutting area a and the cutting contact length b intothe metal cutting force model to finish the prediction calculation of the gear shaping force of the variable transmission ratio rack in each direction. The method provides a novel accurate calculation method for the cutting area and the cutting contact length, and accurate prediction of the gear shaping force of the variable transmission ratio rack is achieved.

The invention relates to a numerical control technology and a neural network technology, in particular to a tool wear detection method. The method comprises the following steps: firstly, collecting vibration and acoustic emission signals in different axial directions during milling of a high-speed cutter, and carrying out data preprocessing; then, adopting an improved 3-K-Means clustering algorithm to cluster three wear state intervals of the cutter, proposing a multi-selection multi-hidden-layer neural network structure to carry out feature learning on the three wear state intervals, and adopting Softmax for classification; and finally, performing parameter fine adjustment on the whole deep network by adopting stochastic gradient descent, and establishing a tool wear detection model. Experimental results show that the accuracy rate of the method provided by the invention on tool wear detection is as high as 95%.

The invention discloses a microtopography icing value prediction method and system. The method comprises the steps: carrying out the classification of microtopography according to the elevation data of a to-be-predicted line tower section and the surroundings of the to-be-predicted line tower section; the classification comprises water body microtopography and non-water body microtopography, and the non-water body microtopography comprises canyons, terrain uplifts, bealocks and high mountains; according to the type of the microtopography, calculating a numerical calculation result of the area with the kilometer-level resolution by adopting a corresponding numerical mode and a parameter scheme, and then interpolating the numerical calculation result of the kilometer-level resolution into the transition layer to extract an initial field and a boundary field of calculation of the microtopography layer; according to the calculated initial field and boundary field, calculating physical parameters of the microtopographic layer; According to the icing model corresponding to the microtopography type, inputting physical parameters of the microtopography layer to obtain the icing thickness under the microtopography of the corresponding type. According to the method, kilometer-level numerical calculation and fluid calculation are combined, and accurate prediction of micro-terrain icing is achieved.

The invention discloses a regional radiative traffic flow prediction method, and the method is characterized in that the method comprises the following steps: S1, collecting the movement track data information of residents in daily life, and converting the track data into a traffic in-out flow graph taking a sub-region as a minimum unit; S2, acquiring traffic flow characteristics of different regions of a city and spatial correlation of flow among different regions of the city; S3, constructing a regional radiative traffic flow prediction model, and adopting a specific full convolutional network to realize gradual expansion prediction by superposing feature extraction-prediction sub-modules by taking a central region as a starting point; and S4, reversely adjusting the full convolutional neural network through the error between the predicted flow diagram and the actual flow diagram, and fitting a regional radiative traffic flow prediction model close to a real condition. According to the method, the problem that prediction is continuously expanded to the peripheral area based on the flow of the central area is solved, the prediction time period is shortened, and real-time prediction of the traffic flow is realized.

The invention discloses a method and device for predicting a destination and a storage medium, and the method comprises the steps: obtaining historical order data, and generating training set data with a label according to the historical order data; constructing a plurality of decision trees according to the training set data, and generating a plurality of multi-feature combinations based on the plurality of decision trees; screening the multi-feature combination of which the prediction accuracy is higher than a first threshold according to the prediction effect of the decision tree; and classifying the users based on the multi-feature combination of which the prediction accuracy is higher than a first threshold, the corresponding decision tree and a set threshold, and performing personalized destination prediction according to a classification result. According to the invention, personalized prediction of the destination can be carried out, and the destination prediction effect is improved while the labor cost is saved.