45results about How to "Forecast in time" patented technology

The invention belongs to the field of intelligent transportation, and in particular relates to a method for predicting urban traffic flow based on time-space correlation. The method includes the following steps: 1) forecasting model training: generate corresponding forecasting models according to different time forecasting granularities; 2) real-time traffic forecasting: consistent with the model training process, the latest collected traffic will be used for traffic flow forecasting The data is added to the input end of the forecasting model, and then processed by the forecasting model to output the forecasted traffic for the next period. The invention accurately predicts urban road traffic flow, can realize intelligent traffic control and management, traffic information service, and provides real-time data for alleviating urban traffic congestion, and has significant social and economic benefits.

The invention discloses a method for predicting ground subsidence caused by metro construction based on neural network horizontal comparison. The method is characterized in that the relation between a horizontal distance from a tunneling face to a ground monitoring point and the ground subsidence at the position of the ground monitoring point is researched. A neural network is applied, rock and soil mechanical parameters of existing tunnels are used as input values, and the ground subsidence amount at the position of the ground monitoring point serves as an output value for training when being located at different positions of the horizontal distance from the tunneling face to the ground monitoring point. A trained network is used for analyzing others, namely the ground subsidence situation above the constructed tunnels. The method mainly comprises the steps of relevant data preparation, process simulation and result prediction and accuracy testing. By adopting the method, the ground subsidence amount at the position of the ground monitoring point can be effectively predicted according to the rock and soil mechanical parameters of the existing tunnels. The method can be widely applied to the construction process, and a measurement basis can be provided for ground building safety and abnormal subsidence prevention.

A regeneration water factory effluent residual chorine risk prediction method comprises: data monitoring and collection are performed so that an index monitored and collected is obtained; the data is transmitted to a server through a communication system; effluent residual chorine risk prediction, analysis and decision are performed by the server according to the collected data: the corresponding monitoring data stored in a data base of the server is read, the monitoring data is input in a water quality model and the effluent residual chorine risk prediction of technology is performed; the index influencing effluent residual chorine risk probability is judged according to a potential of hydrogen (pH) value, the water temperature, ammonia concentration and chemical oxygen demand distribution situation monitored under different risk probabilities; and solutions under different effluent residual chorine risk probabilities are made. The regeneration water factory effluent residual chorine risk prediction method is good in model prediction accuracy (the accuracy is higher than 95%) so that when facing water quality change, a researcher can predict effluent residual chlorine risk accurately in time. The regeneration water factory effluent residual chorine risk prediction method offers a certain reference to operation and management of an actual water factory.

The embodiment of the invention provides a target user prediction method and device, a background server and a storage medium. The method comprises the steps of determining user characteristics of a to-be-tested user, wherein the user characteristics at least comprise user portrait data of the to-be-tested user, behavior characteristics of the to-be-tested user and friends in each first type application of a plurality of first type applications, and behavior characteristics of the to-be-tested user in at least one non-first type application; generating prediction features of the to-be-tested user according to the user features of the to-be-tested user and the application features of the target first type application; and inputting the prediction features of the user to be tested into a pre-constructed target user prediction model, and predicting through the target user prediction model to obtain the probability that the user to be tested is a target user of the target first type application. According to the embodiment of the invention, the prediction accuracy of the target user can be improved, and the timely prediction of the target user and the prediction of the target users ofdifferent first type applications are realized.

The embodiment of the invention provides a quantitative prediction method and device for indicators of wireless network coverage. The method comprises the steps of: obtaining feature data of a wireless network, wherein the feature data includes network structure, wireless parameter data, topographical data, and service distribution data; and inputting the feature data into a trained deep learningmodel to obtain quantitative prediction data of the indicators of the wireless network coverage. According to the embodiment of the quantitative prediction method and the device, the quantitative prediction data of the indicators of the wireless network coverage can be obtained by analyzing and calculating the feature data of the wireless network using the trained deep learning model. The embodiment of the quantitative prediction method and the device includes all the details of the current scene as much as possible, which includes the network structure, the wireless parameter data, the topographical data, and the service distribution data, thereby maximizing the comprehensive analysis of the current scene, and enabling accurate, quantitative, customer-perceived, and timely prediction of the optimized indicators.

The invention relates to a method for predicting the crystallization state of the sugarcane sugar boiling process. The method comprises the following steps of: establishing an original decision table by taking six factors comprising massecuite brix, massecuite temperature, vacuum degree, feeding flow, steam temperature and steam pressure as condition attributes and taking crystallization state level as a decision attribute, and performing discretization treatment on the original decision table; extracting a prediction rule of the crystallization state according to a rough set theory and storing the prediction rule of the crystallization state into an expert system knowledge base; performing rule matching by using an inference machine of an expert system according to the real-time value of the condition attributes and the existing rule of the knowledge base, and outputting the predicting result of the crystallization state; establishing an on-line crystallization state prediction learning model based on a support vector machine, and adding the predicting result of the crystallization state into the expert system knowledge base to realize on-line updating of the expert system knowledge base. The invention also provides a system for realizing the method. The system can accurately predict the crystallization state of the sugarcane sugar boiling process so as to effectively improve the automation level of the sugarcane sugar boiling process.

The invention discloses a thermal power generating unit variable-load speed predicting method based on a wavelet neural network. By establishing the wavelet neural network, timely, effective and active prediction for thermal power generating unit variable-load speed is achieved, prediction method is highly intelligent and prediction precision is quite high.

The invention discloses a cement finished product specific surface area prediction method and system. The method comprises the following steps: obtaining to-be-tested cement finished product data, wherein the to-be-tested cement finished product data comprises granularity data of the to-be-tested cement finished product at the current moment and a specific surface area experimental value of the to-be-tested cement finished product at the previous moment; the granularity data comprises a plurality of granularity values; the particle size values belong to different particle size ranges; inputting the to-be-tested cement finished product data into the trained specific surface area prediction model to obtain a specific surface area prediction value of the to-be-tested cement finished product,wherein the trained specific surface area prediction model is determined through a convolutional neural network algorithm and a back propagation algorithm. The method solves the problem of time-varying delay between the variable data and the to-be-measured index in the cement finished product specific surface area prediction process, and is low in measurement cost.

The embodiment of the invention discloses an energy consumption prediction method and device. A specific embodiment of the method comprises the steps of obtaining historical energy consumption data, historical energy consumption environment data corresponding to the historical energy consumption data and to-be-predicted energy consumption environment data corresponding to the to-be-predicted energy consumption data; and inputting the obtained historical energy consumption data, the historical energy consumption environment data and the to-be-predicted energy consumption environment data into apre-trained energy consumption prediction model to obtain to-be-predicted energy consumption data. The embodiment can be applied to the field of cloud computing, enterprise production conditions areinformed in real time by automatically collecting enterprise energy consumption data, and enterprise production condition prediction is carried out in time.

The invention discloses a neural network-based method for analyzing surface subsidence caused by metro excavation. The method is characterized in that a relationship between a horizontal distance from an excavation face to a ground monitoring point and surface subsidence at the monitoring point is studied; training is carried out by taking rock and soil mechanics parameters of each existing metro as input values and taking surface subsidence quantities at detection points in different positions of the horizontal distance between the excavation face and the monitoring point as output values through applying a neural network; surface subsidence situations above other metros that are about to be constructed are analyzed by using the network after the training; and the method mainly comprises related data preparation as well as simulation process prediction result and accuracy check. According to the method, the surface subsidence quantities at the detection points can be effectively predicted according to the rock and soil mechanics parameters of each existing metro. The method can be widely used in a metro excavation process and provides measurement bases for ensuring safety of ground buildings and preventing non-normal subsidence.

The invention discloses a gear box fault prediction, monitoring and diagnosis method and a corresponding device. The method comprises the steps: collecting working data of a gear box and train operation data; and carrying out intelligent analysis processing on the working data of the gear box and the train operation data, and predicting and obtaining fault data of the gear box. According to the invention, the technical problem that the fault of the train gearbox cannot be accurately predicted in time is solved.

The invention discloses a big data-based water conservancy modular monitoring system comprising a data acquisition end, a data storage and processing end, a monitoring application end and a system management end; the data acquisition end is used for acquiring the data of all monitoring equipment of a water conservancy modular management network and data of investigation, research, measurement anddetection of each related management mechanism, the data storage and processing end is used for storing and processing the big data acquired by the data acquisition end. The invention further discloses a big data-based water conservancy modular monitoring method. According to the system and the method, the monitored content is a collection of the monitored big data of a natural unit, a facility unit, a node unit and a functional area unit and covers the monitored data of a plurality of index systems including the hydrology, the water quality, the water ecology, the engineering quality safety standard and the like; and full coverage, comprehensiveness and science of the big data-based water conservancy modular monitoring system are achieved.

Provided are a product performance prediction modeling method and apparatus, a product performance prediction method, a product performance prediction system, a computer device, and a storage medium.The product performance prediction modeling method includes: acquiring first sample data, wherein the first sample data includes device outlier data generated in a process of manufacturing a product by a device; acquiring a production line configuration simulation parameter of a production line where the device is located, and product information of the product manufactured by the production line;selecting a simulation model to perform simulation test on the performance of the product, so as to obtain product performance simulation data; and inputting the device outlier data, the production line configuration simulation parameter, the product information and the product performance simulation data into a machine learning model to perform machine learning training, so as to obtain a product performance prediction model. The foregoing product performance prediction modeling method and apparatus, product performance prediction method, product performance prediction system, computer device and storage medium can accurately predict product performances during device exception.

The invention belongs to the technical field of mine dynamic water grouting amount prediction and evaluation, and relates to a coal mine water burst dynamic water grouting amount prediction and grouting effect evaluation method. According to the method, on the basis of collection of dynamic water grouting amount data sample materials during mine water burst blocking of diggings, factors influencing grouting amount change are analyzed, a genetic algorithm and a support vector machine learning theory are integrated, and a non-linear data model of the dynamic water grouting amount and influence factors thereof is created and applied to practical engineering projects; the method refers to a evaluation criterion method in which the effect of blocking water burst points by grouting is graded as excellent, good, qualified and unqualified according to the two influence factors including a post-grouting water burst blocking ratio and a grouting prediction ratio; prediction and evaluation processes are simple, the principle is scientific, manpower and finical resources are saved, the mine water burst dynamic water grouting amount can be predicted timely, grouting effects can be evaluated reasonably, and economic benefit is remarkable.

The invention provides a reservoir water and sediment prediction model training method and device, and a reservoir water and sediment prediction method and device. The method comprises the following steps: obtaining reservoir water inflow data; inputting the reservoir water inflow data into a pre-established mathematical model to obtain water and sediment data of a target area, wherein the mathematical model is established by using historical actual measurement data, the historical actual measurement data comprises water flow inflow data and on-way water and sediment data, the water flow inflow data is obtained through upstream water flow inflow observation stations, and the on-way water and sediment data are obtained through downstream observation stations; taking the reservoir water inflow data and the water and sediment data of the target area as training data, and training an initial neural network model through the training data to obtain a water and sediment prediction model of the target area. By implementing the method, water and sediment data of any target area can be quickly and timely predicted, and an important reference can be provided for real-time and fine scheduling of a reservoir.

The invention discloses a financial risk prediction device based on real-time data and a use method. The financial risk prediction device comprises a bottom plate, wherein two sides of the top surface of the bottom plate are fixedly connected with support pipes, and lifting notches are vertically formed in the centers of the front surfaces of the support pipes. Through cooperative use of the infrared signal receiver, the handheld control terminal, the central processing unit, the data receiver, the storage module, the data risk analysis module, the data model drawing module, the screen recording module, the data risk comparison module, the data risk library and the data risk alarm module, financial risks can be predicted in real time; thus, prediction of the prediction device is more timely, and problems that in the financial risk prediction process of the prediction device, due to the fact that real-time data cannot be collected and processed in time, data acquisition and prediction delay is caused, financial risks cannot be known in time, and the prediction device cannot predict the financial risks in time are solved; and huge loss is brought to the user.

The invention discloses a traffic control method and device and electronic equipment. The method comprises the steps of obtaining travel trajectories generated when a plurality of traffic objects movefrom a target departure area to a target arrival area; determining a target track meeting a first preset condition according to the travel track; determining a traffic mode corresponding to the target track as a target traffic mode; and carrying out traffic control on a target path corresponding to the target track according to the target traffic mode. The traffic experience of the traffic objectin the process of moving from the target departure area to the target arrival area can be improved.

The invention relates to an enterprise risk conduction prediction method and device based on graph features and a storage medium, and the method comprises the steps: 1, building enterprise associationgraph data based on bank data, and storing the enterprise association graph data in a graph database; 2, extracting enterprise-enterprise node pairs used as enterprise risk conduction prediction object samples from the graph database; 3, performing column division on the graph mode characteristics of the corresponding data for enterprise-enterprise nodes, and obtaining a risk conduction edge weight by utilizing logistic regression; 4, constructing a characteristic variable of a LightGBM algorithm model for the corresponding topological structure based on the enterprise-enterprise node pairs,and outputting to obtain each unilateral conduction probability result after the model is trained; 5, performing multilateral probability fusion on each unilateral conduction probability result to obtain a risk conduction integration probability result; and 6, performing logistic regression on the risk conduction integration probability result to obtain a final enterprise risk conduction prediction result. The method has the advantages of accurately predicting enterprise risks and the like.

The invention relates to the technical field of computers, and provides an emergency prediction method and a device, a storage medium and a server. The prediction method comprises the following steps:acquiring a monitoring video of a public place; detecting portraits contained in each frame of image in the monitoring video and positions of the portraits in each frame of image; performing posturerecognition on each portrait contained in each detected frame of image to obtain the posture of each portrait; according to the positions of the portraits in the continuous multi-frame images of the monitoring video, respectively drawing advancing routes of the portraits; and inputting the position of each portrait in each frame of image, the posture of each portrait and the advancing route of each portrait into a pre-constructed emergency prediction model to obtain the probability of generating various emergencies. According to the method, emergencies can be predicted in time, and the life and property safety of personnel can be protected.

The invention relates to a steel multi-variety demand prediction method based on an intelligent supply chain. The method comprises the following steps: 1), obtaining a demand data time sequence of steel multi-variety based on industrial product production data of an intelligent supply chain system; 2) constructing an SARIMA time sequence model based on the demand data time sequence of multiple varieties of steel; 3) inputting a to-be-tested training sample set to the SARIMA time sequence model, and obtaining a steel multi-variety demand prediction result; 4) constructing an SARIMAX time sequence model based on the demand data time sequence of multiple varieties of steel; 5) inputting a to-be-tested test sample set to the SARIMAX time sequence model, and obtaining a demand quantity prediction result of multiple varieties of steel after algorithm correction; and 6) obtaining a final prediction result according to a result output by the SARIMA model and a correction result output by the SARIMAX model. Compared with the prior art, the method has the advantages that the time sequence is more stable, the prediction accuracy is higher and the like.

The invention is suitable for the technical field of data processing, and provides a household appliance maintenance order prediction method, a device and terminal equipment. The method comprises the steps: obtaining and processing historical household appliance maintenance order data; carrying out eMD decomposition on the processed historical household appliance maintenance order data to obtain a first component and a residual component, and VMD decomposition is carried out on the first component; obtaining final data according to the residual component and the decomposed first component; dividing the final data into a feature set and a label set, dividing the feature set into a feature training set and a feature test set, and dividing the label set into a label training set and a label test set; constructing and training a to-be-trained order prediction model to obtain a household appliance maintenance order prediction model; and using the home appliance maintenance order prediction model to predict the home appliance maintenance order quantity in a preset time period. According to the invention, accurate data can be provided for the household appliance maintenance manufacturer, and the household appliance maintenance manufacturer schedules the maintenance workers according to the predicted household appliance maintenance order pre-quantity so as to complete the order in time.

The invention discloses a traffic congestion prediction method, equipment and medium based on multiple signal sources. The method includes: obtaining information on each road section on the road network and various training data; performing corresponding preprocessing on the training data according to the data type; Input the preprocessed training data into the congestion prediction model, determine the connection weight on the unit path of the congestion prediction model according to the prediction error and the state distribution of the congestion prediction model; use the trained congestion prediction model to perform congestion prediction on the prediction data; The correlation test is carried out for the adjacent two congestion prediction results. The present invention acquires various types of signal sources as prediction input values, uses neural network to learn multi-dimensional time series, obtains accurate prediction results through multi-signal source comparison, backpropagation and forward propagation adjustment, and timely predicts traffic information, Provide a reference for urban road network planning and urban congestion management.

The invention relates to a cemented filling body damage evaluation method, which comprises the following steps of: filling body sample measurement stress-strain, acoustic emission ringing counting and accumulated ringing counting experimental data in a metal mine are obtained, stress-strain, time-ringing counting and time-accumulated ringing counting curves are drawn, the damage rate D and the damage rate D'of the filling body are defined according to the sound emission ringing count of the filling body, when the value of D 'is gradually increased from 0 to a stage 1, the stage is an internal fine crack rapid increasing stage, when D' = 1, the initial stage is an elastic-plastic stage, after D '= 1, D' is gradually reduced from 1 to 0, the value of D 'is reduced to be smaller than or equal to 0.6, and when D is larger than 0.8, it is indicated that the filling body is developed to the damage stage, and the interior of the filling body is seriously damaged. According to the method, the damage rate D'is defined through sound emission ringing counting to judge the damage speed in the filling body, the damage stage of the filling body is further judged, the damage speed in the filling body can be accurately and timely pre-judged to achieve safe production, and the method can also be used for filling body-surrounding rock early warning in actual engineering.

The invention provides a company event risk prediction method and device, a storage medium and electronic equipment, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining news information containing target company information; based on the news information, updating an existing knowledge graph reflecting the target company information to obtain an updated knowledge graph; and performing risk prediction on the target company based on the updated knowledge graph and a preset risk prediction model to obtain a risk prediction result. According to the technical scheme provided by the invention, the risk of the target company can be automatically, timely and accurately predicted, so that the requirements of wind investment personnel are met.

The invention discloses a fault positioning method suitable for electric power equipment, and the method comprises the steps of building a forward modeling calculation model based on a multi-physical field simulation strategy, and carrying out the forward modeling simulation calculation of each single physical field in the electric power equipment through the forward modeling calculation model; performing inversion calculation according to observable data of the transformer acquired by the monitoring system to obtain loss distribution in the transformer; constructing a power equipment inversion model according to the forward modeling calculation model and the loss distribution; inputting the surface temperature of the transformer into the power equipment inversion model, and performing multi-physical field multi-parameter inversion optimization on the multi-point temperature of each component in the transformer by adopting a genetic algorithm to obtain a temperature field parameter distribution condition in the transformer; performing fault positioning on the electric power equipment according to the temperature field parameter distribution condition in the transformer. According to the invention, the fault abnormity of the power equipment can be accurately predicted in time, and an important basis is provided for operation and maintenance personnel to deal with abnormal phenomena.

The invention discloses a method for predicting the variable load rate of a thermal power unit based on a wavelet neural network. By establishing a wavelet neural network, the timely, effective and active prediction of the variable load rate of a thermal power unit is realized. High precision.

The present invention provides a reservoir water and sediment prediction model training, reservoir water and sediment prediction method and device, wherein, the reservoir water and sediment prediction model training method includes: obtaining inbound water flow data; inputting the inbound water flow data into a pre-established mathematical model, To obtain the water and sediment data in the target area, the mathematical model is established using historical measured data. The historical measured data includes water inflow data and water and sediment data along the way. Acquisition; the inbound water flow data and the water and sediment data in the target area are used as training data, and the initial neural network model is trained through the training data to obtain the water and sediment prediction model in the target area. By implementing the invention, the water and sand data of any target area can be quickly and timely predicted, and an important reference can be provided for real-time and fine scheduling of reservoirs.

The invention discloses a vehicle window fogging prediction method and system, electronic equipment and a storage medium. The method comprises the steps: predicting a corresponding real-time fogging parameter according to an obtained feature parameter, and the feature parameter comprises at least one of a vehicle parameter, a personnel parameter, an environment parameter and a road condition parameter; predicting a future fogging parameter according to the real-time fogging parameter; and judging whether the vehicle window fogs in the future or not according to the future fogging parameters. According to the invention, no additional equipment is needed, the characteristic parameters can be directly obtained through the existing equipment in the vehicle, and the real-time fogging parameters capable of reflecting the temperature and humidity of the vehicle window can be predicted through the directly obtained characteristic data. The future fogging parameter reflecting the future fogging trend can be predicted through the real-time fogging parameter, so that the future fogging condition of the vehicle window can be predicted timely and accurately.

The embodiment of the invention provides a preparation method of an intelligent indication label and a residual shelf life prediction method and device. The residual shelf life prediction method comprises the steps that an intelligent indication label image is collected, and the storage temperature of a to-be-detected product is obtained; wherein the color of the intelligent indication label changes along with the storage of the to-be-detected product; a redness value of the intelligent indication label is obtained according to the image, and a residual shelf life prediction model of the corresponding to-be-detected product is obtained according to the storage temperature; and the remaining shelf life of the to-be-tested product is obtained according to the redness value and the remainingshelf life prediction model. The residual shelf life prediction method and device provided by the embodiment of the invention obtain the redness value of the intelligent indication label by collectingthe image of the intelligent indication label, the remaining shelf life is obtained according to the redness value and the preset remaining shelf life prediction model, the automation level of remaining shelf life prediction is improved, time and labor are saved, and rapid, timely, lossless and low-cost remaining shelf life prediction is achieved; and the preparation of the intelligent indicationlabel based on the anthocyanin is realized.