121results about How to "Real-time forecast" patented technology

The invention discloses a microgrid energy management system capable of realizing load management. The energy management system comprises a man-machine interaction module, a prediction module, a system analysis module, a dispatching and control module, and a system management module. The system performs collection and monitoring on information of the output power of a distributed power generation unit (including a wind turbine generator, a photovoltaic array and a diesel engine set), the charging-discharging states of a storage battery, the load power, the electric energy quality information of unit nodes, and the like, then carries out data analysis and processing under the condition of satisfying electrical characteristic constraint conditions of operating conditions and physical equipment, formulates a corresponding optimal control strategy according to the current microgrid operating condition and external application requests, coordinates and manages the operations of the distributed power supply, the load and other modules in the microgrid system, optimizes the energy flow and utilization of the microgrid system, outputs by using renewable resources, supplies electric energy capable of satisfying the required quality to user load at the lowest operating cost, and realizes the system optimization and dispatching.

A rail transit passenger flow predicting method for predicting passenger travel probability and based on a support vector machine (SVM) includes the following steps: 1 collecting rail transit historical data including a starting station and a destination station of a passenger travel, station entering time and station leaving time; 2 acquiring passenger travel proportion in a statistics mode based on the historical data; 3 training the least square SVM according to the travel proportion obtained by statistics to predict the passenger travel probability; 4 storing the predicted travel proportion for a real-time passenger flow prediction module; 5 collecting real-time station entering passenger flow data which is used as a set of passenger station entering records; 6 acquiring the passenger travel probability at the station and stored in the step 4 and predicting the destination station of the passenger travel; 7 simulating the passenger travel by combining the departure interval of trains, calculating the time when the passengers reach and leave each station and updating full-road-network passenger flow. By means of the method, prediction is conducted by utilizing the passenger travel law, the station entering passenger flow can be predicted in real time, and prediction accuracy is high.

The embodiment of the invention provides a vehicle auxiliary driving system. The vehicle auxiliary driving system comprises a vehicle terminal, a road side unit and a central control system installed on a vehicle, wherein the vehicle terminal comprises a data acquisition module, a data processing module, a first communication module and a second communication module. First vehicle travelling data of the vehicle is collected by the data acquisition module, travelling environment information is collected by the first communication module, and the first vehicle travelling data and the travelling environment information are sent to the data processing module. The data processing module generates operating instructions according to the first vehicle travelling data and the travelling environment information, and sends the operating instructions to the central control system. The central control system controls driving of the vehicle based on the operating instructions, achieving the goal of controlling the vehicle to drive according to the first vehicle travelling data and the travelling environment information. The vehicle auxiliary driving system can real-timely predict and warn probable dangers existing in the current driving environment, thus improve safety during vehicle driving.

The invention relates to a ship conflict early warning method. The ship conflict early warning method comprises the following steps that the real-time position information and the historical position information of a ship are obtained through maritime radar and are processed primarily; ship trajectory data are preprocessed at each sampling moment, clustering is conducted on the ship trajectory data at each sampling moment, parameter training is conducted on the ship trajectory data at each sampling moment by means of a hidden Markov model, the hidden state q corresponding to an observation value at the current moment is obtained at each sampling moment according to parameters of the hidden Markov model by means of the Viterbi algorithm, a predicted value of the position of the ship in the future time period is obtained at each sampling moment through a set prediction time domain W based on the hidden state q at the current moment, the dynamic behaviors of the ship are monitored in real time by establishing a safety rule set, and a warning message is sent out in time. According to the ship conflict early warning method, the trajectory of the ship is predicted in real time in a rolling mode, effective early warning of maritime conflicts is achieved, and the safety of marine traffic is improved.

The invention relates to a real-time weather forecast system and a real-time weather forecast method using a rain sensing device, wherein rainfall information of a specific area is collected through the rain sensing device and is transmitted to the system to be analyzed so as to work out a relatively real-time weather forecast; the real-time weather forecast system using the rain sensing device comprises the rain sensing device, a vehicle-mounted control module, a vehicle-mounted display screen, a system data center and a mobile terminal device, wherein a signal input end of the vehicle-mounted control module is connected with a signal output end of the rain sensing device; a rainfall signal sensed by the rain sensing device is transmitted to the vehicle-mounted control module; the vehicle-mounted control module is in data communication with the system data center through a communication channel; and the system data center is used for transmitting the processed data to the mobile terminal device or the vehicle-mounted display screen connected with the system through the data information channel, so that the purposes of monitoring and issuing real-time weather conditions are achieved.

The invention discloses an air conditioner load control system based on personnel information detection. The air conditioner load control system comprises a BP neural network module used for completing training and predicting the air conditioner load quantity and further comprises a personnel information detection module used for detecting personnel information within the air conditioner working range. The personnel information comprises the personnel types and the corresponding quantities. The BP neural network module is provided with a corresponding relation between an air conditioner load and the personnel information. The invention further discloses an air conditioner load control method based on personnel information detection. According to the air conditioner load control system and method based on personnel information detection, compared with other personnel number acquisition technologies used for air conditioner load predicting, the personnel load requirement difference can be better analyzed, and the air conditioner load control system and method have very important significance for precise load management and control, internal personnel thermal comfort strengthening and comfortable artificial environment creating.

The invention provides an aero-engine main bearing residual life prediction method based on digital twinning. The method comprises: firstly, using a plurality of restricted Boltzmann machines and a regression algorithm to construct a main bearing health monitoring model; comparing the main bearing actual measurement vibration signals with main bearing health state information extracted from the digital twinning model, adjusting and correcting the digital twinning model by utilizing a comparison result, and finally predicting the residual life of the main bearing by utilizing the updated digital twinning model. According to the aero-engine main bearing residual life prediction method based on digital twinning, a digital twinning technology is introduced into the field of main bearing residual life prediction, so that a main bearing digital twinning model applied to the method can be updated in real time along with the change of the working condition of the main bearing of the aero-engine, and a more accurate residual life prediction result can be obtained.

The invention discloses a system and a method for on-line control of a molten steel temperature in LF refining and belongs to the technical field of metallurgy production and control. The system comprises an information acquisition module, a heating judgement module, an electrode heating module, a steel ladle lining heat-dissipation module, an added alloy judgement module, an alloy heat effect module, an added slag judgement module, a slag heat effect module, an argon-blown slag layer judgement module, a molten steel thermal-radiation module, an argon heat-absorption module, a slag layer heat-radiation module, a temperature calculating module, a temperature correction judgement module, a temperature correction module, a temperature forecast module and a temperature control module. The method comprises calculating molten steel temperature variable quantity produced by heating and heat radiation cooling processes in LF refining to obtain a molten steel real-time temperature, controlling the heating process by adjustment of heating time so that the molten steel real-time temperature reaches a molten steel target temperature, and controlling the molten steel real-time temperature in an allowable accuracy range in the LF refining scene.

The invention discloses a concrete crack detector with a reinforcing bar as an electrode. A reinforcing bar emission electrode and a reinforcing bar receiving electrode are embedded in concrete, wherein the emission electrode is connected with an emission circuit through an emission cable; a signal transmitted by the emission circuit is sent out from the concrete through the emission electrode; the receiving electrode is connected with a receiving circuit through a receiving cable; and the receiving circuit is used for amplifying and filtering a signal received by the receiving electrode. An output signal of the emission circuit is connected to a control and computation module through an emission communication control interface, and two-way communication control is performed between the control and computation module and the emission circuit. An output signal of the receiving circuit is connected to the control and computation module through a receiving communication control interface, and two-way communication control is performed between the control and computation module and the receiving circuit. As the reinforcing bar is used as the electrode, the original construction features and stress features of the concrete do not change, a construction process and a construction method have no need to change, and the electrode and the concrete structure have the same service life. The concrete crack detector with the reinforcing bar as the electrode has the advantages of simplicity in design, convenience for construction, low cost and long service life.

The invention relates to a method for constructing a near-infrared model for predicating the attribute of crude oil, and a method for rapidly predicting the attribute of the crude oil based on near-infrared spectroscopy. The model construction method includes the following steps: detecting the attributes of different types of crude oil samples; determining the near-infrared spectrograms of the crude oil samples; and carrying out first-order differential processing and multiplicative scatter correction on the absorbance of the 12500-4000 cm<-1> spectral region in the near-infrared spectrograms;carrying out main component analysis on preprocessed spectra, calculating the T2 value of every crude oil sample, rejecting the values being greater than a threshold, and establishing an initial training set; selecting samples by using the Euclidean distance between the samples with the obtained main component score in the initial training set as a characteristic variable to determine a final training set; selecting one or more wave-number segments by using a moving window partial least squares technology; and establishing the near-infrared model between the attribute and the near-infrared spectrum of the crude oil by using a regression technology. The methods are especially suitable for the online fast prediction analysis of the crude oil.

The invention discloses a vehicle-mounted terminal multi-target recognition tracking prediction method, which is a vehicle-mounted terminal multi-target recognition tracking prediction method based onYOLOv5s (You Only Look Once v5s) and FairMOT (FairMOT) fusion, and detects vehicles in front of a road quickly and accurately in real time by using a YOLOv5s deep learning object detection technology. The YOLOv5s model is fused into the FairMOT architecture detection module to perform target detection and re-identification tracking in a single network so as to realize position detection, type identification and multi-target motion trajectory tracking of the traffic target in front of the vehicle on the road, thereby achieving prediction of driving behaviors of lane change, following, deceleration and the like of the traffic target in front of the vehicle.

The invention discloses an airport traffic flow prediction method based on weather data mining, which comprises the following steps: extracting required weather information from the history data of aselected airport according to preset rules, and identifying the weather information according to preset conditions; Obtain the airport flow statistics data in a specific period of time of the selectedairport, and match the airport flow and weather information, in which, the airport flow is divided into take-off, landing and flying three types; Based on BP neural network, the matching informationof airport traffic is trained and learned respectively, and the influence relationship of different weather types on airport traffic is obtained. The influence relationship is analyzed and verified, and the corresponding BP neural network model is finally obtained. Finally, based on the trained BP neural network model and the forecast weather information, the airport's inflow and outflow are classified and forecasted. The method can effectively solve the normal operation of flights in complex weather and improve the air traffic safety level and efficiency of the terminal airspace.

The invention discloses a ship navigation track prediction method based on a ship type. The method comprises the steps of obtaining AIS data of a to-be-predicted ship, wherein the AIS data comprises anavigation track point sequence and a ship type of the to-be-predicted ship; converting the navigation track point sequence into a sea area grid number sequence according to pre-divided sea area grids; determining a ship type feature vector of the ship to be predicted according to the ship type of the ship to be predicted and a pre-established ship type dictionary matrix; inputting the sea area grid number sequence and the ship type feature matrix into a pre-trained sequence to a sequence model to obtain a prediction result sequence; and determining longitude and latitude coordinates of a central point of each sea area grid in the prediction result sequence to obtain a prediction track point sequence of the to-be-predicted ship. According to the method, the prediction precision of the ship navigation trajectory can be improved, and different trajectory prediction results can be generated for different types of ships.

The invention discloses a real-time effluent total-phosphorus monitoring system based on a peak radial basis function neural network, and belongs to the water field and the data detection field. Sewage treatment is a complicated biochemical reaction process with various variables; in order to improve operation efficiency of sewage treatment devices, guarantee effluent quality and reduce operation cost, the real-time effluent total-phosphorus monitoring system mainly comprises a real-time sewage treatment data acquisition system, an FAMEVIEW data monitoring system and a peak radial basis function neural network online measuring system. Data are acquired and uploaded to a monitoring interface, and water quality and operation conditions are monitored in real time; the data are transmitted to a radial basis function neural network module in background operation so as to provide the basis for water quality predication. The real-time effluent total-phosphorus monitoring system has the advantages that predicated values of water quality parameters are acquired through data monitor and real-time analysis, so that the water quality can be predicated, potential water pollution can be reduced by measures such as manual medication and aeration, the learning algorithm is applied to an actual sewage treatment process, and an application platform is provided for predication control methods.

The invention relates to an online support vector regression (SVR) method, in particular to a segmental online SVR method applied in traffic prediction. For the online SVR algorithm is difficult to realize both the prediction accuracy and operation efficiency, the invention provides a method for predicting segmental online SVR time series. The method comprises the following steps: realizing rapid training by reducing the data length of the online modeling; carrying out the segmental storage on the online SVR; selecting and outputting the optimal sub-segmental model prediction according to the predicted matching degree between the neighborhood samples and each sub-segmental SVR model, thereby improving the prediction accuracy. Compared with the common online SVR algorithm, the algorithm of the invention can improve the prediction accuracy by 5% to 10% while maintaining the execution efficiency of the online prediction; by adopting the segmental strategy and shorter modeling data length, the efficiency of the algorithm is high; the invention is capable of online, real-time and rapid modeling and predicting the time series of mobile communication traffic.

The invention relates to the field of tunnel construction, and discloses a tunnel geological exploring method and system and a mini-type heading machine. The positions of at least two exploring holes are determined on a tunnel face of a tunnel; the mini-type heading machine is placed in the exploring holes, a tunneling path and a positioning program are input through a computer in a control chamber of a shield tunneling machine, the mini-type heading machine is started, and a rock mass is broken; and when a bad stratum is encountered or rock mass parameters need to be measured, the geological condition of the rock mass is measured; and measuring data are sent to the computer in the control chamber of the shield tunneling machine, and accordingly the analysis is conducted. The running performance of the tunnel geological exploring method and system and the mini-type heading machine is stable, the interference resisting capability of data transmission is high , surrounding rock information parameters can be accurately obtained, the tunnel geological exploring method and system and the mini-type heading machine can be combined with the shield tunneling machine, and the advanced prediction of the real-time geological condition is achieved.

The invention provides a construction method for an intermittent dynamic prediction model for microorganisms of coal-chain meat products, and belongs to the field of intelligent prediction of the quality of the cold-chain meat products. The method comprises the following steps of setting dynamic fluctuating temperature; carrying out microbial culture under different temperature conditions; constructing an intermittent dynamic growth prediction model for the microorganisms; carrying out model verification; and constructing a prediction model for the shelf lives of different meat products in a cold chain process. The construction method for the intermittent dynamic prediction model for the microorganisms of the coal-chain meat products is easy to operate and reliable in result, the method iscapable of predicting the quality of the cold-chain meat products under the conditions of fluctuating and constant temperature in real time, and the quality of the meat products in the cold chain process can be monitored in real time, a rapid and efficient prediction method for the microorganisms is provided for enterprises and consumers, and the intelligent prediction model for the microorganisms is established, so that the model prediction precision and efficiency are improved, and an effective means is provided for evaluating the edible safety of the meat products in the cold-chain meat product process.

The invention discloses a deformation monitoring method, comprising the following steps of: 1. installation of a monitoring device and establishment of a space coordinate system; 2. detection of an inclination angle of an object to be tested; 3. compensation of an inclination state of the object to be tested; 4. monitoring of a settlement state of the object to be tested; and 5. monitoring of a cracking state of the object to be tested. The deformation monitoring method disclosed by the invention has the advantages of reasonable design, low cost, time saving, labor saving, small cost, and capability of obtaining deformation states such as inclination, settlement, cracking and the like of buildings, structures and slopes, and performing real-time monitoring, so that forecast can be proposedin time, prevention measures can be conveniently taken, and the practicability is high.

The invention discloses a short-term tourist traffic and trend prediction system based on streaming data extraction. The system comprises a client terminal and scenic zone data terminal acquiring data; a database is established for multi-dimensional statistics; input variable parameters are acquired by converting a computation formula; a prediction model is established through a radial basis function neural network. On the basis of the system, the adopted data resource includes data acquired by an intelligent terminal, scenic zone operation and business platform backstage data and the like, the model established according to the radial basis function neural network can be used for real-time traffic prediction, and the auxiliary determination effect is achieved.

The invention discloses a method for predicating a travel destination by using GPS travel data. The concrete predication method comprises the following steps: 1, determining travel destination and dividing travel destination predication time intervals; 2, determining the frequently visited travel destinations in different travel destination predication time intervals and the supporting point corresponding to every frequently visited travel destination; and 3, respectively establishing and demarcating travel destination predication models in the workday time interval and in the non-workday time interval, and predicating the travel destination. The predication method allows the travel destination in the workday time interval and in the non-workday time interval to be predicated by using the GPS travel data, so the travel destination in all travel modes can be predicated without traditional household large-scale resident travel investigation or road network graph obtaining, and the method has high identification accuracy and strong applicability. Predication results can be applied to vehicle navigation equipment or mobile phone navigation software in order to carry out real-time travel destination recommending.

The invention relates to a method for ultra-short-term prediction of solar irradiance in clear and cloudy weather, which is based on an all-sky imager (TSI-880) ground cloud picture, including cloud retrieval of real-time ground-based cloud images, using the cloud image retrieved from the time-series cloud cover to predict the cloud image; Then, the multiple linear regression equation is constructed by using the cloud amount and clear sky factor data of n time cloud pictures before t time. Finally, the cloud amount data of t + Delta t time cloud pictures in the future are substituted into theregression model to obtain the forecast clear sky factor, and then the forecast surface irradiance of the corresponding future time is calculated. The invention can effectively predict the variation of the ground solar irradiance caused by the movement and change of the clouds in the clear and cloudy weather condition in real time, overcomes the shortcomings of the existing solar radiation prediction method such as randomness, uncertainty, large error and the like in the clear and cloudy weather condition, and realizes the advantages of good real-time performance and high accuracy.

A power transmission line icing thickness predicting method based on an LSTM artificial neural network adopts historical icing thickness data of power transmission line poles and towers, data monitored by tensile force terminals arranged on the power transmission line poles and towers, data from real-time voltage monitoring terminal and data monitored by weather terminals, utilizes an LSTM artificial neural network algorithm technology to dynamically obtain the icing thickness on a power transmission line to be predicted according to real-time change of weather through data fusion, combines with designed borne icing thickness of the power transmission line to conduct early warning on the safety of iced poles and towers. Compared with a traditional artificial ice observing method, labor andmaterial resources are saved, better timeliness is also obtained, the fund input is smaller and the maintenance cost is lower compared with monitoring based on cameras, and power transmission line poles and towers having potential safety hazards can be found more timely.

The invention relates to a method and a system for real-time large-scale scene three-dimensional scanning modeling. The method comprises the following steps: an RGB-D video stream is acquired; global camera pose optimization is carried out on the above video stream to acquire a global camera prediction position; and according to the global camera prediction position, globally-optimized three-dimensional scanning modeling is carried out. Real-time and terminal-to-terminal modeling can be realized. Local camera tracking does not need to rely on, but the globally-optimized camera positioning is kept constantly. A parallel optimization modeling system combining sparse feature points, dense geometry, and color information is brought forward and developed, the globally-optimized camera position can be predicted in real time, camera repositioning is supported, a globally-consistent optimal three-dimensional model can be updated in real time, the system of the invention is applied to large-scale indoor scanning, and a high-quality three-dimensional model can be acquired.

A real-time flood prevention monitoring system based on edge computing and big data comprises a real-time monitoring unit, more than two edge calculation modules. The real-time monitoring unit is usedfor acquiring images acquired in a real-time manner to the edge calculation modules. Each edge calculation module performs single-point pre-calculation corresponding to at least one monitoring point,identifies and classifying the data at the network edge, and reports the abnormal trend data after identifying abnormal trend data. The system also comprises a data storage and calculation center which is only connected with the changed picture data sent back by all the edge receiving calculation modules, carries out overall continuous flood trend prediction and flood control planning, and sendsa result; a real-time monitoring and early warning center used for receiving the single-point flood trend abnormity identified by the edge calculation module in real time; and a big data center usedfor transmitting alarm and prediction data and generating an overall flood real-time trend and flood control arrangement. The system has the advantages that the network transmission quantity between the terminal and the data center can be reduced, and the storage cost and calculation cost of the big data center are reduced.

The invention provides a system and a method for predicting influence of music on behaviors of a driver. The method comprises the following steps: when playing music in a driving process, recording road conditions ahead and a driving control operation as well as physiological characteristics of a driver during driving; determining music characteristics of music played and whether the driving control operation is dangerous or not; establishing a model for predicting influence of the music on behaviors of the driver; and by virtue of the model for predicting influence of the music on behaviors of the driver, predicting the driving control operation and judging whether the driving control operation is dangerous or not in real time according to real-time road condition images, physiological characteristics of the driver and the music characteristics of the music played, and reminding in the case of the control operation is a dangerous operation. The system and the method for predicting the influence of the music on the behaviors of the driver disclosed by the invention can remind dangerous behaviors which may be adopted by the driver, so that the probability of traffic accidents is reduced.

The invention relates to a flight conflict early warning method for an air traffic control system. The air traffic control system comprises a data communication module, a monitor data fusion module, an airborne terminal module and a control terminal module. The monitor data fusion module is used for realizing fusion between air traffic control radar monitor data and automatic related monitor data, and providing real-time flight path information for the control terminal module. The control terminal module comprises a preflight conflict-free 4D flight path generation submodule, an in-flight short-term 4D flight path generation submodule and a real-time flight conflict monitor and alarm submodule. According to the flight conflict early warning method for the air traffic control system, flight planning data are processed through the control terminal module, a 4D flight path is generated through a hidden Markov model, and potential traffic conflicts of airspace traffic situations are analyzed. The flight conflict early warning method for the air traffic control system can effectively and early warn flight conflicts and improve safety of air traffic.

The invention discloses a forest fire meteorological warning method and system. The system comprises a meteorological station, a data receiving terminal, a forest fire meteorological index calculation module, a forest fire meteorological level calculation module, a forest fire warning information calculating and editing module, and a warning information transmitting module. The data receiving module receives information of the meteorological station and analyzes the information, the forest fire meteorological index calculation module calculates forest fire meteorological indexes of warning sub-areas and stores the indexes, the forest fire meteorological level calculation module further calculates forest fire meteorological levels, the forest fire warning information calculating and editing module calculates warning levels and edits warning information, and the warning information transmitting module transmits the warning information to terminal users.

The invention discloses a method for predicting mobile communication telephone traffic based on clustered LS-SVM (Least Squares-Support Vector Machine), belonging to the field of communication and solving the problem that the LS-SVM has high computational complexity and lowered generalization capability in practical application. The method comprises the following steps of: 1. selecting the historical data of telephone traffic of 4-6 months at the current moment, taking the historical data of telephone traffic as a training sample, pre-processing the historical data, clustering, carrying out LS-SVM modeling, and obtaining C LS-SVM prediction models; 2. pre-processing a newly-input sample, and carrying out reconstruction of phase space and normalization processing; 3. classifying the reconstructed newly-input sample according to C clustered results obtained in the step 1, and determining the type of the sample; and 4. according to the classification result obtained in the step 3, inputting the newly-input sample into the LS-SVM prediction model of the corresponding class, outputting a prediction value, and quickly predicting the input sample to be processed.

The invention discloses a packing tower flooding prediction method based on pressure drop standard deviation analysis. The method includes 1, arranging pressure drop detection ports of packing layers of a packing tower; 2, allowing an intelligent pressure transmitter to connect to the pressure drop detection ports through pressure detection pipes and perform detection in real time; 3, allowing the pressure drop detection ports to transmit the pressure drop data to a data collector, processing the data, and acquiring the changing curve of pressure drop with tower operation flux and the pressure drop standard deviation distribution curve; 4, adopting the pressure drop standard deviation distribution curve as the basis for judging the packing tower flooding, and determining the flooding state when the slope of the pressure drop standard deviation distribution curve exceeds a critical value. The acquiring frequency of the pressure drop detection ports is larger than 20 hertz.

The embodiment of the invention discloses a method for predicting the movement track of a special crowd and electronic equipment, and the method comprises the steps: obtaining the activity behavior characteristics of a user of a target crowd, and obtaining the track activity data of a base station coverage area of a plurality of specified frequency bands of a plurality of specified time periods of the user; on the basis of the communication identifiers of the users in the target crowd, associating the activity behavior characteristics of the users in the target crowd with the base station coverage area track activity data of the users in a plurality of specified time periods; taking the activity behavior characteristics of the user and the specified time period as the input of a movement track prediction model, taking the activity behavior characteristics of the user and the base station coverage area corresponding to the specified time period as the output of the movement track prediction model, and training the movement track prediction model; and according to wireless measurement information and known position information, establishing a moving track prediction model, and reversely positioning the position of a user only containing the wireless measurement information, so that the position continuity requirement is met, and therefore real-time and accurate prediction is achieved.