183 results about "Traffic forecast" patented technology

A Dynamic Personal Trip Routing System (DPTRS) which provides users with routes recommendations as a factor of weather and traffic conditions, as well as periodic and historical collected data. The DPTRS also includes a subsystem architecture which provides users the ability to contribute to data collection and update data to be used in providing real-time traffic forecasts. The DPTRS allows for the use of a unique revenue model.

The invention provides an active control method of an oversaturated traffic situation at an intersection group, and aims to realize the active prevention of the oversaturated traffic situation at the intersection group. First, analysis is conducted to the traffic development situation. The possible traffic flow in the balanced dynamic situation of users and the ideal traffic flow of the optimal dynamic situation of a system are predicted based on the needs of the starting point and the ending point of a short time in the future on going out in variable time. Second, traffic signals are actively controlled and optimized. The information of the oversaturated intersection group which possibly occurs in a short time in the future, key crowded paths, recommended shut paths and the optimal transfer flow are extracted, a secondary optimal road network under the requirement of the expected optimal transfer flow is formed, and a signal control scheme is dynamically optimized through a vague matching method by integrating a historical traffic signal control strategy library. Finally, the traffic forecast information is generated. The travel choice behaviors of a driver are analyzed, the corrected value of the saturation of the possible flow of the key crowded paths is worked out by integrating the requirements on the expected optimal transfer flow, and the corrected value of the saturation of the possible flow of the key crowded paths and the saturation of the possible flow of the recommended shunt paths are released as traffic forecast information.

The invention discloses an urban traffic flow prediction method with regional vitality integrated. The method comprises steps: S1, regional division is carried out on an urban road network, and the traffic flow in each region is calculated; S2, a regional vitality model is designed: by using distribution of urban points of interest, holidays and weather information, a 3D convolution neural network(3D CNN) is used to study the dynamic changes of the vitality in each urban region; S3, a flow prediction model is designed: the regional vitality and the traffic flow are integrated, and a convolution long short-term memory network (ConvLSTM) is used for flow prediction; and S4, according to the historical data, the regional vitality model and the flow prediction model are trained at the same time, and the well-trained models are used for predicting the traffic flow in each region in real time. Through integrating the regional vitality and considering influences from a driving force behind crowd activities and external factors, high prediction precision can be acquired.

A computer-implemented method and system are provided for forecasting traffic load on a communications network driven by market factors.

The invention provides a passenger volume prediction method of urban rail transit. The passenger volume prediction method comprises the following steps: accounting OD (Origin-Destination) distributionmatrices based on current time of target rail transit lines; acquiring a passenger volume prediction model by building and using historical OD distribution matrix training of the target rail transitlines according to a prediction requirement; and predicting the passenger volume of specified road sections of the target rail transit lines in a prediction time period. According to the passenger volume prediction method of the urban rail transit disclosed by the invention, passenger volume prediction processes can be effectively simplified, calculation speed and calculation precision are improved, and a powerful support is provided for reasonably carrying out management and dispatching of transport in time.

The embodiment of the invention provides a traffic predicting method and system. The method comprises the steps that current traffic flow data are acquired, wherein the current traffic flow data comprise current traffic flow data of a road segment to be measured and current traffic flow data road segments related to the road segment to be measured; a first predicting model is determined; the current traffic flow data are input to the first predicting model, and a predicting traffic flow value of the road segment to be measured is obtained, wherein the first predicting model carries out discrete processing on historical traffic flow data by acquiring the historical traffic flow data to obtain discrete historical data, a space-time association rule is obtained by mining the discrete historical data, a set of the road segments related to the road segment to be measured is obtained through the space-time association rule, and the historical traffic flow data of the set of the road segments are generated in a training mode through a preset method.

Method for controlling the elevators in an elevator bank in a building divided into zones has a plurality of floors via a process whereby destination floor calls are issued to the elevators by destination floor call input devices in the lobby of departure and the calls are distributed internally among different zones in the building by the elevator group control system, the elevators and the floors to be served being divided dynamically within the aforesaid elevator group control system into aforesaid zones, varying the numbers of elevators and the zone limits (r) according to traffic forecasts and transportation need.

A method and system for distributing labor based upon determinations of traffic in a facility, such as a store. Daily traffic forecast information is obtained for each day within a given time period from a source of such information. Calendar and event information for the facility for each day within the given time period is determined. Baseline days are selected from historical traffic data, and baseline averages and percentages for a predetermined time interval are also selected. The distribution of traffic for each day within the given time period at each time interval is determined using the baseline percentages for each time interval. Labor data is distributed. Labor recommendations are provided at each time interval for the given time period based on the distribution of traffic for each day, and at least one of other user-defined workforce requirements. The results of any or all the these steps are displayed to a user.

A computer-implemented telecommunications spending analysis system extracts telecommunications traffic detail data from billing statements provided by telecommunications carriers and converts the traffic detail data to a generic data format for aggregation in a uniform customer traffic history database. The spending analysis system compares billed rates against estimated market prices stored in a best of class (BOC) database and notifies the user of possible cost savings attainable in a renegotiated contract. A web-based RFP system facilitates preparation of a request for proposals (RFP) and solicitation of vendors to bid on the RFP in an online reverse auction environment, both for telecommunications services and other commodities. The RFP system may access the uniform customer traffic history database to facilitate preparation of a traffic forecast of the RFP. Upon completion of an RFP auction, the RFP system updates the estimated market prices of the BOC database based on the vendor bids.

Generally, embodiments of the present invention provides for methods, systems and computer program products for utilizing non-compete advertisement tags in an advertisement serving system. The method according to one embodiment of the present invention comprises identifying one or more web pages that contain a first advertisement tag and selecting a given web page from the one or more web pages. A second inactive advertisement tag is then inserted into the web page, which requests an advertisement for the web page that is subsequently ignored. The method further comprises monitoring a number of impressions the web page receives using the second advertisement tag in order to determine a traffic forecast. The method further comprises setting the second advertisement tag to an active status, causing the first advertisement tag from the web page to be removed and an advertisement to be returned in response to the request by the second advertisement tag.

The invention relates to real-time service transmission quality assurance in the communication of an Ethernet passive optical network, in particular to a dynamic bandwidth allocation method for self-adapting service quality assurance in the Ethernet passive optical network. The method adopts the technical scheme that: on the premise of assuring the service quality promised by a system, such as maximum time delay and time delay dithering of a real-time service, according to uplink capacity and downlink capacity of the system, the polling cycle is automatically regulated to enable the system performances including throughput, time delay and time delay dithering to be optimal. The method comprises the following steps: firstly, determining a maximum polling cycle of the system according to the maximum time delay and the time delay dithering of the real-time service which can be assured by the Ethernet passive optical network (EPON) system; secondly, determining the minimum polling cycle according to the downlink capacity of the system; thirdly, predicting the traffic of an optical network unit (ONU); and fourthly, allocating the bandwidth. The invention is mainly applied to the communication of the Ethernet passive optical network.

The invention discloses a three-layer satellite network load balancing routing method based on traffic forecast and belongs to the technical field of a satellite network routing policy. According to the method, through adoption of a virtual topology policy, business classification, inter-satellite distance, current satellite node activity degree and a traffic forecast value within a next time period are taken into consideration, routing computation comprises two steps of 1, carrying out time slice switching rerouting, namely, when a time slice is switched, computing the shortest path by takingdistance as weight; and 2, carrying out time slice routing, namely, computing the shortest path by taking the traffic forecast value as the weight in the time slice. The method provided by the invention is satellite node-oriented; traffic is guided to the satellite nodes with low traffic and low activity degree; the traffic load balancing is realized; the throughput of a whole network is improved; and the mean transmission delay of the whole network is reduced.

The invention discloses an urban people flow prediction method based on a Seq2Seq generative adversarial network, and the method comprises the steps: abstracting the urban people flow data at different times into image frames, and representing the people flow through a thermodynamic diagram; Dividing the observation data into training data and labels according to time, and converting the problem into an image problem; The idea of WGAN generative adversarial network is generally adopted, a generator generates pedestrian flow in a certain period of time in the future on the basis of historical data by using a Seq2Seq method, and external factors such as weather are added at the same time; The discriminator uses a Waserstein distance to discriminate true and false data; In the training process, the generator and the discriminator are continuously optimized by combining the generative adversarial loss and back propagation. And finally, when the discriminator cannot discriminate the authenticity, the optimized generator is used for predicting the future urban pedestrian flow. According to the method provided by the invention, the generative adversarial network is used for carrying out urban people flow prediction for the first time, and the problems of fuzzy prediction and slow algorithm convergence are solved in combination with external environment factors.

Methods of providing trip status information periodically to a user in transit to a destination are disclosed. Trip status information comprises information and alerts based on estimates of various time-of-arrival metrics, such as expected time-of-arrival and earliest time-of-arrival. The estimates are based on a plurality of data, including calendrical time (the time and date), historical statistics, average speed, current weather, weather forecasts, current traffic, and traffic forecasts.

The invention discloses a road condition identifier based on a license plate recognition technology and an identification method thereof. The identifier mainly comprises a front-end video capture device, an automatic license plate identifier, a vehicle detection device and a multifunctional repair state identification module. The intelligent control of traffic lights is realized finally through collecting traffic parameter data and combining massive historical traffic data. The invention can realize the short-term traffic forecast, detection of special traffic incidents and dynamic control of traffic signals according to the current traffic state, and has the advantages of simple installation and convenient use.

The invention relates to a depth learning-based regional traffic signal lamp control system and method. The depth learning-based regional traffic signal lamp control system comprises an information acquisition unit, a storage unit, a 5G communication unit, a cloud end data processing and database unit, a regional road network model building unit, a Synchro-based simulation calculation unit, a depth learning-based traffic forecast unit a traffic signal control unit, wherein data such as traffic flow is acquired by the acquisition unit, 5G communication is used for transmission, data informationsuch as each intersection traffic flow is integrated at a cloud server, forecast is performed according to depth learning, an optimal control scheme is obtained by simulation of the forecasted traffic flow data according to Synchro and is finally transmitted to each intersection traffic light for execution, the traffic light management and control in the region can be effectively optimized, and apractical method is provided for traffic congestion reduction.

The invention relates to a merchant passenger flow prediction method fusing a historical mean value and a boosted tree. The method is characterized by comprising the following steps of preprocessing merchant complete behavior data in a time period; constructing features for the preprocessed data; based on the historical mean value and the boosted tree, building a passenger flow prediction model; and performing passenger flow prediction. The invention provides an internet merchant passenger flow prediction model fusing the historical mean value and the boosted tree. The essence of the model isa weighted sum after weight coefficients, calculated by a calculation formula, of a boosted tree model and a historical mean value model are fused according to a certain ratio. The method considers not only how to improve the prediction precision of the model but also a dependency relationship between passenger flow prediction and time, and performs comparative analysis on prediction results of different models.

The invention discloses a telephone traffic forecasting method of an electric power call center. The telephone traffic forecasting method is characterized by comprising the following steps: 1) collecting 96-point historical data of telephone traffic in a forecasting area, and simultaneously collecting relevant factor information influencing telephone traffic load changing corresponding to the historical data, wherein the relevant factor information includes a weather factor and electrical load information data; 2) identifying and correcting bad data; 3) forecasting the telephone traffic; 4) performing weighted averaging on the forecast results of several optimal models during the forecasting process of medium and long term telephone traffic forecasting, short term telephone traffic forecasting or super-short term telephone traffic forecasting to obtain the final forecasting results by choosing proper weight; and 5) performing accuracy rate comparison on the telephone traffic forecast results. The telephone traffic forecasting method can improve telephone traffic forecasting accuracy.

The invention discloses a long-time-sequence traffic flow prediction method based on a graph convolution-Informer model, and belongs to the technical field of long-time-sequence traffic flow prediction. The technical problem to be solved is to provide an improvement of the long-time-sequence traffic flow prediction method based on the graph convolution-Informer model. According to the technical scheme, the method comprises the following steps that speed information of all passing vehicles at expressway stations and provincial and arterial highway intermodulation stations is collected in unit time, and a traffic flow time sequence information data set is established after data preprocessing; a site network structure topological graph is established according to the relative geographical location information of the expressway stations and the provincial and arterial highway intermodulation stations; a two-layer graph convolutional neural network model structure is constructed, a road network topological structure and traffic flow time sequence information are coded, and spatial dependency feature information of data is learned; coding information obtained through image convolution is input into an Informer layer for training, and data long-time-sequence dependence feature information is learned. The method is applied to traffic flow prediction.

The invention discloses a traffic control method based on load prediction in a SDN network, belonging to the technical field of communication. The traffic control method based on load prediction in the SDN network of the invention comprises the following steps: S1: a switch feeds back traffic information of each link within a time slot T to a SDN controller; S2: the SDN controller performs statistics on a link traffic periodic change rule according to the link information fed back by the switch; S3: when a prediction time point arrives, the SDN controller performs statistics on a traffic load range on each link to obtain an average traffic parameter F<1> within a time period T<1>; S4: the SDN controller obtains a short period average traffic parameter F<2> within a time period T<2> according to real-time link traffic information fed back by the switch. In the traffic control method based on load prediction in the SDN network of the invention, a traffic forwarding strategy is regulated dynamically through a link traffic predicted value and a real-time link state, link load is reduced effectively and a network congestion problem is solved, thus, the method has better popularization and application value.

The invention discloses a tourist flow prediction method based on machine learning, which comprises the following steps: collecting historical tourist flow data of tourist scenic spots and sorting thedata according to year, month and day; acquiring correlation data corresponding to the period of time of the historical tourist flow data, wherein the correlation data comprises at least one of a maximum temperature, a minimum temperature, a weather, a wind direction, a wind force and a working day, and summarizing the historical tourist flow data and the correlation data in a unit of days; converting the associated data into numerical values and fusing them with historical tourist flow data; inputting the related data and historical tourist flow into the learner for training to realize the prediction of tourist flow. The technical scheme utilizes machine learning method, comprehensively considers the intrinsic correlation of various factors affecting tourist flow in scenic spots, and assists the weighting calculation method, in order to improve the accuracy, scientificity and convenience of tourist flow prediction.

The invention discloses a deep learning-based space-time long-short-term urban pedestrian volume prediction method. The method is used for pedestrian volume prediction. According to the method, when the spatial correlation is extracted, and the local convolution is carried out on the correlation of the adjacent regions. The correlation of the remote regions is extracted by using graph convolution,so that the integrity of the spatial correlation is considered while parameters and calculated amount are reduced. According to the method, short-term time dependence and long-term time dependence are captured at the same time, so that a prediction result is more accurate in a time dimension. The regional semantic information distribution is considered. Each type of region of interest (POI) in each region is endowed with a corresponding proportion weight, so that the influence of regional semantics on urban pedestrian flow is utilized more accurately.

The invention discloses a LightGBT algorithm-based traffic forecast method. The LightGBT algorithm-based traffic forecast method comprises the following steps of S1, acquiring traffic data, performingdata normalization, and dividing the data into training data and test data; S2, performing model training on the training data by a LightGBT algorithm, and determining a model parameter; S3, inputting a LightGBM model parameter and the test data, and forecasting traffic flow; and S4, performing error estimation on a LightGBM model forecast result, reducing forecast data, and outputting the forecast data. With the LightGBM model, the forecast time is substantially reduced as well as the forecast accuracy is improved, and the LightGBT algorithm-based traffic forecast method has better forecastperformance and generalization capability in forecast of expressway traffic.

The invention provides an unmanned aerial vehicle spraying control system and method based on flow dynamic prediction. The system comprises a pressure sensor and a controller, wherein the pressure sensor is respectively connected with the controller and a spraying system on an unmanned aerial vehicle; the controller is connected with the spraying system on the unmanned aerial vehicle; the pressure sensor is used for obtaining a real-time pipeline pressure of the spraying system and transmitting the real-time pipeline pressure to the controller; and the controller is used for predicating pipeline flow according to the real-time pipeline pressure and a corresponding relation between the pipeline pressure and the nozzle spraying flow of the spraying system and also used for controlling the pipeline flow of the spraying system on the unmanned aerial vehicle according to the obtained pipeline flow predicated value and a preset pipeline flow. Through adoption of the system provided by the invention, the spraying amount of each acre of operation area can be accurately and dynamically controlled in the speed changing process and a situation of large spraying flow error caused by manual operation can be also avoided.

The invention discloses an urban area road network traffic flow prediction method and system based on a hybrid deep learning model, and the method comprises the steps: carrying out traffic flow statistics based on vehicle passing data of a checkpoint; performing spatial-temporal distribution characteristic analysis on the traffic flow data of the checkpoint, and performing characteristic extraction according to an analysis result to obtain spatial-temporal influence factors; according to the space-time influence factors, constructing and training a ConvLSTM and BiLSTM mixed deep learning model; performing synchronous prediction on the traffic flow of an urban regional road network, selecting a prediction loss function and an evaluation index, and performing visual expression on a result; calculating the traffic flow change degree through a linear time sequence prediction model Prophet, carrying out traffic state recognition, and achieving traffic state pre-judgment. According to the invention, a traffic management department can be helped to carry out dynamic management scheduling on urban roads, optimization management is carried out on an urban road network from the overall situation, a management strategy and a management scheme are formulated, and effective data support is provided for traffic managers and decision makers.

The invention relates to a short-time flow prediction method based on an ETC portal system in the technical field of traffic information. The method comprises steps of extracting historical vehicle passing data in a to-be-predicted road segment, and collecting sample flow data according to lanes; flow change floating values of all lanes in different time periods of two or more adjacent days beingcalculated respectively, and then a change floating average value being obtained through calculation; selecting two or more than two adjacent ETC gantries in the same passing direction of the expressway section to be predicted, and counting historical traffic flow data; acquiring RSU antenna data of each lane of the ETC portal in the same time period according to the ETC portal in the to-be-predicted road section, converting the RSU antenna data into a traffic volume data matrix, and calculating real-time flow of each lane of the road network in different time periods; multiplying the obtainedlane real-time flow data by the corresponding floating average value to obtain a flow prediction value. The method is advantaged in that an ETC portal system-based short-time flow prediction method provided by the invention is rapid and convenient in data acquisition, high in flow data updating efficiency and rapid and accurate in prediction.

The invention provides a scenic spot tourist amount prediction system comprising a data collection module, a data prediction module and a flow display module, wherein the data collection module is used for collecting predicted data; the data prediction module is provided with a passenger flow prediction model based on LSTM and is used for predicting a scenic spot tourist amount according to the predicted data collected by the data collection module; and the flow display module is used for displaying the scenic spot tourist amount data. The scheme can predict the tourist amounts within different time periods.

The invention discloses a method for forecasting monthly urban rail transit passenger flow through an ARIMA model based on a seasonal index. The method comprises the steps that original monthly passenger flow sample data is selected, the seasonal index of each month in a year is obtained through calculation according to a direct average seasonal index method, seasonal adjustment is conducted on an original monthly passenger flow sequence in samples by means of the seasonal index, monthly passenger flow sequence data subjected to conversion is subjected to stabilization processing, model identification and parameter estimation and inspection in sequence so as to construct the proper ARIMA model, and finally, a forecast result output by the model is subjected to reverse seasonal adjustment, that is to say, a final monthly passenger flow forecast value is obtained. The method can be used for improving the accuracy of the monthly urban rail transit passenger flow forecast and providing reliable data support for large production plans such as adjustment of a monthly vehicle maintenance plan, a serviceable car allocation plan and a daily transportation plan.

The embodiment of the invention relates to the technical field of networks, and discloses a scenic spot passenger flow prediction method and device, a server and a storage medium. The scenic spot passenger flow prediction method provided by the invention comprises the steps of acquiring real-time passenger flow data in a preset geographic range wherein the real-time passenger flow data is passenger flow data in a first preset time period before the current moment; judging whether a prediction day belongs to holidays or not; if judging that the prediction day belongs to the holiday, determining the passenger flow prediction value of the to-be-tested scenic area in the prediction day according to the real-time passenger flow data and the holiday prediction model; and if judging that the prediction day belongs to a non-holiday, determining a passenger flow prediction value of the to-be-tested scenic area in the prediction day according to the real-time passenger flow data and a non-holiday prediction model. According to the embodiment, the passenger flow volume of a scenic area to be detected can be accurately predicted.