82results about How to "Reasonable prediction" patented technology

The invention provides a regional traffic dynamic regulation-control method, which comprises the steps of: selecting certain intersections to be set as feedback gate control intersections; ranking the intersections to obtain traffic flows of the intersections; establishing a regional macroscopic fundamental diagram; and regulating the intersections when a network macroscopic net flow reaches a threshold value. In the scheme provided by the invention, the traffic control and traffic guidance are combined to form a feedback type system, and dynamic flow interception is carried out on the periphery of a central region for controlling the traffic flow entering the central region by utilizing a macroscopic fundamental diagram function obtained through data of a detector, thereby reducing traffic pressure in important regions.

The invention relates to the technical field of transport line planning, and discloses a method for determining customized bus traffic zones and get-on / get-off stations, a customized bus line planningmethod which is based on the determined get-on / get-off stations and enables each single station to correspond to another single station, a method for evaluating rationality of customized bus lines onthe basis of K-fold cross validation, and a taxi GPS big data-based customized bus line planning method. Compared with the prior art, existing taxi GPS big data is utilized to a greater extent to putforward a better traffic zone division method and a better line planning method and a better line evaluation method, so that better reference basis is provided for the development of customized buses.

The invention discloses a single-index prediction and early warning method based on an artificial intelligent big data platform. The method includes the steps: (1) single-index prediction based on theartificial intelligent big data platform; (2) single-index warning based on the artificial intelligent big data platform. Based on the range interval of historical upper and lower bound predictive index values, warning is performed once historical upper and lower bound threshold ranges are continuously and repeatedly broken through within a period of time, namely, abnormity exists within the period of time. Intelligent analysis is performed based on historical index data, index abnormity trend can be judged in real time according to corresponding early warning strategies by the aid of index data of indexes monitored by the artificial intelligent big data platform in real time, potential risks are discovered before a monitoring system, early warning for a production system is realized, enough time is reserved for operation and maintenance experts to remove system faults, and production system risks are avoided as far as possible.

The embodiment of the invention discloses a pavement performance prediction and maintenance method and device based on maintenance history correction. The method is based on road conditions, traffic volume data and maintenance history data of a route over the years, and adopts a modeling idea of "model grouping, year-by-year accumulation and elimination of maintenance history influences, data update dynamic correction" to finally obtain a pavement performance decay model that is based on untreated actual road condition data over the years and considers traffic volume cumulative axle-load influences. The model established by the method not only can be used to accurately predict the natural attenuation index of road conditions of a pavement in future years, but also can be used to assist incalculating preventive maintenance benefits and judging the best maintenance time in the aspects of pavement maintenance planning and decision-making, and meanwhile, when performing multi-year planning, more scientific and reasonable maintenance schemes and plans can be formulated based on accurately predicted road condition attenuation values, and the reasonable forecast of pavement maintenance funds can be achieved.

The invention provides a method for estimating film thickness based on an even function seismic response. The method comprises the steps of carrying out 90-degree phase shifting on a zero-phase odd function seismic response to form the even function seismic response, carrying out matching pursuit time-frequency analysis on the even function seismic response to obtain a peak amplitude and peak frequency, and estimating the film thickness through the combination of the peak amplitude and peak frequency. A device with the method used comprises a preprocessing module used for carrying out 90-degree phase shifting on the zero-phase odd function seismic response to form the even function seismic response, a matching pursuit time-frequency analysis module used for carrying out analysis on the even function seismic response to obtain the peak amplitude and peak frequency, and a film thickness obtaining module used for estimating the film thickness through the combination of the peak amplitude and peak frequency. According to the method and device for estimating the film thickness based on the even function seismic response, the equivalence between 90-degree phase shifting and film odd-even relation conversion is verified, the advantage of an even function on film thickness estimation is determined, and prediction of the film thickness is facilitated through the combination of the peak amplitude and peak frequency.

The invention discloses a Floodlight-based management platform for an SDN-based wireless network and an authentication method and belongs to the field of wireless network management. The wireless network management platform is realized on an Apache server and comprises a system management module, a topology management module, a configuration management module, a user management module, an AP management module and a data statistics module. A thread used for periodically monitoring user information changes is added into a controller. The AP management module is taken as a wireless AP while functions of an OpenFlow switch are realized. The data statistics module periodically acquires network data and predicts the network status according to a data analysis module. Distributed databases synchronize the added / modified user information with each other, so that the consistency of users of the whole network is kept. The frequent user changes can be handled in real time to adapt to the management requirements for the wireless network, and the SDN controller data storage and feedback ability is added to realize unified authentication of network users.

The invention provides a sulfur deposition prediction method for an acidic natural gas pipeline. According to the technical scheme of the invention, based on a sulfur solubility model, the classic nucleation theory and a particle motion equation, a thermodynamic model and a dynamical model for sulfur precipitation and deposition are established. Meanwhile, the saturated distance, the supersaturated distance and the particle maximal migration distance are calculated, while the sulfur deposition position, the sulfur deposition yield and the sulfur deposition average thickness of the acidic natural gas pipeline are figured out. In this way, the sulfur deposition prediction for the acidic natural gas pipeline can be realized. Based on the above method, a supersaturated distance L2 exists between a saturation point and the appearance location of sulfur particles, so that the precipitation location of sulfur particles can be calculated more accurately. The motion trajectory of sulfur particles can be figured out based on the particle motion equation, and then the maximal migration distance L3 of particles is determined. In view of the mechanism, the above method is more reasonable in predicting the deposition location of sulfur particles, compared with the conventional sulfur deposition prediction method based on a critical velocity calculation model for sulfur particles in a vertical well bore. In this way, the prediction result is more scientific, more accurate and more reliable.

The invention belongs to the technical field of information retrieval and database structures, and discloses an Internet topic trend auxiliary prediction method and system and an information processing terminal. The method comprises the steps: incorporating the content and emotional characteristics of news event related reports into a calculation process of a moving average index and an index smoothing similarity moving average index; calculating the moving average for smoothing calculation through the statistical operation of the frequency information of the event in different time periods; carrying out the predictive analysis of the trend of the event through the intersection and tilt angle of the fast and slow lines; and achieving the quantitative calculation for predictive analysis ofa network report. According to the invention, a generated result can reflect the development trend of news more accurately, and a better prediction result can be obtained under the support of new bigdata.

The invention discloses a cloud platform security situation awareness method. The cloud platform security situation awareness method comprises the steps of 1) selecting a plurality of basic evaluation points and determining a quantized value of each basic evaluation point; and selecting a plurality of floating evaluation points and determining an initial value of each floating evaluation point; 2) updating the quantized value of each floating evaluation point periodically based on collected cloud platform information, wherein the quantized value of each floating evaluation point in the current period is updated based on the evaluation value of each floating evaluation points in the previous period and a security floating factor in the current period; 3) dividing the cloud platform into a plurality of security layers, fusing the quantized values of the evaluation points belonging to the same security layer to obtain an evaluation value C of the security layer, wherein the evaluation points include the basic evaluation points and the floating evaluation points; and 4) calculating a security evaluation result Q of the cloud platform based on the evaluation value of each security layer of the cloud platform and determining the security situation of the cloud platform. According to the cloud platform security situation awareness method, the security situation of the whole cloud platform can be monitored and evaluated, the operation is simple and convenient, and high security and reliability are achieved.

The invention discloses a network security situation awareness model and method based on a time sequence. The model comprises a data preprocessing module, a situation analysis module and a situation prediction module. The method comprises the following steps: firstly, data sets of different sources are collected; principal component information is extracted in the data; fusion analysis is carriedout on the principal component data by utilizing a D-S evidence theory; secondly, asset importance is added on the basis of vulnerability threats, a security situation value of the network is obtained, finally, a perception algorithm based on Fourier curve fitting is selected to predict the situation value in the future time period according to different situations, and a network security situation prediction curve is drawn; or a perception algorithm based on curve contour similarity is selected to predict the situation value of the future time period. According to the invention, the prediction of the network security situation is more accurate, convenient, effective and reasonable, and the network security situation in a future period of time can be reflected more clearly and visually through the network security situation prediction map.

The invention relates to a power battery air cooling control strategy based on a vehicle-mounted navigation system. The power battery air cooling control strategy specifically comprises the steps thatS1, the vehicle-mounted navigation system forecasts working condition information of a future road section; S2, the average speed and gradient of the future road section are calculated; S3, an automobile kinetic model is established by utilizing the average speed and the gradient; S4, the future required power of the power battery is calculated; S5, the future charging and discharging current ofthe power battery is calculated; S6, a future temperature rise prediction model of the power battery is established by combining heat generation and heat transfer mechanisms of the power battery; S7,the control system determines whether to start a fan or not according to the future temperature rise; S8, if the fan needs to be started, a dynamic planning algorithm determines the forced fan starting opportunity and the starting wind speed; and S9, air cooling heat dissipation treatment is conducted on the power battery according to the fan starting time and the wind speed which are obtained through dynamic planning. The invention provides a universally applicable air cooling control strategy for the power battery, and the energy consumption of the fan can be reduced as much as possible while the heat dissipation requirement of the power battery is met.

The invention relates to a floor system vibration response computing method under jumping loads. The floor system vibration response computing method under the jumping loads includes the steps that (1) a large number of single jumping load curves are collected to form an actually measured jumping load data base; (2) actually measured jumping load data are normalized and then are input into a standard excitation system for time-procedure analysis, and an acceleration response spectrum of the actually measured jumping loads is obtained through computation; (3) classification is carried out with testers as objects, statistics and fitting are carried out with an outer envelope line of a tester response spectrum as a representative curve, and an integral piecewise function expression of a design response spectrum is built; (4), a power response of each order of vibration mode of a floor system structure under the jumping loads is computed through the design response spectrum, the power responses of all orders are combined according to an SRSS method, and floor system power responses are finally obtained. Compared with the prior art, the floor system vibration response computing method can estimate the acceleration response of the floor system structure under the jumping loads rapidly and accurately.

The invention discloses a cell detection method which comprises the following steps: s1, manually labeling point labeling information on a test image, and extracting a plurality of labels from the point labeling information; s2, performing joint learning on the plurality of labels to obtain a plurality of teacher models; s3, performing learning optimization on the weights of the plurality of teacher models; s4, based on the weight of the teacher model obtained in the step s3, generating a weighted label, and using the weighted label to train a single student model; and s5, completing cell detection test based on the trained student model obtained in the step s4 to obtain a detection result. Based on a knowledge distillation model method, the cell detection performance is improved by more efficiently utilizing point labeling information.

The invention belonging to the field of rock exploration geophysics discloses a physical elastic template of a fracture-porosity type rock. In order to study the compact sandstone reservoir with high saturated gas, micro-fracture development and high heterogeneity, the model uses a Voigt-Reuss-Hill model to calculate the elastic modulus of mixed minerals; a differential equivalent medium (DEM) model is used for describing a skeleton elastic modulus of a facture and porosity rock; and on the basis of a Biot-Rayleigh theory, local fluid characteristics of the pore are described and a relationship between the elastic parameters and physical properties of the compact sandstone reservoir is disclosed.

The invention discloses a vehicle occupancy statistics system based on a WIFI (Wireless Fidelity) technology. The vehicle occupancy statistics system comprises a GPS (Global Positioning System) subsystem, a WIFI subsystem and a data publishing subsystem installed on a vehicle, wherein the GPS subsystem is used for acquiring real-time running GPS data of the vehicle at fixed frequency, processing and screening the GPS data, setting an information acquisition point, triggering an information trigger module of the information acquisition point according to the GPS data, and storing the GPS data and data of the information acquisition point; the WIFI subsystem is used for acquiring MAC (Media Access Control) data of passengers in the vehicle, processing an acquisition trigger signal transmitted by the information trigger module, controlling a WIFI probe to acquire the MAC data according to the acquisition trigger signal, calculating current number of people data according to the MAC data,and storing the MAC data; and the data publishing subsystem is used for visually outputting a real-time location and the real-time passenger carrying situation of a bus. By means of the acquisition ability of the WiFi probe and the WiFi coverage of the bus, the MAC data of the passengers are acquired in real time, and reasonable prediction of the bus arrival distance, arrival time and real-time number of passengers are realized in conjunction with the GPS data.

A method of providing solution heat treatment to an aluminum alloy. A non-isothermal process is used to provide a faster heat treatment cycle time while maintaining or further improving the alloy mechanical properties after subsequent aging hardening. The process includes establishing a temperature inside a processing vessel that is greater than a soaking temperature but less than a liquidus temperature of the alloy, rapidly heating the alloy to the soaking temperature in a first heating operation, reducing the temperature inside of the processing vessel to the soaking temperature, then heating the alloy to a temperature above the soaking temperature through a gradually increasing temperature in a second heating operation. Protocols for the improved solution heat treatment may be based on one or more of computational thermodynamics, dissolution kinetics and coarsening kinetics.

The invention discloses a distributed reinforcement learning social navigation method based on an image hidden variable probability model. According to the method, the image hidden variable probability prediction model is introduced to replace a traditional decisive prediction model, so that on one hand, the reasonability of the model is enhanced, prediction is closer to a pedestrian motion model with randomness, and on the other hand, the strategy performance is further enhanced by enhancing the exploration capability of the model, and overfitting is avoided. Meanwhile, the method achieves discrimination of a dynamic obstacle by decoupling movement of the mobile robot through an image sequence, a high-layer pedestrian detection module with instability is omitted, and sound migration is realized. In addition, a strategy sharing multi-agent simulation environment is designed to simulate a pedestrian dynamic environment, and the similarity between simulation and real crowd interaction is enhanced. In the environment, multiple agents synchronously perform data acquisition, so that the training time can be further shortened.

The invention provides a method for measuring in-situ simulated water quality in a pipe network under a water source switching condition. The method comprises the following steps: through separation treatment to a selected pipe section of a water supply system, an in-situ test pipe section, used for testing, of the water supply system is obtained, then testing is performed on the in-situ test pipe section, and detection is performed on water quality of each of local water of the water supply system and to-be-switched replacement water used for replacing the local water, so that a result about change of a water quality parameter under water source switching is obtained. According to the method provided by the invention, part of pipe sections are separated on an existing pipe, and test pipes do not need to be digged out completely, so that excavation workload is reduced, improvement cost is low, pipe sections in the excavation-limited area can be still tested, movement of the pipe sections is avoided, influence on existing physical, chemical and microbiological systems in the pipe is small, and obtained test data can fully simulate the situation about change of water quality in the existing pipe under the water source switching condition, which has very important guiding significance in stability of water quality in the water supply pipe network under the predicted water source switching condition.

The invention provides an esophageal squamous cell carcinoma prognosis index model construction method based on clinical phenotypes and LASSO. The method comprises the steps: firstly, collecting M types of clinical phenotype index information and lifetime information of an esophageal carcinoma patient, and enabling the information to serve as an original data set; then, by using a KaplanMeier method and a logrank method, researching the relationship between the clinical phenotypic index and lifetime information of the esophageal cancer patient; analyzing clinical phenotypic indexes influencing survival prognosis of the patient by utilizing single-factor COX regression and multi-factor COX regression; and then, analyzing and screening indexes with higher survival relevancy with the patient through an LASSO regression method, constructing a prognosis index of a patient prognosis survival evaluation model, solving the prognosis index of the patient through a clinical phenotypic index of the patient, and further judging the prognosis survival risk of the patient. According to the method, the postoperative survival condition of the esophageal squamous cell carcinoma patient can be accurately predicted, the prognosis risk prediction capability is improved, and the prognosis risk prediction cost is reduced.

The invention relates to big data processing, and aims to provide a power sales amount intelligent prediction method based on a deep recurrent neural network. The method comprises the steps of readinghistorical data of sales flow and electricity consumption of an electric power department, performing information mining and analysis after preprocessing, and evaluating a relation between the amountpayment time and user payment time to obtain distribution information; organizing a historical data structure, taking normalized n-day data as input, learning high-dimensional characteristics by using a multi-layer recurrent neural network (GRU), and inputting the high-dimensional characteristics into a softmax discriminator to carry out sale amount level classification in a certain period of time; and traversing hyper-parameters of the deep circulation network model by using a grid method, recording the optimal hyper-parameters after multiple experiments, constructing a final amount prediction deep circulation neural network model, and intelligently predicting the electric power sales amount by using the final amount prediction deep circulation neural network model. The method is more accurate and reasonable, manual intervention is less, the result is more robust, the method is more suitable for big data, and automatic learning can be realized.

The invention discloses an unconformity trap reservoir lithology prediction method based on prestack seismic ray impedance inversion. A common image point gather subjected to prestack offset is converted into a radial parameter domain from an offset distance domain; after all common image point gathers are converted, data of a constant radial parameter is extracted and the part of the data is stacked to form a co-radial parameter profile; a high precision seismic reflection coefficient sequence is acquired by iterative inversion; after a relative amplitude value on a reflection coefficient profile is converted into an actual reflection coefficient value, a generalized linear inversion algorithm is used for performing inversion on the reflection coefficient value to acquire radial wave impedance; and radial impendence capable of reflecting the characteristic that the lithology parameter changes along with frequency is used and combined with the exploratory area reservoir physical property and geologic feature to achieve lithology prediction of the unconformity trap reservoir. The radial impendence used by the method is less affected by the range of incidence angle, and the method is more applicable to the unconformity trap reservoir with strong anisotropism, large lithological changes and complex reservoir physical property, and can predict the lithology more accurately.

The invention relates to a low-permeability reservoir comprehensive geological-engineering classification evaluation method. The method comprises the steps that reservoir geological classification evaluation indexes are constructed, specifically, a gray correlation method is used for obtaining reservoir geological main control factors, influencing the fracturing effect, of a low-permeability reservoir; based on the reservoir geological main control factors, low-permeability reservoir geological classification evaluation indexes are constructed by using a principal component analysis method; reservoir engineering classification evaluation indexes are constructed, specifically, geomechanical main control factors, influencing the fracturing effect, of the low-permeability reservoir are obtained by using a grey correlation method; based on the geomechanical main control factors, low-permeability reservoir engineering classification evaluation indexes are constructed by using an analytic hierarchy process; and according to the classification threshold values of the reservoir geological classification evaluation indexes and the reservoir engineering classification evaluation indexes, reservoir categories are divided. According to the method, comprehensive evaluation is carried out from two aspects of geology and engineering, the fracturing effect of the low-permeability reservoir canbe prejudged and evaluated more scientifically and reasonably, and a more scientific and accurate basis is provided for fracturing well selection and fracturing construction parameter selection.

The invention belongs to the field of hydraulic engineering, and particularly relates to a dredging channel back-silting amount predicting method based on a timing sequence analysis-Markov chain method. The method includes the following steps that an original sequence of spatial dispersion is selected; a predicted value sequence is generated; the relative residual sequence of the original sequence and the predicted value sequence is calculated; the relative residual sequence is divided into intervals; Markov chain prediction is performed among the intervals; the water depth prediction interval is averaged to obtain a predicted water depth value, and an original water depth value is substracted from the water depth value to obtain a back-silting value. A timing sequence analysis method and the Markov chain are combined to obtain the timing sequence analysis-Markov chain prediction method, the change trends and random fluctuation of the sequences can be reflected at the same time, and the predicting method conforms to the actual change conditions of the dredging channel back-silting sequence better. By using the predicting method, the relative error rate is only 1.04%, the timing sequence analysis-Markov chain method can improve the precision by 97% compared with a single timing sequence method, and the prediction of the timing sequence analysis-Markov chain method conforms to actual measurement data better compared with the grey theory.

The invention discloses a clastic rock reservoir porosity prediction method based on a diagenetic process. Based on comprehensive research of clastic rock reservoir sedimentary facies, lithology, fluid property, and a diagenetic stage, the method predicts temporal-spatial matching relations of reservoir physical properties and diagenetic actions in a reservoir evolution process, and establishes ageological parameter-diagenesis-porosity prediction model under different geological environment conditions and different diagenesis, so as to predict porosity sizes of the clastic rock reservoir, determine pore spatial distribution of the clastic rock reservoir, and achieves an objective of advantaged reservoir quantitative evaluation.

A photovoltaic output 2D section prediction method based on fuzzy rules comprises the following steps: 1, inputting local electric and meteorology data of a to-be-predicted photovoltaic power generation system, and inputting internet meteorology data; 2, preprocessing various types of data of the inputted data; 3, obtaining the sunny inclined surface global radiation intensity of a to-be-predicted date, and using same to correct the power data of a power sunny model; 4, building a lasting 2D section prediction model; 5, building a 2D section prediction vector, and building a support vector regression 2D section prediction model; 6, building a first fuzzy rule, and using the built fuzzy membership function to modify the lasting 2D section prediction model and the support vector regression 2D section prediction model combined coefficient; 7, building a second fuzzy rule, and adjusting the deviations of the2D section prediction result according to the rules. The method can effectively predict the 2D section, and the prediction result is high in credibility.

The invention provides building and predicting methods of a net photosynthetic rate predicting model based on woody plant leaf phenotypic characteristics and belongs to the field of bioinformatics. The building and predicting methods have the advantages that an average linkage clustering method and a PAM clustering method are combined to divide woody plants at different geographical positions intodifferent sub-varieties, and the differences of the different sub-varieties are eliminated to reasonably predict net photosynthetic rate; meanwhile, a gradient increasing algorithm is used in each sub-variety to build the net photosynthetic rate predicting model on the basis of the phenotype data of leaf samples, the gradient increasing algorithm iterative stopping criteria is put forward for thefirst time during modeling, algorithm iteration residual critical values are provided for the woody plant leaf phenotype data of different sub-varieties, the populus simonii net photosynthetic rate predicting model is built by using the improved gradient increasing algorithm based on the iterative stopping criteria, and an ideal predicting effect on the net photosynthetic rate of different populus simonii varieties can be achieved under a condition that the leaf phenotype data is known.

The invention relates to the technical field of intelligent electric meter fault prediction, and provides an intelligent electric meter fault prediction method and system based on a BiLSTM-CNN model, and the method comprises the steps: obtaining historical fault data of an intelligent electric meter as feature data, including equipment data, operation data, a fault type, fault time, temperature and humidity; the feature data is subjected to data cleaning, the data cleaning comprises cleaning of duplicate values, missing values and abnormal values contained in historical fault data features, and then a BiLSTM-CNN model is constructed; according to the technical scheme, the feature set influencing the fault of the intelligent electric meter is encoded by using the advantage of extracting sequence time sequence information by the bidirectional long and short time memory network and the advantage of extracting local feature information by the convolutional neural network, and finally, the encoded vector is used for classifying and predicting the fault of the intelligent electric meter. The problem that effective information in the fault data of the intelligent electric meter cannot be well extracted by a single neural network model in the prior art is solved.

The invention provides a self-adaptive active sensor tracking method based on tracking precision and risk control, and the method comprises the steps: firstly obtaining the parameters and state information of a tracking target through measurement data sampling, and carrying out the evaluation of the threat degree of the target through a dynamic fuzzy Bayesian network (DFBN) method; predicting thethreat risk in combination with the probability that an active sensor signal is intercepted by a tracking target; secondly, evaluating the tracking precision of the active sensor by using the target prediction covariance and the measurement noise covariance; and finally, constructing an efficiency function by fusing the tracking precision and risk control to reasonably allocate the active sensor resources. According to the method, the threat risk assessment method is improved, and the target threat risk is predicted more accurately and reasonably; secondly, the tracking precision and risk control are fused to construct a self-adaptive active sensor tracking method, the method is more suitable for a complex and changeable real environment, and the target tracking capability is improved.