63results about How to "Predictive" patented technology

The invention provides a traffic jam avoiding prompting system based on a collective intelligence network. The traffic jam avoiding prompting system consists of a webpage traffic hot spot digging module, a microblog traffic hot spot digging module, a road condition rule predicting module, a jam hot spot verification and management module, a real-time traffic hot spot server, a traveling route planning module based on hot spot avoiding, a traffic hot spot and avoiding route display module and intelligent mobile phone end application software. The traffic jam avoiding prompting system faces the collective intelligence sensing network, is combined with the traditional traffic information sources, carries out digging and prediction on the urban traffic hot spots on the basis of the data activation technology, and further provides the traffic jam hot spot display and avoiding prompt service for intelligent terminal users. The traffic jam avoiding prompting system has the advantages that the data is from the multi-source collective intelligence network, the accuracy is good, the checking comprehensive rate is high, the hot spots are captured and fed back in real time, high efficiency and practicability are realized, the design is good, the operation is stable, the installation is convenient, the traffic jam avoiding prompting system can become a tool capable of providing wide and practical intelligent traffic information for urban residents, and the efficient information prompt is provided for the users to realize the smooth traveling.

The present invention provides a method for automatically controlling partial functions of an air conditioner and a heat pump water heater based on weather forecast data. According to the method, mainly the control on an air conditioner and the big data of weather forecast are combined, such that the air conditioner can predict and automatically control the opening and the closing of the electricheating tape of a compressor during the standby of the air conditioner according to the temperature change in weather forecast, the amount of new air introduced by an indoor unit can be automaticallycontrolled according to outdoor air quality report data while whether an indoor heat exchanger is clogged by dirt can be judged by combining historical air quality data and the gas discharging temperature or the pressure data of the compressor, the dehumidification period or logic can be automatically established according to local seasonal climate, or the evaporation pressure during cooling can be adjusted sp as to make the air conditioner dehumidify according to the concrete humidity condition, the temperature change at the current day can be predicted, the water boiling temperature can be adjusted, the running period achieving the highest energy efficiency can be determined, and the time for boiling water can be calculated and is transmitted to a client terminal.

The invention discloses a multi-phase gait switching control system and control method for a power-assisted exoskeleton robot. The multi-phase gait switching control system is characterized in that the multi-phase gait switching control system includes a sensor module, a signal acquisition and processing module, a gait switching motion control module and a motor driving module; and sensors comprise inertial measurement units, plantar pressure sensors and motor encoders. With the multi-phase gait switching control system and control method for the power-assisted exoskeleton robot of the invention adopted, the output of lower hip joint and knee-joint motor power-assisted torque under various of phase gaits and smooth switching between the phase gaits can be planned under a condition that the exoskeleton robot bears a heavy object, and the coordination and flexibility of the movement of the limbs of the weight-bearing exoskeleton equipment can be realized.

The invention discloses an intelligent Operation and Maintenance Analysis System Based on Multi-source Heterogeneous Data Fusion, Machine Learning and Customer Service RobotBased on multi-source heterogeneous data fusion, In the intelligent operation and maintenance analysis system of machine learning and customer service robot, the analysis and processing module uses the multi-source heterogeneous data fusion technology to preprocess and normalize the real-time data. After the correlation analysis, the model matching is used to output the results, and the root cause of the problem is determined by root cause analysis. The model management module is used for acquiring data for machine learning and outputting various reasoning models to the analysis processing module and the recommendationdecision module. The recommendation decision module recognizes the context and matches the corresponding subsequent processing behavior according to the policy. The invention is based on the multi-source heterogeneous data fusion technology and the operation and maintenance mode of machine learning, which gives consideration to the compatibility problem of the big data, improves the automation degree of the operation and maintenance, guarantees the network safety and the service quality, simultaneously has certain prediction ability, and reduces the related operation and maintenance cost. Theoutput mode of the customer service robot further realizes the intelligence of operation and maintenance.

The invention discloses a self-adaptive control method for diagonal gait of a four-footed robot, which is used for self-adaptively controlling the walking of the four-footed robot in the nonstructural terrain with the diagonal gait. The self-adaptive control method bases on a fuzzy reasoning learning method and a foot trajectory real-time correction method and takes the diagonal gait as the motion mode so as to comprehensively control a robot body to adapt for the nonstructural environment. The gait planning is carried out on the environmental information collected by the sensor through the fuzzy reasoning learning method; the stability judgment is carried out on the gait planning information; if the planned four-footed robot has stable diagonal gait, the joint corner information is sent to a control system; if not, the supporting time of the diagonal feet is corrected by the foot trajectory real-time correction method so that the four-footed robot has better stability when walking under the changed gait in the nonstructural terrain. The self-adaptive control method is used for correcting foot trajectory planning and walking stability under changed gait when the four-footed robot walks diagonally in a dynamic way, thus realizing the self-perception, self-correction and self-adjustment of the four-footed robot.

The invention proposes a method for planning the real-time motion of a mobile robot based on human behavior simulation, and the main contents of the method comprise a multilayer personification behavior path planning frame: simulating the functions of a human brain, human eyes and human legs based on the behavior mode of human planning, correspondingly setting a global layer path planning, sensinglayer path planning, and execution layer path planning; a completeness adjustability path planning algorithm which guarantees the completeness and is high in efficiency, wherein a track planned through the algorithm can meet the kinetics kinetic constraint and environment constraint conditions of the robot; and a dynamic monitoring and recovery behavior strategy: carrying out the real-time dynamic monitoring of a safety distance range in front of the robot, and taking a recovery behavior strategy during emergency, thereby effectively guaranteeing the safety and improving the robustness. The method can achieve the real-time motion planning of the mobile robot in large-scale, dynamic and unstructured indoor environments with an unknown part.

The invention discloses a PEMFC electric pile temperature control method and system. The PEMFC electric pile temperature control method comprises the steps of: acquiring a volt-ampere characteristic curve of a PEMFC electric pile model, and enabling the target optimal temperature under different current values to serve as an initial temperature value; regulating and controlling PID control parameters of the PEMFC stack model according to the gain change rate of the PEMFC stack inlet and outlet temperature relative to the PEMFC stack state change at the adjacent moment and the target current value; and controlling the inlet and outlet temperature of the PEMFC stack according to the PID control parameter and the initial temperature value. A model prediction control method and a PID parameter adjustment method are adopted to accurately control the output value of the inlet and outlet temperature of the PEMFC stack, temperature oscillation is reduced, the hysteresis quality of temperature control is solved, short-time high temperature in the PEMFC stack is avoided, and the real-time performance of temperature control is improved.

A process gas temperature control method for a loose resurgence roller comprises the following steps of firstly setting a process gas temperature value SP, a water adding amount standard value d and awater adding amount acquisition period T of a roller, then detecting the process gas real-time temperature PV and instantaneous water adding amount dn of the roller; performing real-time comparison on PV and SP, and performing real-time output of steam positioning valve aperture y1 through operation of a temperature PID controller; according to a preset water adding amount acquisition period T, calculating a difference between the instantaneous water adding amount dn and a water adding amount dn-1 of a previous period, and performing real-time output of a steam positioning valve aperture ywater for compensating through operation of a water adding amount compensation module; and finally calculating a steam valve aperture setting value y and outputting to the steam valve positioning devicefor controlling the injection amount of steam. According to the method, pre-estimation can be performed on water adding amount change of a tobacco block, and aperture of the steam fixing valve can beadjusted in advance, thereby quickly estimating influence of water adding amount to the process gas temperature, and reducing fluctuation amplitude of the process gas temperature.

The invention discloses a method for determining carbon sinking capacity of a mangrove forest. The method comprises the following steps: collecting surface soil of the lower covert of the mangrove forest and bare land nearest to the mangrove forest in an area to be measured and mangrove plant leaves; determining a ratio of organic carbon in the soil and stable carbon isotopes (delta <13>C) of the plant leaves; taking the bare land as the background value, and calculating a ratio of mangrove-sourced carbon in the soil organic carbon in the area to be detected and in the artificial mangrove area with known ages of the mangrove according to a relevant formula; obtaining a multiplier index of the organic carbon of the mangrove plants according to the age of mangrove forest, establishing an index multiplier equation, obtaining a mangrove forest organic carbon multiplier rate model by virtue of derivation, and calculating the carbon sinking capacity of the target mangrove forest at a specific age of the forest according to the content of the organic carbon in the surface soil under the lower convert of the target mangrove species. The method is simple, easy and free from damaging the forest; and the obtained result has a prediction function, and the carbon sinking capacity of different species with different ages in different environments can be accurately determined, so that comparative data can be provided to the research of the global variation.

The invention is applicable to the technical field of wind tunnel tests, and provides a wind tunnel test scheduling method and system based on deep reinforcement learning. In the wind tunnel test scheduling method and system, the utilization rate of the branch pipeline and the valve opening delay time of the branch pipeline are considered, wherein one variable of the target function is the utilization rate of the branch pipeline, therefore, the distribution of the wind tunnel test scheduling pipelines can be globally considered; in the distribution, the selection of the next execution action is obtained by maximizing the target function, so that the scheduling method has predictability, the utilization rate of power resources can be maximized, the opening/closing times of branch pipelinescan be minimized, and the loss of power equipment is reduced; and the other variable of the target function is the valve opening delay time of the branch pipeline, so that the prediction accuracy canbe improved, and the smooth experiment can be ensured.

The invention discloses a double-lane vehicle lane changing guiding method under vehicle-road cooperation, and the method comprises the steps: describing the state information of a vehicle based on acellular automaton model, carrying out the real-time effective evaluation of the threat to the vehicle in real time according to the information of surrounding vehicles, carrying out the traversal ofthe elements of a threat evaluation function, obtaining the position of a minimum value, and giving the vehicle lane changing guidance; in a double-lane environment, subjecting the observation vehicleto vehicle speed induction according to the movement trend of the forward vehicle after lane changing, so that the vehicle speed of the observation vehicle is associated with the acceleration trend of the forward vehicle after lane changing; referring to the motion trend of the backward vehicle, observing that the vehicle adopts a leading acceleration motion rule, and avoiding rear-end collisionaccidents while ensuring motion synchronization; providing early-stage speed guidance for the vehicle close to the blocking point, and avoiding the phenomenon of vehicle starting and stopping; and finally, giving a double-lane vehicle lane changing rule under vehicle-road cooperation to complete lane changing. According to the invention, the problems of vehicle congestion and road section congestion point increase in the future traffic environment are solved.

The invention discloses a method for detecting environmental pollution caused by fertilization, and establishes a critical threshold model for environmental pollution of organic fertilizers, by pre-detecting the content of various nutrients and heavy metal components in an organic fertilizer, and the loss of various nutrients and the heavy metal components in the crop planting environment, and combining with the demand for nutrients and the heavy metal components in crops, the amount of the organic fertilizer needed to meet the growth of crops is calculated, and the method can predict whetherthe application amount of the organic fertilizer brings heavy metal pollution to the planted soil or not. The method of the present invention is predictive, and the accuracy for assessing the heavy metal pollution caused by fertilization is as high as 100%, and the method can be used to pre-predict and adjust a fertilization scheme before fertilization, and avoids the problem of heavy metal pollution caused by fertilization and the problems of nitrogen saturation and P pollution.

The invention provides a method and a device for detecting pe file abnormal behavior, a medium and equipment. The method comprises the following steps: receiving a file to be detected; extracting thefeature data of the file to be detected; processing the feature data to obtain valid data; predicting whether the file to be detected is an abnormal file or not on the basis of a prediction model trained by a support vector machine ONECLASS model according to the valid data to obtain a prediction result; outputting the prediction result to a user. By adopting the prediction model trained by the support vector machine ONECLASS model, 0-day attacks can be recognized accurately. The method has the advantages of automation, timeliness, effectiveness and the like which are incomparable with the traditional method.

The invention relates to a metering method of an intelligent metering device for the water content of oil well produced fluid. The metering method is technically characterized in that a liquid-phase multi-element water content fluctuation signal of oil-water mixed liquid obtained after gas-liquid separation is obtained through a multi-electrode high-frequency capacitive sensor, and a deep LSTM-CNN multi-channel fusion network is adopted for extracting water content features in the multi-element water content fluctuation signal collected by the sensor layer by layer and carrying out feature fusion. The water content of the oil well produced fluid is predicted by taking a water content test value as a label and taking the features obtained after network fusion as network input. The adopted deep neural network has objectivity in predicting the water content, uncertainty and subjectivity of manual operation in a test are eliminated, and after the deep convolutional neural network and the long and short-term memory network are trained through a large amount of data and the number of iterations, the data can be predicted strongly and objectively.

The invention relates to a road adhesion coefficient estimation method and system based on a laser radar. The method comprises the following steps: S1, acquiring point cloud information acquired by a vehicle-mounted laser radar in vehicle driving under different illumination conditions and different road types; s2, road surface area point cloud in the point cloud information is extracted; s3, establishing a road surface reflection intensity distribution model parameter database based on the reflection intensity probability distribution model parameters of the road surface area point cloud; and S4, obtaining actually measured road surface point cloud data, and performing road surface type identification and adhesion coefficient estimation by using the distribution similarity. Compared with the prior art, the method has the advantages of high estimation precision, strong robustness and the like.

The invention discloses a method for solving a time-varying continuous algebraic Riccati equation based on an error adaptive coefficient zero neural network, which comprises the following steps: firstly, abstracting and modeling original problems, such as mechanical arm motion planning, image and signal processing and the like, and converting the problems into a mathematical model for solving the time-varying continuous algebraic Riccati equation; converting the mathematical model into a time-varying matrix linear equation problem through derivation and Kronecker product, using an error-based adaptive coefficient zero neural network for iterative solution, performing mapping and adaptive transformation continuously on system residual errors and state variables in the solution process, and adjusting the solution strategy of the network; finally, stopping iteration and outputting the model system when the model system obtained by iteration solution meets predefined conditions and precision. Compared with other traditional methods for solving the time-varying continuous algebraic Riccati equation, the algorithm provided by the invention has higher convergence precision and robustness.