303 results about "Weather factor" patented technology

The invention discloses an online car-hailing service demand prediction method based on a deep neural network. The method comprises the following steps: S1, dividing a region of a city road network, and calculating and counting the network car-hailing demand of each region; S2, designing a prediction model of the demand of the network car-hailing: based on the deep neural network, learning and training the spatio-temporal characteristics of the demand of the network car-hailing, combining the influence factors of the weather factors on the willingness of the network car-hailing users, and predicting the regional demand of the next period of time; S3, according to the historical data, training the prediction model of regional car call demand, and then predicting the demand of network car call in each region by using the trained model. By combining external factors and spatio-temporal correlation, this method uses depth neural network to express the complex non-linear spatio-temporal correlation characteristics of the demand for network car-hailing, and can achieve high prediction accuracy.

A weather-based financial index is based at least in part on weather. The index may take into account any of a variety of weather factors, such as temperature, precipitation, humidity, number of sunny or overcast days in a period of time, number of freeze days in a period of time, etc. Weather factor value(s) are combined with one or more financial components to provide the weather-based financial index. The index may be traded on an exchange, such as the New York Mercantile Exchange (NYMEX). The value of the index may be calculated based on any period, such as daily, weekly, monthly, yearly, etc. Values of the index may provide insight into the direction of a component of the index, a market or industry corresponding to the component, or the index itself.

The invention discloses a method for predicating the short-period output power of photovoltaic power generation based on a BP (Back Propagation) neural network. According to the method, the BP neural network is adopted for predicating the output power of a photovoltaic power generation system, and the influence of weather factors on the output power of the photovoltaic power generation system is subjected to statistical analysis. The method comprises the following steps of firstly mapping weather types as day types used as the input data of the BP neural network, and utilizing power generation power during each time period of a prediction day as output data; then determining the quantity of hidden layer nodes through formula calculation and repeated cut-and-try operations according to the quantity of input and output units; performing normalization treatment on the input data, performing reverse normalization treatment on the output data, and training the BP neural network by utilizing the treated operating data; and finally predicating the power generation power of the predication day by utilizing a trained model, thereby obtaining the predication result. The data processing method and a prediction model can be used for effectively predicating the short-period output power in photovoltaic power generation under multiple weather types.

A weather-based financial index is based at least in part on weather. The index may take into account any of a variety of weather factors, such as temperature, precipitation, humidity, number of sunny or overcast days in a period of time, number of freeze days in a period of time, etc. Weather factor value(s) are combined with one or more financial components to provide the weather-based financial index. The index may be traded on an exchange, such as the New York Mercantile Exchange (NYMEX). The value of the index may be calculated based on any period or on any geography (or combination of geographies). Historical values of the index or component(s) thereof may be used to predict future values of the index. Values of the index may provide insight into the direction of a component of the index, a market or industry corresponding to the component, or the index itself.

The invention discloses a vehicle detecting device based on multi-kind sensor information. The vehicle detecting device based on the multi-kind sensor information comprises an acceleration detecting unit, a photosensitive signal detecting unit, a geomagnetism detecting unit, a data collection unit, a microprocessor, a storage unit and a wireless transmitting module. When a vehicle passes the detecting device, the acceleration detecting unit collects a road surface shock signal, a photosensitive element induces the illumination variations and generates a photosensitive signal, and the geomagnetism detecting unit induces the changing of the magnetic field and outputs a corresponding electrical signal; analysis and computing are conducted on the signal data by a microcontroller, and relevant parameter information of the vehicle is acquired, transferred to the storage unit, and is sent to a receiving terminal through the wireless transmitting module. The vehicle detecting device based on the multi-kind sensor information has the advantages that vehicle information is detected through the theory of fusing the multi-sensor information, so that detecting accuracy can be improved effectively without being affected by weather factors and sizes of vehicles, installing and maintenance are easy, size is small, cost is low and power consumption is low.

The invention provides an intelligent device and a method for irrigation rapid diagnosis. The irrigation rapid diagnosis method comprises: obtaining soil moisture data, obtaining weather data of the land on the day and in a forecast period, obtaining information of crops, and types of soil; according to the information of crops, the soil moisture data, the types of soil, and weather data, calculating irrigation lower limit rate of the crops and root zone average moisture content in each forecast day in the forecast period, analyzing and comparing the root zone average moisture content and the irrigation lower limit rate, to generate an irrigation suggestion; and showing the irrigation suggestion. Compared with an existing irrigation rapid diagnosis device, the provided diagnosis method is aimed at conditions of different crops in different regions, and gives comprehensive consideration to weather factors and actual water consumption characteristics, and provides relatively scientific and rational irrigation suggestion, so as to overcome defects of irrigating by manual experience to a certain extent.

The invention discloses a load prediction system of a regional power grid and a method thereof, wherein the load prediction system comprises a reverse isolation device, a data processing module, a load classification module, a load prediction module and a database; the data processing module is used for preliminarily processing load data collected by the reverse isolation device into load sample data; the load classification module is used for utilizing FCM fuzzy clustering to carry out classification on the load sample data and selecting mean values of all kinds of load curves as a typical daily load curve; and the load prediction module is used for using the typical daily load curve as an input variable of a BP nerve network, carrying out prediction, and obtaining a load predicted value of the regional power grid. With comprehensively considering influences of season and weather factors on loads, the pre-processing module is improved, the loads are classified based on the FCM fuzzy clustering, the typical daily load curve is accurately found, the load predicted value of the regional power grid is obtained by the BP nerve network, and the inherent law of history load data is fully extracted, so that the prediction precision is greatly improved.

The invention discloses a method for forecasting public building air conditioner short-time base wire load with consideration of real-time weather factors. The method comprises the following steps: S1, calculating correlation between temperature, humidity, air speed, rainfall amount and the air conditioner load by using a Pearson correlation coefficient formula, and selecting strong linear correlation parameters as the real-time weather factors considered during forecasting of the public building air conditioner short-time load base wire; S2, respectively calculating the daily temperature-humidity index THI and a weighed temperature-humidity index THI of the forecasting date and two months before the forecasting date and selecting a typically-similar date of the forecasting date; and S3, calculating the air conditioner load value of 24 hours of the forecasting date by using a BP neural network method. The method can be used for forecasting the public building air conditioner short-time base wire load on a national scale and can also be used for providing theoretical support and data support for applying the typical public building air conditioner load to the regulation of power grid load.

The invention relates to a system and method for managing roadside parking spaces by means of the linkage of a radar and intelligent cameras. According to the system and method, whether vehicles entering or leaving parking spaces exist is detected through the radar, the information of the vehicles entering or leaving the parking spaces is tracked and shot through an intelligent camera group; if the parking space entering or leaving information of the vehicles detected by the radar is consistent with the information of the vehicles tracked and shot by the intelligent camera group, a main controller sends the parking space entering or leaving information of the vehicles and the information of the vehicles to a data processing platform; and if the parking space entering or leaving information of the vehicles detected by the radar is inconsistent with the information of the vehicles tracked and shot by the intelligent camera group, the main controller sends an instruction to an enhancement processing module for secondary processing. According to the system and method of the invention, the radar, intelligent camera group and storage equipment are adopted to perform real-time monitoring on the roadside parking spaces, and therefore, hidden dangers caused by the inaccuracy of the information of the vehicles entering or leaving parking spaces due to factors such as instrument errors, shooting angles and weather factors can be eliminated, and the accuracy of parking management can be improved.

The invention belongs to the field of traffic engineering, and discloses a multi-factor short-term traffic flow prediction method based on neural network LSTM. The multi-factor short-term traffic flowprediction method based on the neural network LSTM comprises the following steps: step 1, obtaining traffic flow data of a period of time, and preprocessing the traffic flow data to obtain short-termtraffic flow data; step 2, screening the short-term traffic flow data according to weather records and holiday records, and dividing data sets; step 3, performing data cleaning, data reconstruction,and normalization; and step 4, establishing an LSTM neural network model, selecting the data set according to the weather conditions and holiday conditions of the date to be predicted, using the selected data set to train the LSTM neural network model and adjust the LSTM parameters, and obtaining the traffic flow of the date to be predicted based on the established LSTM neural network model. The invention provides a more detailed idea, excludes influences of other factors on the traffic flow, such as weather factors and holiday factors, and relatively improves the prediction accuracy, so thatthe traffic flow prediction of a certain period in the future is more accurate and effective.

The invention provides a short-term load prediction method. The short-term load prediction method comprises the following steps: acquiring a value of a precast weather factor of a to-be-predicted date and a value of a weather factor of a plurality of historical dates; correcting the value of the precast weather factor of the to-be-predicted date and the value of the real weather factor; acquiring an electric network load value of a plurality of historical dates and short-term load prediction data, carrying out the comparison, and solving the error; fitting the corrected value of the real weather factor of a plurality of historical dates and the electric net load value to obtain a relation model; utilizing the error of the electric network load value and the short-term load prediction data to correct the relation model; predicting the electric network load of the to-be-predicted date according to the corrected relation model. Since the accumulation effect of the weather factor and the error of the historical short-term load prediction data and the real electric network load data are considered in the method, the prediction precision of the electric network load can be further improved.

The invention discloses a road travel time forecasting method based on a random forest and a clustering algorithm. In the road travel time forecasting method, according to the time sequence rule of historical traffic data, combined with the road property, weather factors, holiday information and states of road upstream-and-downstream traffic flow, and through a hybrid forecasting model of the density-based clustering algorithm (DBSCAN) and the random forest (RF), travel time of all key road sections at some time interval is accurately forecasted. The forecasting result can be used for pre-judging a traffic state development tendency and making a control scheme for potential congestion roads, can also be used for dynamic path induction, can project best travel plans for travelers, and can assist in social intelligence traveling. According to the road travel time forecasting method, the forecasting accuracy of all trees in the random forest is increased through density clustering, and therefore the forecasted whole accuracy is increased.

The invention provides a near infrared spectrum technology-based multiband soil nitrogen detection device and a near infrared spectrum technology-based multiband soil nitrogen detection method. A regression equation of near infrared spectrum data and soil sample nitrogen content is established by establishing a soil nitrogen prediction model and using soil spectrum absorbance of seven bands so asto accurately measure the soil nitrogen level. The device and the method have high practical precision, are not affected by weather factors, are suitable for real-time field measurement, provide storage function of a U disk, and are convenient for storing historic records, calling the data and realizing a soil nitrogen distribution graph.

The invention discloses a vehicle detecting device based on multiple shock detecting sensors. The vehicle detecting device based on the multiple shock detecting sensors comprises a shock detecting unit, a shock signal adjusting circuit, a data collection unit, a microprocessor, a storage unit and a wireless transmitting module. The shock detecting unit, the shock signal adjusting circuit, the data collection unit and the microprocessor are connected in sequence, and the storage unit and the wireless transmitting module are connected with the microprocessor respectively. In the running process of a vehicle, continuous stimulating is generated to the ground by vehicle shafts through vehicle wheels, the generated road surface shock is detected by the multiple shock detecting sensors, the detected signal after passing through the signal adjusting circuit is input to a microcontroller, shock data are processed by the microcontroller to acquire shock source positioning, vehicle shaft positioning data are further acquired, and vehicle information such as the number of the vehicle shafts, wheel base, speed of the vehicle, the vehicle type classification and the like is acquired. The vehicle detecting device based on the multiple shock detecting sensors has the advantages of not being affected by weather factors such as rain, snow, fog and the like and boundary dimensions of the vehicles, and being simple to install and maintain and stable in operation.

The embodiment of the invention provides a phase component fault range finding method based on mu PMU distribution line parameter identification; the method mainly comprises the following steps: usingthe mu PMU on two ends of the line to gather the voltage current phasor data before and after the fault; carrying out Fourier decomposition for the extracted stable state voltage current so as to obtain the fundamental wave component of the phase value; using the fundamental wave components of the three-phase voltage currents on two ends of the line to determine the line parameters; using the phase components on two ends of the line after the fault to build a matrix equation, and combining the constrained condition and the line parameters so as to determine the unique fault distance. The method introduces the mu PMU to determine the line parameters in real time, thus reducing line parameter errors caused by field construction, line aging and weather factors, solving the range finding errors caused by non-synchronization of the information on two ends, preventing the complex circuit transient analysis, and reducing the computational complexity; the range finding method can use the dual-end impedance method to eliminate the range finding errors caused by transition impedance, and the range finding precision is not affected by the neutral point grounding mode.

The invention discloses an enterprise power consumption maximum demand prediction method based on ARIMA and SVM, and the method is characterized in that the method comprises the steps: reading from apower grid company charging table TSDB; supplementing missing values by using a moving average method; removing outliers by using a K-Means clustering algorithm; predicting power consumption by usingan ARIMA time sequence; in combination with prediction of weather factors and production conditions, using a trained SVM model to perform decision making; and taking the maximum value of the predictedmonth, and calculating the maximum demand of the future month. According to the invention, the time sequence of historical electric quantity is fully considered, which includes trends, periodic and seasonal; meanwhile, prediction of weather factors and production conditions is also considered; two factors are unified in one model by using a machine learning decision and a time sequence predictionalgorithm, an accurate result is obtained through training, and along with more and more data and longer and longer use years, the prediction error of the maximum demand of the future month is smaller, so that more electric charge is saved for enterprises.

The present invention discloses a traffic accident scene safety protection and monitoring method and device. The method comprises the steps of: employing a camera to collect road surface images of a traffic accident scene, and determining a limit of the prohibited area according to the position information and the shooting direction of the camera and the road surface images; and employing a radarto monitor whether there is a vehicle entering the limit of the prohibited area or not, and if it is monitored that there is a vehicle entering the limit of the prohibited area, performing alarm processing. The traffic accident scene safety protection and monitoring method and device achieve construction of an accident lane prevention area boundary with clear boundaries in logic so that the traffic accident scene protection early warning response efficiency is improved, dangers when the people's police perform reconnaissance and cone barrels are arranged at the accident scene are effectively avoided, influences of weather factors on a current safety protection mechanism are avoided, injuries and deaths of the people's police and the masses caused by secondary accidents are avoided.

The invention relates to a short-term power prediction method for a photovoltaic power generation system. The method comprises the steps of re-dividing weather types: according to the factors that whether the variables of the natural environment are continuous and stable or not and the averaged values of the variables, dividing weather types in the generalized point of view; determining a similar day of a to-be-predicted day: determining a generalized weather type corresponding to the to-be-predicted day according to weather factors, selecting a historical day similar to the to-be-predicted day according to the grey correlation method, adopting the photovoltaic power output of the historical day as a model input, and selecting daily weather feature vectors; defining the degree of similarity between the to-be-predicted day and the above particular historical day; establishing a photovoltaic power generation short-term power prediction model based on a support vector machine, and optimizing the parameters of the prediction model according to the genetic algorithm. The above precision method is high in precision, and fast in convergence.

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.

Space-based power system and method of altering weather using space-born energy. The space-based power system maintains proper positioning and alignment of system components without using connecting structures. Power system elements are launched into orbit, and the free-floating power system elements are maintained in proper relative alignment, e.g., position, orientation, and shape, using a control system. Energy from the space-based power system is applied to a weather element, such as a hurricane, and alters the weather element to weaken or dissipate the weather element. The weather element can be altered by changing a temperature of a section of a weather element, such as the eye of a hurricane, changing airflows, or changing a path of the weather element.

The invention discloses an intelligent power grid park terminal user energy demand condition dynamic prediction system and a method. The method includes: conducting a main constituent analysis of meteorological factors for influencing the cooling and heating load demand of intelligent power grid park terminal users; converting related variables to a few of linear independent random variables; quantifying the weather factor and the day type, conducting an analysis with historical load data by employing a fuzzy clustering method, and forming a sample; representing load characteristics of various types of loads and various types of distributed energy supply systems in the intelligent power grid park in load curves; and finally solving a model according to the process of a BP neural network algorithm, and obtaining a cooling and heating load prediction result. The system comprises a main constituent analysis module, an analysis sample formation module, a load characteristic curve module, and a load prediction module. According to the method and the system, the network size is reduced, the prediction precision is improved, and advantages of a BP neural network for large-scale parallel processing and adaptive learning ability are fully developed.

The invention discloses a coordinated operation optimization method for a comprehensive energy system containing a heat pump, and the method comprises the following steps: building a daily load prediction model of the comprehensive energy system based on a nonlinear regression analysis method and a neural network prediction theory according to weather and season historical data and cold, heat andelectric load historical data; analyzing the influence of weather factors, seasonal factors and hot water load on the energy efficiency of the heat pump unit, and establishing an operation characteristic mathematical model of the heat pump under different working conditions; inputting actual weather and season information to obtain a daily load prediction result, and determining a matched heat pump operation characteristic model according to the weather and season information and the hot water load prediction result; and constructing an operation optimization model taking the minimum operationcost of the comprehensive energy system as a target, and carrying out optimization solution on the target function to obtain an optimized operation scheme of the comprehensive energy system. The influence of various factors on the actual energy efficiency of the heat pump is fully considered, the source side heat pump and other devices can be coordinated, and optimal operation of the comprehensive energy system is achieved.

A street lamp illumination control system comprises node controllers, a concentrator and a control center; each node controller controls and connects a street lamp; the node controllers are connected with the concentrator; the concentrator is connected with the control center through a wired mode or a GPRS wireless mode; the street lamp control system uses a terminator control mode to ensure a lighting time to precisely follow seasonal changes; an infrared ray measurement and control instrument on the node controller can monitor road vehicle number in real time, and light illumination or brightness adjustment can be made according to the vehicle flow in real time; the system is flexible, and energy saving effect is obvious; an illumination sensor on the concentrator can detect illuminance in real time, so the system can provide corresponding light supplement when visibility is insufficient because of nature weather factors like black clouds or haze or solar eclipse in day time, thus satisfying illumination needs for road driving safety.

The invention provides a medical big data based medical insurance actuarial system and a medical big data based medical insurance actuarial method. The method includes: searching disease dates in medical data; acquiring historical weather information corresponding to the disease dates in the medical data from a weather information platform, and associating the medical data with the historical weather information; analyzing the medical data associated with the historical weather information to obtain patients susceptible to weather factors; classifying the patients susceptible to the weather factors according to preset age group dividing rules, and calculating incidences of diseases susceptible to the weather factors according to quantity of the patients, corresponding to each age group, susceptible to the weather factors, and calculating health insurance premiums of the patients, of each age group, susceptible to the weather factors according to the incidences of the diseases and preset medical insurance actuarial algorithms. By implementation of the medical big data based medical insurance actuarial system and the medical big data based medical insurance actuarial method, claim risks of medical insurances are reduced, and profitability of insurance companies is improved.

Weather metrics data and a weather factor relationship knowledgebase are received and used to forecast the weather-based demand. The weather factor relationship knowledgebase can be a weather-impact model. The weather-impact model can comprise at least one of an empirical scoring matrix, a weather indices template, and a proxy model conditions template. The weather-impact model can be derived from an analysis of normalized proxy sales history data. Optionally, the weather-based demand can be scaled, deaggregated, or both.

A method and an apparatus for controlling a temperature based on scheduling of an indoor temperature are provided. The apparatus obtains set temperature history information of an air handling unit from previously stored information. The set temperature history information is mapped to weather factor history information about at least one weather factor. The apparatus also obtains weather factor information regarding a scheduling target time. Based on the obtained set temperature history information and the obtained weather factor information, the apparatus schedules a set temperature of the air handling unit with regard to the scheduling target time. The present disclosure relates to a sensor network, Machine Type Communication (MTC), Machine-to-Machine (M2M) communication, and technology for Internet of Things (IoT). The present disclosure may be applied to intelligent services based on the above technologies, such as smart home, smart building, smart city, smart car, connected car, health care, digital education, smart retail, security and safety services.

The invention discloses a preprocessing method based on an inspection image, and the method comprises: an image defogging algorithm which removes the impact on the image quality from a weather factor,and guarantees the original features of an image; an image segmentation algorithm which is used to obtain a condition area satisfying uniformity and connectivity, and extract an interested target ora meaningful area; an image denoising algorithm for removing a large amount of noise introduced in the imaging process and ensuring the original characteristics of the target object; an image enhancement algorithm, in whch image enhancement processing is used for optimizing the image quality, emphasizing the region of interest in the image and improving the readability of the image; and an image restoration algorithm which is used for processing the image which is degraded due to distortion, blurring, distortion or noise mixing to obtain a source image which is restored as much as possible. The technical scheme of the embodiment of the invention is used for preprocessing the inspection image, improving the quality of the inspection image, highlighting the characteristics of the target object, facilitating the later intelligent recognition and classification, and greatly reducing the cost compared with a mode of processing only depending on manual discrimination.

Disclosed is an automatic feeding system for an aquaculture pond. The aquaculture pond is rectangular, the system comprises a feeding hull, a controller and a feeder, the feeding hull is of a twin-hull structure and comprises two parallel boat-shaped buoys and a deck, the two parallel boat-shaped buoys float on the water and are covered and connected into a whole by the deck, an electronic compass, a wireless signal receiver, the controller and the feeder are mounted on the deck, two wireless signal transmitters are mounted on a buttress of the pond at intervals, two screw propellers are symmetrically mounted on the left side and the right side of a stern of the feeding hull, one screw propeller is arranged on each boat-shaped buoy, the electronic compass monitors the instant course of the feeding hull and feeds the instant course back to the controller, the wireless signal receiver receives wireless signals transmitted by the wireless signal transmitters and feeds the wireless signals back to the controller, and the controller controls running of the two screw propellers. Positioning and orienting of the feeding hull in the patrolling process are not affected by the buttress of the pond and weather factors, and sailing and steering stability is better.

The invention discloses a power grid high low temperature refined early warning method in combination with dynamic correction, comprising the steps of: S1, forecasting local weather to generate highest and lowest temperature forecast results and extracting live weather factors influencing local temperature forecast results; S2, establishing highest and lowest temperature forecast equations in dependence on generated temperature forecast results and selected live weather factors; S3, counting highest and lowest temperature forecast errors based on actually measured highest and lowest temperature information and forecast results, and correcting the highest and lowest temperature forecast results based on the temperature forecast errors to obtain final forecast results of highest and lowest temperatures; and S4, performing electric power equipment priority forecasting based on the final forecast results of highest and lowest temperatures and publishing early warning information. The power grid high low temperature refined early warning method can effectively solve the problem of great differences between a power grid system high temperature and low temperature forecast effect and an actually measured result, timely and effectively take corresponding measures to cope with extremely high and low temperatures, and mitigate damage caused by extreme weather.

The invention discloses a remote sensing satellite fire point identification method, device and equipment and a storage medium, and relates to the technical field of artificial intelligence. Fire point recognition is carried out through the fire point recognition model, high accuracy and recall rate are achieved, complex preprocessing does not need to be carried out on satellite remote sensing data, dependence on expert knowledge is low, and the method is suitable for different remote sensing satellites. Weather information and earth surface type information are considered on the basis of a remote sensing satellite image, the remote sensing satellite image is prevented from being influenced by weather factors and earth surface types, and the method is suitable for fire point identificationof different area scenes. In the fire point recognition model training stage, method of learning by migration, on the basis of a first fire point identification model trained by a first training dataset formed by disclosed fire point identification product data, the first fire point recognition model is finely adjusted through the second training data set formed by a small number of real fire cases, the problem of lack of real data is solved, and the finally obtained fire point recognition model has high accuracy and recall rate.