139results about How to "Solve the hysteresis" patented technology

The invention provides a method and a device for detecting malicious flash files. The method for detecting malicious flash files includes: collecting operation information, generating shell codes, of flash files to be detected by hooking function hooks, executing different detection operations according to function operation information generating a memory reading operation and function operation information generating a memory writing operation; performing legality detection on parameter calling of the function operation information generating the memory reading operation, performing injection attach detection on effective memory space of the function generating the memory writing operation, and accordingly, determining whether the flash files to be detected are malicious or not. The malicious flash files can be detected automatically without manual involvement and maintenance of a virus base by the dynamic detection, maintenance cost is reduced greatly, and coverage rate is increased. Further, whether any flash files to be detected are malicious can be detected without depending on a pre-collected virus base, and the problem of lagging is solved.

The invention discloses a typhoon center positioning method and device, and a typhoon path generation method. The typhoon center positioning method comprises the steps of firstly, acquiring bottom-layer wind field data, height field data and sea level air pressure data; then, identifying a low-voltage center of a monitoring area; judging whether an overlapping area exists between the low-pressurecenter and a vortex center of a bottom-layer wind field or not, whether a peripheral maximum wind speed value is greater than or equal to a preset threshold value or not and whether a central air pressure value is less than a preset air pressure value or not; and if so, taking as a candidate point of the typhoon center, judging whether a maximum value center and a minimum value center of an air pressure gradient in a horizontal direction and a vertical direction exist around the candidate point of the typhoon center at the same time, if so, conforming to typhoon characteristics, and taking thecandidate point of the typhoon as the typhoon center point. By implementing the embodiment of the invention, the typhoon center can be positioned before the formed typhoon is observed through a real-time cloud picture, then the real-time typhoon path is detected, and a problem of hysteresis of an existing typhoon center positioning technology is solved.

The invention discloses a rock and earth mass frost heaving parameter in-situ tester. The tester comprises a test probe, a control device and a data collection device. The test probe comprises a hollow shaft, an elastic air bag, a thermal insulation layer, a heat exchange sheet and a protection sleeve which are distributed in order from inside to outside. The upper end and the lower end of the elastic air bag are fixed on the hollow shaft through flanges. A frost heaving force monitoring device and a frost heaving displacement monitoring device are arranged between the hollow shaft and the air bag. The control device comprises a temperature control device and a pressure control device. The temperature control device is connected with the heat exchange sheet through a thermal insulation pipe. The pressure control device is connected with the elastic air bag through a pressure pipe. The data collection device is connected with a sensor arranged on the test probe. The tester has advantages of accurate result, low cost, easy operation and the like.

The invention discloses a webpage backdoor detection method and system based on non-credit-granting. The method includes the steps that firstly, a credit-granting framework and a non-credit-granting framework of a website are established; URLs and URL pages in the website are traversed and recognized sequentially, whether the URLs are in the credit-granting framework or not is judged, if the URLs are in the non-credit-granting framework, whether the URLs are completely matched with the URL pages stored in the credit-granting framework in related attribute or not is judged, if the URLs are not completely matched with the URL pages stored in the credit-granting framework in related attribute, the URL pages are tampered, and related processing operations are conducted; if the URLs are not in the credit-granting framework, the hashed values of the URL pages are calculated, whether the URLs are in the non-credit-granting framework or not is judged, if the URLs are in the non-credit-granting framework, the URL pages are webpage backdoors, otherwise, whether the URL pages have website script characteristics and executable permissions or not is judged, and further judgment is conducted. According to the technical scheme, the webpage backdoor detection method and system can perform effective reorganization on the webpage backdoors in the non-credit-granting framework, and the webpage backdoors comprise unknown or encrypted webpage backdoors. According to the webpage backdoor detection method and system, the detection efficiency is high, the false alarm rate is low, and implementation is more convenient.

The invention discloses a financial automation method based on a business economic service process, comprising the steps of: determining the type of a business economic service; obtaining a variationparameter corresponding to the business economic service, selecting a built-in parameter according to the variation parameter, and automatically calculating a financial value; generating an accountingdocument according to the financial parameter and the business economic service, wherein an accounting subject in the accounting document corresponds to the business economic service; and automatically adjusting the general ledger and the subsidiary ledger corresponding to the accounting subject and related information in a standard financial statement according to the accounting document. Themethod achieves the integrated management of service process and financial automation.

The invention discloses a tidal lane driving direction dynamic control method. The method comprises the steps that road basic data, real-time traffic flow rate and predicted number of queue vehicles are input; according to saturation in two directions, road traffic conditions are divided into five modes; according to a corresponding condition, a corresponding control model is called to solve an optimal control scheme; whether the optimal control scheme meets an optimization condition is determined; and if the optimal control scheme meets the optimization condition, the optimal scheme is output to road infrastructure and is carried out, otherwise the existing scheme is output. According to the invention, a model which dynamically changes the driving direction of a lane according to the real-time traffic flow rate and the predicted number of queue vehicles is provided for the first time; the method is suitable for a variable-lane road with more than 2 lanes; and the method has the advantages of high road resource utilization, good economic benefit, being universal and the like.

The invention discloses a network detection optimization method based on ant colony algorithm. At first, according to the traffic distribution of each node in the network, When the flowing ant arrivesat a node, the flowing ant generates the detection ant to detect the surrounding network. After the pheromone of the path, node energy consumption and link delay information is mastered, the flowingant chooses the next hop with certain strategy, and constructs the cruising path to the target heuristically. After arriving at the destination node or exceeding the maximum lifetime, the flowing antsubmits the recorded path information, which constitutes the path set of this iteration, calculates the total energy consumption of the topological subset, and then determines the amount of pheromone,so that the flowing ant returns to the original path to release pheromone. After several iterations, traffic gradually concentrates on part of the path to generate a low-energy topology subset to meet the data transmission needs, so that other links can be shut down to achieve the purpose of energy conservation.

The invention relates to a system and an application which can precisely control the content of dissolved oxygen in water. The system is composed of a dissolved oxygen sensor, a transmitter, a data acquisition module, a programmable controller, a signal outlet module, two proportional electromagnetic valves, and the like, the dissolved oxygen sensor can detect the content of the dissolved oxygen in the water, detected signals are integrated to 4 to 20mA analog signals through the transmitter, the analog signals are acquired to the programmable controller through the data acquisition module, the programmable controller calculates 4 to 20mA control signals according to the measured dissolved oxygen value and the set value, and the control signals are further transmitted onto the proportional electromagnetic valves to regulate the opening degree of regulating valves to control the type and the flow rate of the introduced gas. The system and the application can solve the serious lag and the unbalanced forward and reverse direction control of the controlled system through the optimization of the output range of the set forward and the reverse direction control signals and the proportion of the on-off time of the control signals, precisely control the (concentration or) content of the dissolved oxygen and control the precision to be within 5ppb.

The invention discloses an indoor air quality control system based on personnel positioning. The control system comprises a light scattering type PM2.5 sensor, a CO2 sensor, a wind speed sensor, a wireless transmission module, a terminal, a microprocessor, a human-computer interaction panel and a mechanical ventilation air volume control module, wherein a building model database and a mechanical ventilation air volume calculation platform are installed on the terminal. The invention further discloses a ventilation calculation method. According to the method, based on personnel positioning, calculation is carried out on the functional zones with different pollution levels by using models, so that the optimal ventilation capacity of each functional area is achieved, unnecessary energy consumption and more pollution caused by blind opening of a whole-area mechanical ventilation system are avoided, good energy-saving and environment-friendly benefits are achieved, and when the outdoor pollution is large, more intelligent ventilation selection is provided for residential users or public building energy managers.

The invention proposes a dark link detection method and system. The method comprises: establishing a search engine spider user agent information library, a dark link feature library and a website basic information library; traversing each webpage in the website basic information library; comparing webpage information with the dark link feature library, and if features in the dark link feature library exist, then indicating that a dark link exists in the page, otherwise, simulating search engine spider user agent information and re-requesting for the page; and comparing whether a HASH value of the page is the same as that in the website basic information library or not, and if the HASH value of the page is the same as that in the website basic information library, then indicating that the dark link does not exist in the page, otherwise, indicating that the dark link exists in the page. According to the method, the dark link in a website can be detected and deleted without the need for artificial participation, so that the detection efficiency of the dark link is greatly improved.

An electronic experiment report automatic generating, intelligent correcting, storing and downloading system comprises a student operation subsystem, a teacher operation subsystem and a project principal operation subsystem; each subsystem comprises a plurality of modules, different systems and different modules interact with one another, and the functions of automatic generating, intelligent correcting, storing and downloading of the electronic experiment reports are achieved. The electronic experiment report can be edited at a PC terminal and can also be edited at a mobile terminal. Studentscan fill in the experiment report at any time at the mobile terminal and check the correction condition of the experiment report, thereby saving time. And the correction of the experiment report canbe completed at the mobile terminal without time limitation, thereby saving the time of teachers and improving the working efficiency. The problem of information communication hysteresis between teachers and students is solved, role setting is free and flexible, the teachers can serve as both experiment teachers and project managers, and the teachers or the project managers can check the experiment report filling submission condition and the experiment report reviewing condition at any time. Meanwhile, paperless and green experiment reports are achieved, and the requirement for sustainable development of the modern society is met.

The invention discloses a method and a system for predicting scrapped vehicles, which belong to the field of intelligent traffic big data. The method includes the following steps: extracting the characteristic information of vehicles from the track data of vehicles in a first preset time period; generating training samples according to the characteristic information of vehicles and a preset vehicle scrapping standard, and dividing the training samples into a training set and a testing set; training a prediction model according to the training set, and using the testing set to test the prediction model; and using the tested prediction model to predict scrapped vehicles in the future. In the invention, the prediction model is trained based on the historical track data of vehicles in a certain period, and the prediction model and the prediction result obtained finally are more scientific and reliable. Meanwhile, the operation rules of vehicles before scrapping are found out by combining the method with machine learning, and the problem of lagging in the prior art is solved. The vehicle manufacturers can also determine the marketing strategies more accurately.

The invention discloses a towboat fuel oil on-line monitoring management system, comprising a boat end monitoring system which is used for collecting and processing data and a shore end monitoring system which is used for analyzing and storing data. The boat end monitoring system comprises a rotating speed sensor, a mass flowmenter, a volume flowmeter, a collection cabinet and a boat end server; the boat end server comprises a dynamic information generation unit, a deck log voyage generation unit and a daily report generation unit; the shore end monitoring system comprises a network switch and a shore end server; the boat end server exchanges data with the shore end server through a 3G network; the shore end server comprises a fuel quantity consumption form generation unit, a staff appraisal report form generation unit and an inventory statistic unit. The towboat fuel oil on-line monitoring management system guarantees the data accuracy. All the data is backed up, which effectively obtains the scientific fuel consumption data of each fuel consumption unit of the towboat. The real-time data transmission solves the hysteresis of the data information transmission and greatly improves the working efficiency and decision efficiency.

The invention discloses a flue fly ash constant-speed sampling device as well as a flue fly ash constant-speed sampling system based on a BP neural network and a control method, and belongs to the field of detection of flue fly ash. The flue fly ash constant-speed sampling device comprises a sampling nozzle, a sampling pipe, a cyclone separator, a sampling bottle, an ejector pipe, an adjusting nozzle and an exhaustion pipe. The flue fly ash constant-speed sampling system based on the BP neural network comprises a sampling device, a PLC controller, an executor, a pressure sensor and a BP neural network module. The control method of the flue fly ash constant-speed sampling system based on the BP neural network is characterized in that the flue fly ash constant-speed sampling is realized based on the BP neural network, and flue static pressure approach to the actual pressure is predicted by using mass data which are measured in real time, so that the problem of hysteresis quality of the constant-speed sampling is solved, the sampling accuracy is improved, and the advantages of small calculation errors, high data processing ability and low design cost are achieved.

The invention discloses a greenhouse temperature control method. The greenhouse temperature control method comprises the following steps: acquiring an outdoor prediction temperature value at a future time point; predicting an indoor prediction temperature value according to the outdoor prediction temperature value; judging whether the indoor prediction temperature value is accordant with a temperature threshold suitable for the growth of plants; and if not, regulating an indoor temperature in a preset time period before the future time point according to the indoor prediction temperature value. The invention further discloses a greenhouse temperature control device. The greenhouse temperature control device comprises a prediction temperature value acquiring device, an indoor temperature sensor, temperature regulating equipment and a controller, wherein the controller comprises a pre-control module, the pre-control module is used for judging whether the indoor prediction temperature value is accordant with a temperature threshold suitable for the growth of plants at the future time point, and if no, regulating the indoor temperature of a greenhouse by virtue of the temperature regulating equipment in a preset time period before the future time point according to the indoor prediction temperature value. Therefore, the problem of temperature control hysteresis of the greenhouse can be avoided, the energy is saved, and the growth of the plants is benefited.

An image processing method and apparatus are provided which are applied to scheduling nodes in an image processing system. The image processing system also includes multiple image processing nodes. The method comprises: when an image processing task is received, selecting one image processing node from the plurality of image processing nodes as a target image processing node according to a presetimage processing scheduling strategy; assigning the image processing task to the target image processing nodes to so that the target image processing nodes perform the image processing task. The technical scheme of the application can improve the image processing efficiency.

An embodiment of the invention provides a method and a system for monitoring a fan and a wind turbine monitoring system, wherein the fan monitoring system comprises a fan monitor, a linkage module and a video monitor. The fan monitor is used for receiving an alarm signal sent by a fan controller installed on the fan and sending the alarm signal to the linage module, and the alarm signal comprises alarm time and a fan logo; the linkage module is used for analyzing the alarm time and the fan logo in the alarm signal and sending the analyzed alarm signal to the video monitor; and the video monitor is used for determining fan image information matched with the fan logo and the alarm time according to the fan logo and the alarm time and displaying the fan image information and can solve the problem of hysteresis of a wind turbine monitoring mode.

The embodiment of the invention provides an event information acquisition method and device. The scheme comprises the steps of obtaining a to-be-detected video; determining all targets in each frame of image in the to-be-detected video and types of all the targets; determining multiple pieces of triple information corresponding to each frame of image according to the types of all targets in each frame of image, the type of a preset action executor, the type of a driven target and the type of an interaction tool; intercepting at least one continuous image sequence corresponding to each piece oftriple information from the to-be-detected video; and respectively inputting each continuous image sequence corresponding to each piece of triple information into the behavior event detection model to obtain event information corresponding to each piece of triple information. By applying the technical scheme provided by the embodiment of the invention, the hysteresis problem existing in event information acquisition is solved, and the event information acquisition efficiency is improved.

The invention provides a flow prediction model training method and device, a flow prediction method and device, equipment and a medium. The flow prediction model training method comprises the steps: obtaining a flow data set, intercepting the flow data set through a sliding window operation, and obtaining the sample data features of the flow of the size of a sliding window and the sample labels corresponding to the sample data features of the flow; obtaining a next flow value after the sample data characteristics of the sample label flow; obtaining sample attribute characteristics corresponding to the sample label; processing the sample data characteristics of the flow by using a gated recurrent neural network of the model, and inputting the output memory state and the sample attribute characteristics into a full-connection network of the model for regression prediction; and according to the sample label and the prediction result, updating the parameter weight through a back propagation algorithm to obtain a trained flow prediction model. Therefore, a prediction result is associated with the attributes of different flows, and the problems of hysteresis quality and low accuracy of flow peak prediction are solved.

The invention discloses a textile workshop air conditioner system PLC automatic control method. The method comprises the steps that firstly, a PLC control system is initialized; various parameters ofa textile workshop are collected into the PLC control system; and finally, the PLC control system carries out analysis calculation according to the collected data, finally, an air conditioner system is controlled to run, and the problems that an existing textile air conditioner automatic control system is poor in control strategy control precision, and adjusting of an air conditioner automatic control system for the workshop temperature and humidity is delayed are solved.

The invention provides a Yellow River basin pollution source monitoring and early warning method based on electric power-environmental protection data fusion analysis, and the method comprises the steps: carrying out the data fitting of the correlation degree between power utilization data and environmental protection index data, and obtaining the correlation mapping of the power utilization data and environmental protection index data of different enterprises; constructing an enterprise portrait of each enterprise in the sample set from two dimensions of environmental protection index data and power utilization characteristic data, and realizing enterprise sample pollution discharge monitoring data complementation based on the enterprise portraits in combination with own actual sampling point data and enterprise actual power utilization load data; and on the basis of the short-term power load prediction value of each enterprise, in combination with the association mapping of the power consumption data and the environmental protection index data of the enterprise, obtaining short-term pollution discharge prediction data of each enterprise in the region, and taking the short-term pollution discharge prediction data as the input of the dynamic monitoring and early warning model of the drainage basin, thereby predicting the short-term environmental index in the region. And environmental protection early warning monitoring is realized.

The invention provides a pipe network siltation risk prediction modeling method based on a PNN neural network and an SWMM technology. The method comprises the following steps: A, collecting pipe network drainage parameters, and carrying out the preprocessing of the drainage parameters based on an SWMM model; b, constructing historical data including parameters influencing the pipe network sedimentation condition and reflecting the pipe network sedimentation condition based on the preprocessing result in the step A; and C, inputting historical data into the PNN neural network to obtain a siltation risk prediction model. The pipe network siltation risk prediction modeling method based on the PNN neural network and the SWMM technology has the advantages that the pipe network siltation risk prediction modeling method is based on the PNN neural network and the SWMM technology; through the combination of the PNN neural network and the SWMM technology, more accurate historical data are constructed for model training, the problems of singleness and hysteresis of an existing siltation judgment method are solved, a management and maintenance unit can conveniently make a detailed desilting maintenance scheme, and the pipe network maintenance efficiency is improved.

The invention provides a liquid-cooled heat radiator, a novel server heat dissipation system and a regulation and control method. The liquid-cooled heat radiator comprises a heat dissipation base anda plurality of heat dissipation fin groups arranged on the heat dissipation base, wherein each heat dissipation fin group comprises a hollow pipe and two heat dissipation fins, the heat dissipation fins are of hollow structures, ends, away from the heat dissipation base, of the first heat dissipation fins and the second heat dissipation fins are connected through the hollow pipes, the first heat dissipation fins, the second heat dissipation fins and the hollow pipe form a communicated cavity, two ends of each of a plurality of radiating fin groups are respectively communicated with a water inlet and a water outlet after a plurality of radiating fin groups are communicated in series or in parallel or in series-parallel combination, the heat dissipation base is fixed to a heat generation element through fixing holes. When a liquid-cooled radiator works, a communicated cavity formed by the first radiating fins, the second radiating fins and the hollow pipe is filled with cooling liquid, heat dissipation efficiency of an air cooling system is improved, and the effect of saving energy can be achieved.

The invention provides an anomaly prediction method, an anomaly prediction device, electronic equipment and a computer readable storage medium, and belongs to the technical field of intelligent medical treatment. The method comprises the steps of obtaining monitoring data of a target object in a current period; acquiring monitoring data of the target object in a historical period and a pre-trainedanomaly prediction model from a shared storage area; based on the monitoring data of the current period and the historical period, predicting an abnormal condition of the target object in a future period through the abnormality prediction model; and when the current period is ended, writing the monitoring data of the current period into the shared storage area. According to the invention, the possible abnormal condition of the patient in the future can be predicted, so that measures can be taken in advance and the medical risk is reduced.

The invention discloses a cement grinding mill system power consumption index prediction method based on an extreme learning machine, and the method comprises the steps: firstly collecting related variables, carrying out the operation analysis of a cement grinding mill system, selecting eight variables related to the power consumption of a cement grinding mill as input variables, selecting the ton power consumption as an output variable, and constructing an input and output layer of an ELM model; optimizing the number L of neurons of the ELM, the weight w of an input layer and the bias b of a hidden layer by using an IPSO algorithm; the IPSO algorithm taking a mean square error of the model as a fitness function in a parameter optimization process; ELM related parameters obtained through IPSO optimization being combined with sample data to complete training of an IPSO-ELM prediction model, and industrial field actual data being substituted into the trained model to complete online prediction of the power consumption of the cement grinding mill system. According to the method, the IPSO-ELM model is trained by using the sample data to obtain the power consumption prediction model, and variable data of an actual cement production site is input into the trained model, so that online prediction of the power consumption index of the cement grinding mill is realized.

The invention discloses a dynamic 3D city model construction method for city canyon environment navigation. The method comprises two steps of static refined 3D city model construction and dynamic optimization of a 3D city model based on position information generated by visual / GNSS / 3D model fusion. According to the method, the fusion result based on the airborne visual information / GNSS / 3D map in the current epoch is obtained in real time and is applied to the visual loopback detection step, so that the urban 3D model database is dynamically updated by utilizing the information after final map optimization. In addition, a result obtained in real time and fused in the current epoch based on airborne visual information / GNSS / 3D map is applied to the visual loopback detection step, so that the urban 3D model database is dynamically updated by using the information after final map optimization, the precision of the 3D urban model is improved, and the accuracy of the urban model is improved. And moreover, the problem of hysteresis of traditional 3D city model information is solved, and finally, important guarantee is provided for high-precision positioning of the city environment.

The invention discloses an exoskeleton rehabilitation robot control system. The control system includes a mobile platform system, a lower-limb exoskeleton system and a data terminal, wherein the mobile platform system includes a first single-chip microcomputer, a wheel hub motor and a weight reduction system; the lower-limb exoskeleton system includes a body driving system, a second single-chip microcomputer and a perception control system. By means of the control system, the problem of hysteresis of an exoskeleton is well solved, and the human-computer consistency is achieved; the applicationrange of the exoskeleton is enlarged; the problems of information exchange between patients and doctors and between patients and their families, and patients' training experience are solved.

The invention discloses a screen plate falling monitoring device of a medium drainage screen. The screen plate falling monitoring device of the medium drainage screen comprises a foreign object detection box, an undersize pipeline, a detection module and a control unit. The foreign object detection box is connected with the undersize pipeline, a grate is arranged at the bottom of the foreign object detection box, and a detection box cover plate is arranged on the top of the foreign object detection box. The detection module is arranged on the foreign object detection box and used for detectingthe position of the detection box cover plate. When the height of materials accumulated in the foreign object detection box is equal to a certain value, the detection box cover plate is prompted to move, and the control unit outputs an alarm signal. The hysteresis problem of human experience fault judgment is solved, and the timeliness and practicability advantages of the device are highlighted.

The invention discloses a method for predicting the failure of a virtual machine. The method includes: acquiring comprehensive performance data of a current virtual machine in real time, the comprehensive performance data comprising external data and internal data of the virtual machine, wherein the external data comprises performance data of a host where the virtual machine is located, a load ofthe host where the virtual machine is located and a load of the virtual machine, and the internal data comprises performance data of the virtual machine; determining a virtual machine category to which the virtual machine belongs through the external data, and calling a prediction model established in advance corresponding to the virtual machine category; inputting the internal data into the prediction model to predict the failure of the virtual machine. This scheme is to solve the problem of lag in the process of discovering virtual machine faults.

The invention discloses an apartment house management system and method. The apartment house management system comprises a cloud-end server, a management end and at least one lock control assembly, wherein the cloud-end server can be configured to transmit user identification information and house renting information to the lock control assembly according to instruction information generated by the management end; and the lock control assembly identifies and matches practical check-in person number with house renting information and judges whether user identification information is matched ornot; and a house door is controlled to open if matching fits; and the house door is controlled to be in a closed state if matching does not fit.