88results about How to "Improve the accuracy of judgment" patented technology

The invention relates to a parking lot management system and management method. The parking lot management system comprises a server, a number of UWB positioning parking cards and at least three UWB positioning base stations. The server comprises a storage unit which is used for storing a virtualized parking lot digital model which models all parking spaces in a parking lot to coordinate the parking spaces. The UWB positioning base stations are connected with the server through cables. The precise coordinates of the UWB positioning base stations are measured in advance and stored in the server as known parameters. The real-time position coordinates of the UWB positioning parking cards are calculated through a three-point positioning method. The server further comprises a comparing calculation unit which is used for comparing parking space coordinates in the parking lot digital model according to the real-time position coordinates of the UWB positioning parking cards to determine whether a car is parked in a parking space.

The invention discloses a pair of intelligent glasses with the blinking detection function and an implementation method of the intelligent glasses. The intelligent glasses comprise a glasses frame, a glasses bridge, eyeglasses, nose supports and glasses legs, wherein the nose supports and the glasses legs are respectively provided with a myoelectricity sensor; the glasses bridge is provided with a vibration reminder, and each glasses leg is provided with a myoelectricity signal processing device and a camera. The implementation method includes the following steps that 1, an eye change myoelectricity signal is collected through the myoelectricity sensors; 2, the collected myoelectricity signal is sent to a signal processing circuit for processing, and myoelectricity signal features are obtained; 3, the obtained myoelectricity signal features are compared with preset myoelectricity signal features, one-time blinking is judged if the obtained myoelectricity signal features are coincident with the preset myoelectricity signal features, and next judgment is carried if the obtained myoelectricity signal features are not coincident with the preset myoelectricity signal features; 4, the number of blinking times within a unit time is obtained, and the blinking frequency is obtained; 5, whether a user blinks abnormally or not is judged according to the blinking frequency, and the vibration reminder vibrates for reminding if it is judged that the user blinks abnormally. The intelligent glasses with the blinking detection function and the implementation method of the intelligent glasses can effectively protect the eyes in time.

An air conditioner refrigerant circuit performs a cooling operation in which an outdoor heat exchanger functions as a condenser of the refrigerant compressed in a compressor and an indoor heat exchanger functions as an evaporator of the refrigerant condensed in the outdoor heat exchanger. Further, an outdoor expansion valve is disposed at a position that is at once downstream of the outdoor heat exchanger and upstream of a liquid refrigerant communication pipe in the refrigerant flow direction in the refrigerant circuit in the cooling operation, and shuts off the refrigerant flow. A refrigerant detection unit is disposed upstream of the outdoor expansion valve and detects the amount or the amount-related value of refrigerant accumulated upstream of the outdoor expansion valve.

The invention provides a refrigeration anti-freezing protection control method and device for an air conditioner and the air conditioner. The control method comprises the following steps that after the air conditioner is started and is subjected to refrigeration operation, the temperature of an inner disk is acquired; whether the condition of entering an anti-freezing protection mode is met or notis judged according to the temperature of the inner disc; and if the condition of entering the anti-freezing protection mode is met, the air conditioner enters the anti-freezing protection mode, andthe operation state of an inner fan, a bypass control valve and a compressor is controlled according to the temperature of the inner disk, the exhaust temperature of the compressor of the air conditioner and a current wind gear of the inner fan of the air conditioner after the air conditioner enters the anti-freezing protection mode so as to prevent an indoor heat exchanger from freezing, whereinthe bypass control valve is arranged between the indoor heat exchanger and an outdoor heat exchanger. According to the refrigeration anti-freezing protection control method for the air conditioner, the evaporation temperature is improved through the adjustment of the air conditioner so as to achieve the defrosting purpose of an inner machine, frequent restart of the compressor is avoided, the pressure reduction and defrosting effects of the air conditioner are achieved.

The invention discloses a method for longitudinal zero-sequence power direction protection of a double circuit line on a same pole, which comprises an own-side zero-sequence power direction criterion and an opposite-side zero-sequence power direction criterion, and the own-side zero-sequence power direction criterion is shown by a formula (I) while the opposite-side zero-sequence power direction criterion is shown by a formula (II), wherein 3I0M and 3I0N are zero-sequence currents of an M side and an N side of one loop, 3I0M' and 3I0N' are the zero-sequence currents of an M side and an N side of the other loop, 3UM0 and 3UN0 are bus zero-sequence voltages of an M side and an N side at a protective mounting part, Arg shows a subsequent phasor phase-taking angle, and K is a zero-sequence mutual inductance influence coefficient. By using the AND gate outputs of the two criterions, when the two formulae are satisfied at the same time, ground faults can be determined to appear in a protected area, and the criterions of the longitudinal zero-sequence power direction protection are acted.

The invention provides a method and device for detecting junk posts based on an occurrence rule of posts in a community network. The method comprises the following step: a, recognizing the posts, and judging whether the posts are junk posts according to the content characteristics and the occurrence rule in one or a plurality of community networks. Preferably, the step a comprises the following steps: a1, recognizing the posts according to the preset semantic rule, and extracting the content characteristics of the posts; a2, inquiring the occurrence rule of the posts in the community network according to the content characteristics of the posts; and a3, judging whether the posts are junk posts based on a first preset rule according to the occurrence rule of the posts in the community network. In the prior art, a large number of repeated posts occurring in the community network can not be detected, which is caused by carrying out the dirty word matching or semantic analysis on a single post individually in general, and compared with the prior art, the method and the device can be used to improve the judgment accuracy of the junk posts.

The invention relates to a reactor dynamic interlock system and method based on a digital instrumentation and control system. The reactor dynamic interlock system comprises a fault diagnosis module, a decision support module and a dynamic interlock module. When a reactor apparatus generates an abnormal status, the fault diagnosis module outputs possible fault reasons, the decision support module lists corresponding emergency operation procedures, after an operator determines an emergency operation procedure according to a judgment, and the dynamic interlock module automatically executes the emergency operation procedure through a controller and downloads a monitoring configuration shielding dangerous operation for an operator station, such that the execution efficiency of the emergency operation procedure can be effectively improved, wrong operation of the operator can be prevented, and safe operation of a reactor is guaranteed.

The invention discloses a nuclear fuel subassembly video detection method. The nuclear fuel subassembly video detection method comprises following steps: (1) distance calibration in a photographed area is carried out, and nuclear fuel subassembly video images are collected; (2) the video images are processed, displayed, and stored, two calibration points of distance calibration in the video images are determined, and a measuring area in the video images is acquired; (3) the size of the measuring area is calculated, wherein actual calibration distance Lb between the two calibration points is acquired, image calibration distance Np between the two calibration points in the video images is calculated, image defect size Nc in the measuring area in the video images is calculated, and actual defect size Lc of the measuring area is calculated and stored based on the actual calibration distance Lb, the image calibration distance Np, and the image defect size Nc. The nuclear fuel subassembly video detection method can be used for observing situations of nuclear fuel subassemblies via video images, calculating defect size in the nuclear fuel assemblies, accurately determining that whether deformation of the nuclear fuel subassemblies is caused, and determining deformation degree accurately.

The embodiment of the invention provides a sleeping posture identification method and a sleeping posture identification device. The sleeping posture identification method comprises the following steps of acquiring the joint information of each joint of a nursed object in a measurement space; and judging whether the sleeping posture of the nursed object is abnormal or not according to the acquired joint information, and giving an alarm if the sleeping posture of the nursed object is abnormal. According to the sleeping posture identification method and the sleeping posture identification device, the joint information of the nursed object is used for judging whether the sleeping posture of the nursed object is abnormal or not in a sleeping posture identification process, so that the abnormal sleeping posture judgment accuracy is high, the false alarm probability is lower, and the time and the energy of a caregiver are saved; and moreover, the sleeping posture identification device is not required to contact with the nursed object, so that constraints to the nursed object are avoided.

The invention discloses a patrol inspection method for a power transmission line of a power grid. The patrol inspection method comprises the following steps: a camera takes video images of a power line or a pole in a flying patrol inspection line of an unmanned plane and transmits the video images to onboard communication equipment; the onboard communication equipment communicates with a plurality of ground control consoles to acquire position information of each ground control console, selects one ground control console closest to the unmanned plane at present according to different positions of different ground control consoles, and transmits the video images to the closest ground control console; the closest ground control console receives the video images, and judges damage to the power line and the pole of the power transmission line according to the video images; a background control center communicates with the multiple ground control consoles and is used for displaying the video images and a judgment result of each ground control console. According to the patrol inspection method, the accuracy of judging the damage to the power line and the pole of the power transmission line can be improved, measures can be conveniently taken in time, and safe operation of a power system is guaranteed.

The invention discloses a method for detecting loading conditions of an elevator car. The method comprises the following steps of: dividing a bottom plate of the elevator car into N regions (N is larger than 1); enabling a pressure sensor used for detecting the load of each region to be arranged at each region; receiving the load of each region of the bottom plate of the elevator car, which is detected by the corresponding pressure sensor; summing up the received loads of N regions to obtain a total load value; and judging whether the elevator car is in an abnormal loading status according to the obtained total load value and the received distribution condition of the loads of the N regions. Accordingly, the invention further discloses a device for detecting loading conditions of the elevator car. Through the adoption of the method disclosed by the embodiment of the invention, the loading conditions of the regions of the bottom plate of the elevator car can be detected by the pressure sensors distributed on the bottom plate of the elevator car, so that the accuracy for judging the loading conditions of the elevator car is improved, and the service efficiency of an elevator is further improved.

The invention discloses a filter method of rubbish mail, said method includes: collecting the mail sample set and rule set; the rule training set is generated based on said mail sample set and rule set; the decision tree model is built according to said rule training set, said decision tree model includes rules and its statistic value which are in various nodes; judging whether the mail is rubbish mail to be filtered based on said decision tree model.

The application discloses a motor controller over-current protection device, a motor controller and an electric car. The motor controller over-current protection device actually collects a two-phase current signal output by a motor controller through a current sensing module, converts the two-phase current signal into a voltage signal which can be measured by an operational amplifier through a current conversion module, and compares the voltage signal with an upper threshold voltage and a lower threshold voltage to judge whether there is an over-current phenomenon. As the voltage signal is directly collected by the current sensing module, the voltage signal represents the actual current value of the motor controller, the situation in which the current value represented by the voltage signal is inconsistent with the actual current of the motor controller is avoided, and the accuracy of judgment on whether there is an over-current phenomenon in the motor controller is increased. Further, the motor controller over-current protection device realizes bilateral detection of the voltage signal, and over-current detection by the motor controller over-current protection device is more comprehensive.

An embodiment of the invention relates to the wireless communication technology field and especially relates to a method for improving mobility robustness, a system and equipment thereof. The invention aims at improving determination accuracy of the mobility robustness in a networking environment of a Macro base station and a Pico base station under a Hetnet scene. In the embodiment of the invention, first network side equipment sends a finally obtained wireless signal quality parameter value of a cell to second network side equipment containing the cell whose parameter is not set up correctly, wherein the value is obtained before user equipment generates failure connection. The second network side equipment determines the decision signal quality value of the corresponding cell according to the wireless signal quality parameter value of the cell and then determines a reason of generation of the failure connection according to the decision signal quality value of the cell. The multilayer coating condition is considered in the Hetnet scene, determination accuracy of the mobility robustness in the networking environment of the Macro base station and the Pico can be improved.

The invention discloses a method and a device for power distribution device operation status analysis. The method comprises the following steps: collecting target data of a target power distribution device, wherein the target data including at least one of device status data, load rate and three-phase imbalance data of the target power distribution device; inputting the target data into a preset supervised learning machine model; obtaining a target fault rate of the target power distribution device output by the preset supervised learning machine model; and according to a preset fault rate threshold and the target fault rate, determining whether the target power distribution device is in a fault state.

The invention discloses a judging method of a bus-coupled or segmented running mode in bus protection, which is characterized by comprising the following steps: (1) at the time of no fault, changing a bus-coupled CT from a non-current state into a current state, and under the condition that the running mode of bus-coupling is judged to be in a separated state, respectively adding the current of the bus-coupling into two buses to calculate difference currents; comparing an obtained calculating result with a calculating result when the running mode of the bus-coupling is judged to be in the separated state, and using the variation trend of the difference current for judging; (2) at the time of a fault, changing the bus-coupled CT from the non-current state into the current state, and under the condition that the running mode of the bus-coupling is judged to be in the separated state, turning off a differential outlet trip for a short period of time; and carrying out reasonable logical judgment through the changes of an SHJ node and a TWJ node of the bus-coupling and the differential condition action. The invention has high judge accuracy and cannot make an erroneous judgment on the mistake current or abnormal conditions of a bus-coupled current collection channel. An effective valve and a sampling value are adopted to carry out a compound judgment so as to make a complementation on judge accuracy and judge speed.

Embodiments of the invention provide a method and a device for determining high-risk users. The method comprises the steps of firstly, based on an exception rule set, determining a high-risk businessrequest event, and a corresponding first user with a high risk; and then carrying out risk diffusion based on a crowd relation graph, and adding a second user with a specific association relation withthe first user in the graph, as the high-risk user, into a high-risk user set. Further, the exception rule set can be updated based on the high-risk user set. Therefore, the high-risk users can be identified more efficiently.

The invention relates to an adaptive adjustment method for a landing detection threshold value of a leveling supporting leg, and the method comprises the steps: powering on; starting an automatic leveling system; sampling and storing a motor current at a T2 time interval through the controller after the leveling supporting leg extends out for T1, wherein the number of collection times is N; calculating a mean value A of N sampling values of the motor current; taking the sum of the mean value A and a fixed threshold value B as a landing detection threshold value at this moment; repeatedly carrying out steps 2, 3 and 4; enabling a new landing detection threshold value to be compared with the former landing detection threshold value: if the new landing detection threshold value is less than the former landing detection threshold value, the new landing detection threshold value is taken as the current landing detection threshold value, and the next step is carried out; or else, directly switching to the next step; judging whether the supporting leg is landed or not: switching to automatic leveling if the supporting leg is landed, or else returning to step 2. The method is higher in judgment accuracy, avoids false landing of the leveling supporting leg in a process of automatic leveling, is higher in feasibility, is easy to achieve engineering, and is low in implementation cost.

The invention discloses a secret-associated network information leakage risk monitoring method. User operations are recorded in real time in each client, a complete behavior, that a file flows out ofthe client, of a user this time is recovered by taking the event that the file flows out of the client as a starting point, and then whether or not a single behavior has a leakage hazard is analyzed and judged in the client according to security level management provisions in a secret-associated network; and the behaviors of the various clients in the secret-associated network are uploaded to a behavior library of a server, all leakage hazard behaviors related to the behaviors of the clients are tracked and analyzed out, therefore, the secret-associated files are effectively prevented from flowing out of the secret-associated network, and leakage events are avoided. According to the method, the dilemma that current secret-associated network security protection means are powerless for the behaviors that malicious leakers leak secrets by using normal printing and burning ways is solved, and the defects of an existing monitoring method are overcome. The method is wider in monitoring coverage, higher in judgment accuracy rate and convenient for audit management of security protection personnel.

The invention relates to a single-phase fault judging method for a small-current grounding system of an overhead line. The single-phase fault judging method comprises the following steps of: respectively installing a fault indicator on a phase A, a phase B and a phase C of the same site of the overhead line, wherein each fault indicator is in wireless communication with the other two fault indicators by virtue of a wireless communication module; detecting the voltage and current information of the phase, and judging whether the phase has small-current grounding faults or not by virtue of each fault indicator; if the phase is judged to have small-current grounding faults, acquiring voltage information of the phase of the other two fault indicators by virtue of the wireless communication module; judging whether the voltage of other two phases is higher than the normal value or not, if higher, judging that the phase has true small-current grounding faults and sending the small-current grounding fault information of the phase to a communication terminal by virtue of the communication module, otherwise, not sending the small-current grounding fault information.

The invention belongs to safe driving detection technology, and relates to wearable fatigue driving monitoring system and method based on forehead electroencephalogram signals. The system is capable of monitoring fatigue state of a driver in real time, high in judgment accuracy, timely in early warning and intervention, comfortable to wear and convenient to carry. The system comprises a forehead electroencephalogram signal acquisition module, a fatigue state identification module and a result feedback and intervention module which are sequentially connected and all arranged on a wearable headband, wherein the forehead electroencephalogram signal acquisition module is used to acquire the forehead electroencephalogram signals of the driver and transmit the acquired electroencephalogram signals to the fatigue state identification module in real time; the fatigue state recognition module is used to analyze the received electroencephalogram signals, carry out data preprocessing and featureextraction, perform fatigue state recognition by adopting an algorithm model obtained by off-line training and send recognition results to the result feedback and intervention module; and the result feedback and intervention module is used to remind and intervene the driver when receiving the fatigue state transmitted by the identification module.

The invention discloses a processing method of chat information, and a server. The method comprises the following steps: receiving a chat information publish message sent by each client, computing the score of the chat information for the currently received chat information publish message, and filtering the chat information publish message when the score exceeds a first preset threshold value; otherwise, computing the score summation of the corresponding chat information in the chat information database by use of a user identifier, and filtering the chat information publish message when the score summation exceeds the second preset threshold value; otherwise pushing the chat information publish message to each corresponding client according to publish range information in the chat information publish message. According to the scheme disclosed by the invention, each chat information is scored, and each user is accumulatively scored, a score standard is set to balancing whether the intention of the chat information is rightful and whether the chat behavior of the user is reasonable, the judgment granularity for unreasonable chat behavior is refined, thereby precisely and effectively filtering the unreasonable chat behavior.

The invention relates to a method for detecting integrity of a filter element of an ultrafiltration membrane. The granule counting method is adopted, aerosol with particle size ranging from 0.1 to 1 micrometer and particle concentration ranging from 30000 to 200000 each liter serves as an air source, the aerosol is conveyed to the filter element of the ultrafiltration membrane and filtrated through the filter element through an air sampling pump and a pipeline, the gas particle concentration before and after entering the filter element of the ultrafiltration membrane is detected through a particle counting machine, the removal rate of particulate matters in the air source before being filtered through the filter element and the air after being filtered is calculated, and the qualified limiting value of the removal rate is high than 99.9%. The method for detecting the integrity of the filter element of the ultrafiltration membrane is fast and high in efficiency, and the identification of the integrity of the filter element of the ultrafiltration membrane is enhanced.

The invention provides a method for detecting an adhesive tape at the edge of banknotes, which comprises the steps of acquiring banknote thickness detection data of thickness sensors in n channels when the banknotes are detected, and finding out a start position and an end position in the banknote thickness detection data; respectively calculating the banknote thickness detection data, and calculating a mean value of an oscillation area near the start position and the end position of each piece of banknote thickness detection data; sorting the mean values, and using a mean value of the mean values locating at the middle positions of the sorting sequence to act as a reference mean value; and judging whether each piece of banknote thickness detection data is greater than the reference mean value and exceeds a set value or not, if so, judging there is an adhesive tape at the edge of the banknote in the channel and alarming, if not, ignoring. Compared with the prior art, the method detects an adhesive tape at the edge of the banknotes, is not only high in judgment accuracy but also simple in detection and improves the detection efficiency.

The invention relates to an old people nursing system and method. The old people nursing system comprises a sound positioning module, an image collection module, a processing module and a nursing device, wherein the sound positioning module performs sound collection on an image collection direction to obtain sound information, performs positioning on a sound production point of a user according tothe sound information to generate positioning information and imports the sound information into a sound tumble model for judgment to generate sound judgment information; the image collection moduleperforms image collection on the user in real time to obtain image information, and imports the image information into an image tumble model for judgment to generate image judgment information; the processing module generates tumble information when both the image judgment information and the image judgment information show that the user tumbles; and the nursing device moves according to the tumble information and the positioning information to a positioning area so as to help the user. Compared with the prior art, the old people nursing system can perform image collection and sound positioning, realize accurate judgment on whether the old people tumble or not and improve the judgment accuracy, and can also carry out temporary rescue through the nursing device.

The invention relates to a detecting method of active voice end, which comprises the following steps including 1) receiving continuous voice and obtaining frame from the continuous voice, 2) calculating the energies of the frame obtained in step 1) and obtaining energy threshold value according to the energies, 3) respectively calculating the zero-crossing rate of the frame obtained in step 1) and obtaining the zero-crossing rate threshold value according to the zero-crossing rates, 4) using linear regression deductive method to judge whether each frame is active voice or inactive voice by taking the energies obtained in step 2) and the zero-crossing rates obtained in step 3) as input parameter of the linear regression deductive method and 5) obtaining active voice starting point and active voice end point in the active voices or inactive voices of the step 4) according to the energy threshold value and the zero-crossing rate threshold value. The invention increases the judging accuracy rate of the active voice starting point and active voice end point, and also improves the correctness rate of voice identification.

The invention discloses a fault monitoring system for an ultralong optical cable and a method and belongs to the technical field of ultralong optical cable detection. The fault monitoring system for the ultralong optical cable disclosed by the invention comprises a first brillouin time domain detection system, a second brillouin time domain detection system, an uplink path optical fiber, a downlink path optical fiber and a plurality of optical repeaters arranged in all the sections of the uplink path optical fiber and the downlink path optical fiber, wherein the first brillouin time domain detection system and the second brillouin time domain detection system are used for emitting detection pulse light to the uplink path optical fiber and the downlink path optical fiber, collecting a backscattered brillouin light signal scattered by the detection pulse in the optical fiber channel and analyzing the monitoring data at each point on the optical fiber channel. According to the fault monitoring system for the ultralong optical cable and the method disclosed by the invention, the fault information at the abnormal point of the optical cable is analyzed by utilizing the original temperature and strain data acquired by the collected back scattered brillouin light signal, so that the high-precision fault positioning and diagnosis can be realized.

The invention provides a transaction risk control method and device. The transaction risk control method in the invention comprises the steps of: receiving a to-be-judged order of an account, obtaining historical transaction data corresponding to the account, then, after the fact that pre-set transaction attributes need to be statistical according to the historical transaction data is judged, extracting transaction attribute data corresponding to the pre-set transaction attributes in the to-be-judged order and the historical transaction data, generating to-be-judged model characteristics according to the transaction attribute data, inputting the to-be-judged model characteristics into a pre-set risk control model, so that a risk result is obtained, and finally, performing risk control on the to-be-judged order according to the risk result. According to the transaction risk control method and device provided by the invention, because the pre-set transaction attributes are transaction attributes that the difference value among corresponding transaction attribute data in a normal order and a fraud order is greater than a pre-set numerical value in a process of training the pre-set risk control model, the transaction attribute data input into the model can comprehensively reflect the to-be-judged order; and thus, the judgement accuracy of the risk control model is increased.

The invention discloses a multifunctional smart cat-eye system. A light intensity analysis module is used for comparing a change curve of environmental light intensity data with a preset threshold curve, and a camera is activated when the change curve exceeds the preset threshold curve. A wide-angle optical anti-shake automatic rotation camera is adopted, and an external unit processor is used for controlling the power-on state of an electromagnetic coil (6) in order to rotate a lens. An electromagnetic coil power-on driving module is used for generating a corresponding electromagnetic coil power-on driving signal according to shooting target direction information, so that the aim of automatically rotating the lens according to a shooting target direction is fulfilled. Through adoption of the multifunctional smart cat-eye system, active focusing and capturing can be performed, and a physical rotation angle being 3 degrees is provided. A wide-angle convex lens is adopted, and an actual effective acquisition range is up to 170 degrees, so that a target image can be acquired effectively and completely. In conjunction with judgments on the changing situation of the light intensity and a target distance, the camera is activated when the light intensity drops suddenly and the target distance is relatively small, so that error judgments caused by turn-off of passage lights (the light intensity drops suddenly) and the like are avoided; the judgment accuracy is increased; and the judgment reliability is enhanced.

A method for automatically counting the ear rate of wheat scab features that the image of wheat scab is easily affected by background, and the influence of illumination and shade is difficult to be recognized, compared with the prior art. The invention comprises the following steps: acquiring and preprocessing the wheat scab image; construction of Wheat Spike Detection Model Based on Wheat Scab Image; training depth convolution neural network; collection and preprocessing of Wheat Fusarium Head Blight Images; counting of normal ear and diseased ear; calculating the rate of diseased panicle. The invention realizes the automatic statistics of wheat scab ear rate through the machine vision method, and improves the intelligence level, robustness and detection precision of wheat scab ear rate statistics.