142results about How to "Avoid missed judgment" patented technology

The invention provides an inter-class variance MODIS forest fire point detection method, which comprises the following steps: pretreating original data required from an MODIS satellite receiving station, acquiring brightness temperature or reflectivity of each channel, and excluding illegal data; identifying image elements of cloud and water; identifying a smog mask; extracting potential fire points from excluded image elements of the cloud, the water and the illegal data; calculating inter-class variance yield of each potential fire point and effective background image elements in a background window; verifying fire points according to the threshold value of the inter-class variance and the smog mask; and outputting an MODIS image containing the fire points. By adopting lower threshold value, the potential fire points comprise true fire points as well as comprise certain false fire points of which the underlying surfaces have higher temperature so as to avoid omission; the threshold value of the inter-class variance is adopted to judge and remove the false fire points so as to prevent from being misjudged; moreover, information of the underlying surfaces around the fire points is considered fully, so that the method can adapt to different areas and seasons.

The embodiment of the invention discloses a collision detection method, device, equipment and storage medium for mechanical arms. The method includes the steps that during the repetitive operation of a mechanical arm, if it is detected that the mechanical arm finishes one working cycle, the dynamic threshold used for judging whether or not the mechanical arm is collided is updated; the actual measured value of output torque of a power device used for controlling the mechanical arm to work is obtained, and a theoretical value corresponding to the output torque of the power device is obtained by a dynamical model, wherein at least one parameter relative to the operation of the mechanical arm inside the dynamic model changes with time; the difference value between the actual measured value and the theoretical value is determined, and whether or not the mechanical arm is collided is judged according to the difference value and the updated dynamic threshold. The collision detection method, device, equipment and storage medium for the mechanical arms can prevent the collision detection precision from decreasing with the changes of time, so that the phenomenon of judgment omission and judgment errors can be avoided.

A real-time monitoring system of faults of overhead lines of a distribution network comprises a fault alarm of overhead lines arranged at the overhead lines of the distribution network, a wireless repeater which is arranged at an overhead line rod of the distribution network, a background monitoring system and a wireless transceiver which is connected between the wireless repeater and the background monitoring system. The fault alarm of the overhead lines transfers the detected real-time abnormal signals on the overhead lines and address information of the abnormal signals to the wireless repeater, the wireless repeater gradually transfers the information to the wireless transceiver through a relay link which is constituted by the wireless repeater, and the wireless transceiver further transfers the information to the background monitoring system for analysis, judgment and processing. The fault alarm of the overhead lines comprises a sensor, a signal acquisition unit, a microprocessor, a wireless RF module, a magnetoelectric conversion circuit, a power control module and a power supply. The real-time monitoring system can accurately and rapidly give out the faults, the abnormal information and occurrence addresses thereof on the overhead lines.

The invention discloses a ship electric power station fault diagnosing method based on a knowledge petri network. The method includes the steps that (1) fault symptom sets of units of a ship power station are obtained and screened according to a ship power station fault Petri network model; (2) by means of an improved Apriori algorithm, strong association rule mining is carried out on the fault symptom sets and the fault units; (3) by means of man-machine conversation, a user inputs fault symptom characteristic quantity and confidence, a system carries out fault symptom identification through fuzzy reasoning by means of a strong association rule to determine the fault units; (4) the fault units serve as a root database, faulty Petri subnets are extracted from the Petri network model, fault reason diagnosis is carried out by means of a forward operation and backward inference method, and according to diagnosis results, fault reasons, fault route graphs and a corresponding fault maintenance method are provided. The ship electric power station fault diagnosing method can avoid false negatives of the fault reasons, generates a fault propagation path, and improves accuracy and efficiency of ship power station fault diagnosis.

The invention pertains to video coding technique, in particular relates to a transformation and quantization method. The technical problem to be solved by the invention is providing a transformation and quantization method which reduces the video coding complexity. The invention can detect the situation in which a residual block system is zero after transformation and quantization in advanced through prediction before the transformation and quantization, thus saving the transformation and quantization process of an all-zero block. The proposal has an outstanding meaning for video coding with a low bit rate, as the proportion of the all-zero block is lager in the situation of low bit rate. Detecting most parts of the all-zero block in advance can greatly reduce the calculating complexity of transformation and quantization, further increase the coding efficiency. Furthermore, the invention can real-timely adjust the judging conditions for the all-zero block, further enhance the coding speed and quality and avoid occurrence of missed judgment.

This invention relates to continuous casting dross detection method and device based on vibration detection. On section if the end of motion arm far away the long nozzle, assemble multiple vibration sensors with different directions to collect vibration signals of different directions; form sampled signal after amplifying, filtering, sampling and A / D transform; take off-line process for the ,mass collected signals to form norm eigen vector; use effective vector quantization algorithm to train codebook corresponding to all states used as codebook KB of all casting states when recognizing on line; by wavelet analysis method, obtain eigen vector of sampled signals; through the search and calculation for code word, decide which codebook that current state belongs to and give dross information. The advantage of this invention is that it needs just to set multiple vibration sensors on motion arm and treat the signal real time, and needs no other auxiliary device.

The invention relates to a safety scanning method and device of a downloaded file. The safety scanning method comprises the steps that when it is monitored that the file is downloaded, the content of the downloaded file is obtained and scanned, and file feature information corresponding to the content of the file is obtained; a preset local feature information library is inquired, whether the file feature information corresponding to the file content is matched with information items stored in the local feature information library is judged according to preset rules; the safety grade of the downloaded file is determined according to the matching result. The safety scanning method can detect malicious information like fishing information contained in the file, thereby avoiding missed judgment of malicious files, and improving accuracy of judgment.

The embodiment of the present application discloses a risk control model training method and a device applying the method. By dividing the historical data of known results into the first and second data sets, the first data set is used to train each risk model with different loss functions. Each risk control model is verified through the second data set, so as to select the optimal model according to the verification result. Subsequently, the optimal model is used to carry out risk control processing on the data to be processed. Realize rapid judgment of data risks such as online transactions, avoid missed judgments of data risks, and ensure the safety of users' property.

The invention brings forward a wireless sensor network-based intelligent parking guidance system and a parking guidance method thereof. The intelligent parking guidance system comprises a parking stall data collection module, a parking stall data management module and a parking stall data issuing module. The parking stall data collection module utilizes a geomagnetic sensor to collect parking stall information and transmits the formation to the parking stall data management module in real time by a wireless ad-hoc network mode; the parking stall data management module carries out processing and statistics on the parking stall information; and then, the parking stall data issuing module issues the parking stall information timely to a parking stall display. According to the invention, the intelligent parking guidance system guides a user to find a nearest empty parking stall, so that a utilization rate of parking resources is improved; information transmission is realized by employing a wireless ad-hoc network mode; therefore, it is convenient for arrangement and maintenance of the system; complexity and cost of the system are reduced; and a favorable condition is created for wide application of the intelligent parking guidance system.

The invention provides a processing method of towrope exploration navigational positioning data. The method includes the steps of reading original positioning data from a P2 / 94 format data file collected by an onboard positioning system, wherein the original positioning data comprise differential global positioning system (DGPS) data, RGPS data, acoustic data, cable compass data and onboard gyrocompass data, carrying out preprocessing which comprises gross error detecting, filtering and data standardization processing on the read original positioning data, using a node calculating method to confirm coordinates of positioning nodes in the preprocessed original positioning data, converting accurate positions of the positioning nodes into navigation data of a P1 / 90 format, and storing the navigation data. The method can process the original navigation data of the P2 / 94 format into the navigation data of the P1 / 90 format fast and efficiently.

The invention relates to a data cleaning and repairing method, which comprises the following steps of: for data errors generated in acquisition and data transmission generated by power distribution network terminal equipment, setting an access module to carry out data cleaning and repairing; For data errors generated by information equipment in the power distribution network information system, setting a conversion module for data cleaning and repairing; For data interaction among a plurality of business information systems of the power distribution network, data badness generated in data processing, setting a rule base cleaning module; For cleaning of historical data stored in a warehouse, setting an offline module for data cleaning and repairing, and carrying out logic checking on the whole historical data. According to the method, data errors and bad data conditions can be efficiently judged, data missed judgment and wrong judgment are prevented, the repair efficiency is ensured, and the upper-layer business application data requirement is met.

The invention relates to the technical field of intelligent building comprehensive security and protection management and control, in particular to an intelligent building management and control system for realizing behavior recognition based on an intelligent video analysis algorithm. The invention provides an intelligent building management and control system for realizing behavior recognition based on an intelligent video analysis algorithm, and the system comprises: a processor which executes a computer implementation assembly stored in a memory; a video analysis assembly which comprises an early warning model and is configured to identify emergencies in the monitoring video and generate early warning information; an emergency service assembly which is configured to call a corresponding emergency plan according to the early warning information; and a data acquisition monitoring assembly which is configured to be used for managing data acquisition, monitoring, data storage, early warning, linkage and system configuration functions of the system and comprises a linkage assembly, wherein the data acquisition monitoring assembly comprises a linkage assembly which is used for carrying out linkage processing on external access equipment of the system according to the emergency plan.

The invention belongs to the field of short text auditing, particularly relates to a short text auditing method and device fusing variant word recognition, and aims to solve the problem of how to fusea variant word recognition technology into a harmful text auditing task and realize automatic updating of a model. The method comprises the steps of constructing a configuration word bank, obtainingto-be-audited text data based on a social media platform, screening the to-be-audited text data to obtain suspicious text data, removing meaningless information, and calculating a text feature vectorand a statistical feature vector; and performing feature fusion on the text feature vectors and the statistical feature vectors, obtaining harmful texts through a trained harmful text classification model based on a support vector machine, obtaining sensitive words of the harmful texts by utilizing a preset keyword extraction algorithm, and writing the sensitive words into a configuration word bank. According to the method, the variant word recognition technology is fused into text feature and statistical feature calculation to carry out harmful text auditing tasks, automatic model updating isachieved, and the text auditing accuracy and updating speed are improved.

The invention discloses a fixed-point fatigue identification method for diagnosing faults of a bearing and a gear of a transmission system, comprising a pair of main gears (1) having a tooth number DC, an engagement pair of small gears (2) having a tooth number XC, a bearing (3) for supporting the main gears, and a bearing (4) for supporting the small gears. Parameters of the bearing (3) supporting the main gears are set as follows: intermediate diameter DDO, roller diameter Dd and roller number DZ; and the parameters of the bearing (4) supporting the small gears are set as follows: intermediate diameter XDO, roller diameter Xd and roller number XZ. The invention provides the method for identifying the faults caused by mutual fixed-point fatigue and quasi-fixed-point fatigue; and the method is characterized in that the parameters of the bearings are selected reasonably, and the mutual fixed-point fatigue and the quasi-fixed-point fatigue of the bearings and the corresponding matched gears are avoided, so that the fault rates of the bearings and the gears can reduced further.

The invention provides a personalized information display method and device based on face recognition, and a user terminal. The method comprises the following steps that: carrying out face recognitionon a key frame in the user image data of a target user, and determining that the key frame contains user face information; according to the user face information in the key frame, judging whether a target user corresponding to the user face information is a known user or not; and if the target user corresponding to the user face information is the known user, displaying personalized information corresponding to the determined known user. By use of the method, on the basis of a face recognition technology, through the steps of user existence confirmation, known user confirmation and relevant information recommendation, personalized information related to the user is recommended for different users, accuracy for recommending and displaying the relevant information is greatly improved, erroneous judgment and missed judgment for user categories can be avoided, and user experience is improved.

The invention relates to a DC fault arc detection and protection device, and belongs to the field of electrical engineering. The device comprises a high-frequency current transformer, a Hall current sensor, an amplifier, a bandpass filter, a high-precision ADC, a digital signal processor (DSP) and an IGBT circuit breaker, an output terminal of the high-frequency current transformer is connected with an input terminal of the amplifier, an output terminal of the amplifier is connected with an input terminal of the bandpass filter, an output terminal of the bandpass filter is connected with an input terminal of an AD sampling circuit, an output terminal of the AD sampling circuit is connected with an input terminal of the digital signal processor, and an output terminal of the digital signalprocessor (DSP) is connected with the IGBT circuit breaker. A current signal obtains a high-frequency current component through coupling via the high-frequency current transformer, the signal is amplified by 10 times through the amplifier and then passes through the bandpass filter, the ADC samples the signal and transmits the signal to the DSP for algorithm processing, the Hall current sensor transmits a detected DC current value to the DSP for dynamic threshold calculation, and an arc is extinguished through the circuit breaker after detection of arc generation so that protection is realized.

The invention relates to an intervention method and device of driving behavior, computer equipment and a storage medium. The intervention method comprises the steps that vehicle driving information and driving behavior information are obtained; a first weight corresponding to the vehicle driving information and a second weight corresponding to the driving behavior information are determined; the first weight and the second weight are added to obtain the driving behavior comprehensive weight; the driving behavior safety level is determined according to the corresponding relation of the preset driving behavior comprehensive weight and the driving behavior safety level; and an intervention instruction corresponding to the driving behavior safety level is obtained, and intervention is conducted on the driving behavior according to the intervention instruction. According to the intervention method, based on a data analysis technology and risk management and control, the purposes that the driving behavior safety level of a driver is determined comprehensively according to the obtained vehicle driving information and the obtained driving behavior information, and the corresponding intervention instruction is executed are realized, thus the accuracy of intervening in the driving behavior of the driver is improved, and the driver is advantageously reminded about safe driving.

The invention provides an image detection method and device, electronic equipment and a storage medium. According to the specific implementation scheme, carrying out feature extraction on a to-be-detected image to obtain a to-be-detected feature image; processing the to-be-detected feature image by using the convolution layer to obtain the probability that the pixel points in the to-be-detected image belong to the text prediction box, the prediction positions of the text prediction box to which the pixel points belong, and the prediction positions of the question stem boxes matched with the pixel points in the answer box, wherein the text prediction box comprises a question stem box and an answer box; and according to the probability that the pixel point belongs to the text prediction box,the prediction position of the text prediction box to which the pixel point belongs, and the prediction position of the question stem box matched with the pixel point in the answer box, obtaining a matching relationship between the question stem box and the answer box of the to-be-detected image. According to the embodiment of the invention, the matching relationship between the question stem boxand the answer box can be detected from the to-be-detected image, so that the subsequent question judgment process is facilitated, and the missed judgment problem caused by the limitation of the fixed area of the answer box can be avoided.

The present invention provides a vanadium redox flow battery control system based on PLC. The control system comprises a vanadium redox flow battery process system and a control cabinet. The control cabinet comprises an alternating current power supply configuration unit, a direct current power supply configuration unit, an ion concentration measurement unit, a man-machine interface unit, a control unit, a signal transmission module group and a power frequency conversion unit. According to the present invention, a power supply circuit of the control system, a PLC module group, a I / O module group, the signal transmission module group and the power frequency conversion unit are grouped and independently designed so as to realize the hot redundancy of the whole system, and greatly improve the stability and the reliability of the system. The system has a plurality of operating modes and an external four-circuit alternating current power supply, such that the charging and the discharging of the vanadium redox flow battery can be performed at any time and any place, the operation is convenient and the application is flexible; the system further has functions of self-diagnostic and self-recovery, such that the misjudgment and the missed judgment of the failure are effectively avoided, the reliability and the availability of the system are improved so as to indeed realize unattended operation; in addition, with the man-machine interface, the related parameters can be monitored and adjusted so as to meet the requirements of various operating conditions.

The invention discloses a method for identifying an irregular spam short message on the basis of Chinese word segmentation. Chinese word segmentation is performed on the same short message in a way of normal transverse reading firstly according to contents of the short message, and then weight is calculated according to number of words of a word segmentation result. Then according to the characteristic that the irregular spam short message is necessarily constrained in number of characters in each row, the range of the contents of the short message is judged, vertical arrangement of the characters in the range of the contents of the irregularly arranged short message is converted into transverse arrangement, then Chinese word segmentation is performed, and weight is calculated according to the number of words of the total word segmentation result. Then according to comparison of two times of weight, whether the short message is the normally arranged short message or the irregularly arranged short message is judged. Then according to the type of arrangement, content analysis matching keywords are adopted to identify whether the short message is the spam short message so that leakage judgment of the spam short message is avoided, all-checking rate and accurate-checking rate of the spam short message are enhanced.

The invention discloses an evaluation system for transient stability of a large-scale power system based on optimal mutual information redundancy. The system firstly obtains a power system large dataset that can approximately represent all operating characteristics of a current power system based on PMU data or software simulation; on the basis, an optimal characteristic set with maximum correlation and minimum redundancy with transient stability margin of the system is constructed through multi-time utilization of an information redundancy optimization algorithm; a function expression of a relationship between a selected variable and TSM is constructed by utilizing MATLAB; and the TSM can be directly estimated by an expression through real-time variable data obtained by the power system,after appropriate compensation is performed on a result, the result is fed back to on-site scheduling personnel in order to make timely decisions. The system can solve the shortcomings that traditional evaluation methods are poor in precision, are slow in speed, are poor in robustness of data loss, are poor in evaluation analyticity, and the like, and effectively meets real-time evaluation requirements of large-scale power systems.

The invention discloses a detection method. By adding an image shooting step after an element carrier sticking working procedure and combining preprocessing information of a computer on a CAD drawing,an image of an element carrier after the sticking working procedure is completed is analyzed, more accurate and practical structure feature information, compared with the CAD drawing, of a tested product is obtained, mounting defects of components such as a metal wire and a metal ball are judged by carrying out comparison on structure features, acquired through analysis on images of element carriers in the same batch after a wire bonding working procedure is finished, by utilizing the structure feature information, defect judgment accuracy is improved, and an error and omission during judgment are avoided.

The invention discloses a small current grounding line selection device and line selection method, thereby improving the line selection precision. The device comprises an AC input conversion module for isolating and transforming an input analog signal and transmitting the converted analog signal to an A / D conversion module, the A / D conversion module for collecting an analog signal and transmittingthe signal to a CPU module, a switch input module for collecting a switch-in amount and transmitting the information to the CPU module, the CPU module for selecting the collected analog amount as line selection sample data, carrying out fault line determination, and outputting a switch-out amount to a switch-out output module, and the switch-out output module for receiving the switch-out amount outputted by the CPU module and controlling turning on / off of an external device.

The invention relates to a field elimination method, especially to a field elimination method for eliminating a continuous multi-day meter reading data deficiency fault of an electric energy meter. Atpresent, some faults need to be dealt with on the scene; the field processing efficiency is low; and wrong determination, determination missing, and repeated determination occur easily. A first fieldfault elimination method with a local communication mode being a non-RS485 mode and a second field fault elimination method with a local communication mode being an RS485 mode are provided. The firstfield fault elimination method includes: step 11, a preparation step is performed; and step 12, an elimination step is performed; to be specific, information check, general checking, and parameter checking are carried out; local communication is investigated; reading of electric energy meter data is carried out; manual fault determination is carried out; and then feeding back to the system is performed. The second field fault elimination method includes: step 21, a preparation step is performed; step 22, an elimination step is performed; to be specific, information check is performed; generalchecking is performed; terminal parameters are investigated; local communication is investigated; reading of electric energy meter data is carried out; manual fault determination is carried out; andthen feeding back to the system is performed. Therefore, the processing speed is increased; the wrong determination, determination missing, and repeated determination are avoided; and a problem of lack of multi-day meter reading data under the concentrator can be solved quickly and accurately.

The invention relates to a weed occlusion diagnosis and troubleshooting determination method of a distributed photovoltaic power station. The method includes: determining whether a weed occlusion diagnosis needs to be carried out on a current date; calculating the daily dispersion rate; comparing the daily dispersion rate with the threshold dispersion rate; calculating the average dispersion rateof this week and comparing the average dispersion rate of this week with the average dispersion rate of the last week; calculating the ratios of the electric power of the power station to the benchmark electric power in the first 15 days; grouping the ratios of the first 15 days according to the adjacent 3 days, and calculating the average values of 5 groups of ratios; comparing the average valuesof two adjacent groups; pushing a weed occlusion failure alarm or a suspected weed occlusion failure prompt; and continuing the determination of weed occlusion troubleshooting from the day after theweed occlusion failure alarm is pushed. The invention realizes the remote accurate positioning of the weed occlusion failure, and automatic recovery can be achieved after weeding on the spot and troubleshooting. Manual spot check and patrol inspection can be avoided, the operation and maintenance cost of the distributed photovoltaic power station is reduced, the operation is simple, and the economic input cost is low.

The invention discloses a magnetic flux leakage defection machine of an oil sleeve. The magnetic flux leakage defection machine comprises a control cabinet, a computer, a production line detection system and a marking machine, wherein the control cabinet is respectively connected with the computer, the production line detection system and the marking machine through a signal line; and the production line detection system comprises a plurality of clamping driving devices, a vertical defect detection device and a transverse defect detection device. Electric signals of sleeve defects, detected by an infrared monitoring sensor on the production line detection system, are processed by a signal amplification circuit and a noise cancelling circuit and are divided into two paths; in one path, defect parts are accurately sprayed with marks by a signal processing circuit through utilizing the computer to drive the marking machine; and in the other path, defect signals are mixed into a simulated signal, and defect degree is displayed on a display screen of the computer through an A / D (Analog / Digital) converting circuit. With the adoption of the technical scheme of the invention, the magnetic flux leakage defection machine disclosed by the invention can carry out defect detection on the oil sleeve in rotation process on a production line, has small power consumption amount, is energy-saving and environment-friendly and has high detection sensitivity; and a probe does not need to be replaced when sleeves with different calibers are detected, the detection efficiency is high, the hooping of the oil sleeve can be smoothly avoided, the probe can be automatically reset after being detected, the service life is long, the cost is reduced, and the detection quality and the detection effect are improved.

The invention provides a self-test method for a temperature and humidity sensor. The self-test method includes the following steps that: a heating element and a heating unit connected with the heatingelement are respectively arranged in a probe and a transmitter; the heating element is controlled to perform heating, and the working state of the probe is judged according to the rise and fall change of collected temperature and humidity data; in a test environment, the heating element is controlled to perform heating according to self-test power, so that a preset temperature and humidity curvecovering the working temperature and humidity interval of the sensor is obtained; during a self-test process, the heating element is controlled to perform heating according to self-test power, so thata self-test temperature and humidity curve is generated; and whether the sensor drifts is judged according to the conformity of the self-test temperature and humidity curve and the preset temperatureand humidity curve which is obtained under the test environment having the same before-heating temperature and humidity environment with the self-test temperature and humidity curve. With the self-test method provided by the invention adopted, the working state and performance of the sensor can be remotely judged, the self-check problem of the temperature and humidity sensor can be solved, and good technical effects can be realized.

The invention discloses a method for on-site eliminating a nil data fault of an electric meter under an I-type concentrator, and relates to on-site defect elimination methods. At present, some faultsneed to be handled on site. However, the efficiency of on-site handling often is low, and conditions of wrong determination, omission of determination, and repetitive determination easily occur. According to the invention, the method includes the following steps: 1. a step of preparation; and 2. a step of defect elimination: information checking, general inspection, parameter investigation, localcommunication investigation, electric meter data reading, manual fault determination, and feedback to system. According to the technical solution processes the fault based on the set steps, and can effectively increase processing speed, the method can prevent wrong determination, omission of determination, and repetitive determination. The method herein can also fast and accurately eliminate the nil data problem of the electric meter that exists for days under the I-type concentrator in accordance with standard on-site fault defect elimination procedures.

The invention relates to the technical field of optical camera dirt detection, in particular to a camera module black spot detection system based on a turntable, which comprises a camera module for shooting an image; a acquisition module used for acquiring an image shot by the camera module; a storage module used for storing the images shot by the camera module; and a detection module used for carrying out black spot detection on the image. The technical problem that misjudgment or missed judgment is likely to occur due to the fact that in the prior art, the junction of a dark area and a bright area is taken as a stain to be captured is effectively solved. The black spots of the camera module can be automatically detected, and the black spot detection efficiency and accuracy of the cameramodule are improved.

The invention discloses a method for detecting the incomplete root penetration of a butt weld of a thick-walled pipe in ferrite, and belongs to the field of ultrasonic flaw detection nondestructive testing. The method for detecting the incomplete root penetration comprises the steps that firstly, a transverse wave high-frequency angle probe is utilized for carrying out qualitative analysis on incomplete penetration reflected waves; then excess weld metal of the weld at the horizontal flaw locating position is ground and polished to be in the same horizontal plane with a base metal; finally, a high-frequency longitudinal wave normal probe is used for carrying out qualitative and quantitative analysis determination on the incomplete root penetration. According to the method, the incomplete root penetration of the butt weld of the thick-walled pipe in the ferrite is precisely judged, and meanwhile the method can be used for the incomplete root penetration of a thick plate, a pressure container and the like in the ferrite in a reference mode.