35 results about "Observation period" patented technology

The application discloses a kind of based on GNSS's dam vertical displacement monitoring reference network automation observation method.It includes the following steps, step one: point site reconnaissance, instrument and observation pier selection;Step two: GNSS data preprocessing and scheme optimization;In GNSS vertical displacement automation monitoring, determine the optimal observation period, repair frequent small cycle jump;Step three: baseline solution and network adjustment method;In GNSS vertical displacement automation monitoring, troposphere delay correction, thermal expansion effect correction these models are applied to baseline solution, introduce prior height difference information in network adjustment, improve the accuracy and reliability of baseline solution.The application has the advantages of improving the vertical direction of GNSS baseline solution precision.The application also discloses a kind of based on GNSS's dam vertical displacement monitoring reference network automation observation platform.

The invention relates to the technical field of track measurement, in particular to a high-precision measurement system for building a track based on a total station, which comprises an acquisition module, an overrun determination module, an analysis module, an identification module, a calculation module, an adjustment module and an execution module. According to the method, through multi-dimensional parameter joint analysis, initial overrun is identified by using an allowable deviation threshold value, and then sections with related geometric features are merged into measurement and control units through an analysis model, so that the aggregation degree of anomaly positioning is improved; then, the abrupt change feature is combined to distinguish the abrupt change feature into linear anomaly and local deformation, and further through a threshold feedback mechanism which can be dynamically adjusted along with an observation period, the change of anomaly detection from static threshold judgment to adaptive evaluation of evolution along with a line state is realized, so that a measurement result more accurately reflects the real deformation feature of a track; the problem that the tiny deviation of the track is difficult to timely and accurately identify due to accumulated errors of the sensor and data response lag is effectively solved.

The invention belongs to the technical field of image processing, and particularly relates to an air permeability evaluation method and system for a seedling substrate, and the method comprises the steps: collecting a photo sequence and a pressure difference sequence of a seedling substrate sample under a specific humidity condition at a fixed frequency in an observation period; constructing a ventilation difference index; generating an adjustment critical value; generating an evaluation grade and an adjustment instruction based on the comparison result; generating a formula adjustment coefficient; obtaining an environment correction index; generating a ratio adjusting instruction; and packaging the ventilation difference index, the proportion adjustment instruction and the structure risk factor into a new training sample, and carrying out iterative learning on the training model. The method effectively solves the technical problem that the sensitivity of the system for identifying the abnormal event of the seedling culture substrate structure is low because macroscopic soil pores and effective air permeability of the seedling culture substrate structure are confused in the existing method.

The invention discloses a permanent resident population identification method and device, equipment, a storage medium and a product. The method comprises the following steps: on the basis of position signaling data of user terminals in a set observation time period in a set area in a statistical period and a predetermined night residence of a user to which each user terminal belongs, determining a single-day staying accumulated duration of each user in a corresponding night residence every day in the statistical period; generating the number of permanent residents under an operator in a set area based on the accumulated duration of the single-day stay of each user and a set permanent resident population identification rule; wherein the night residence place of the user to which each user terminal belongs is generated based on a network topological graph formed by resident base station points of each user terminal in a set area in a set observation time period, and the set permanent resident population identification rule is generated based on inflection point detection of permanent resident population scale change trends under different rules. The method has the advantages of high speed processing, low identification cost, high result precision and the like.

Activity level and fall detection using accelerometer data is described. In one or more implementations, measurements of a user generated by a wearable monitoring device during an observation period are obtained, the measurements including accelerometer data. Physical steps taken by the user are detected based on peaks in the accelerometer data above a threshold, and a step count is generated based on the detected physical steps within predetermined time epochs. Activity level classifications of the user for the predetermined time epochs are generated based on the step count. The step count and the activity level classification for the predetermined time epochs may then be output, such as a notification or in a user interface. Fall predictions may be generated and output by processing the accelerometer data using machine learning models trained to correlate patterns in accelerometer data to fall events.

This invention discloses a network fault alarm analysis method, apparatus, device, and storage medium. The method includes: when several alarm events are generated within several observation periods, extracting key fields of the alarm events and sorting them according to time series to obtain two-dimensional matrix data; extracting alarm data from the two-dimensional matrix data based on a sliding window determined by a sliding window adjustment strategy, and extracting spatiotemporal features from the extracted alarm transaction data to train a first fault type identification model; determining a first fault type through the updated first fault type identification model; using a sequence pattern mining algorithm to obtain several frequent sequences for each observation period and identifying a second fault type; when the first fault type matches any second fault type, determining the faulty device and its corresponding physical location based on the corresponding frequent sequence. This invention can cope with dynamic and ever-changing network attack behaviors and achieve accurate location of fault physical locations.

The present application belongs to the technical field of building settlement observation engineering, and provides a method and system for observing uneven settlement on both sides of a building high-altitude corridor, comprising the following steps: extracting historical observation data of the high-altitude corridor from an observation history database, performing correlation analysis on all observation points of the high-altitude corridor, and constructing an observation point correlation system; performing trend analysis on the historical observation data of the observation points, extracting observation points with an observation period to be adjusted according to the trend analysis results, as observation points to be adjusted, comparing and analyzing the observation points to be adjusted with the observation point correlation system, extracting observation points to be adjusted with correlated observation points, as linked observation points, and observation points to be adjusted without correlated observation points, as non-linked observation points. The method optimizes the problem of missing sensitive point measurement caused by fixed observation period and data mismatch caused by single adjustment of correlated observation points, thereby causing settlement amount calculation distortion.

The application provides a traffic situation deduction method and system based on multi-source data fusion, relates to the technical field of traffic information processing, and obtains source observation records formed by at least two traffic perception sources for a target import lane in a target observation period, wherein the source observation records include a queue tail position and an observation coverage range; according to the positional relationship between the observation coverage range and the queue tail position, available source observation records are selected from the source observation records; according to the queue tail position in the available source observation records, a fused queue tail position of the target import lane in the target observation period is obtained; and the fused queue tail position is compared with a queue storage boundary of the target import lane to obtain a queue overflow prediction result of the target import lane. Thus, the influence of the source observation records with insufficient observation coverage ranges on the fusion result of the queue tail position can be reduced, and the reliability of the queue overflow prediction can be improved.

The application discloses a quality control method and device of marine observation data and a medium, relates to the technical field of marine observation data processing, and comprises the following steps: acquiring marine observation data, wherein the marine observation data comprises sampling data of each observation period in multiple observation periods; sequentially performing missing value inspection, global range inspection and stratified statistical inspection on the sampling data of the observation period to obtain first inspection data; performing average processing on the first inspection data to obtain an observation value; sequentially performing empirical range inspection and quartile range inspection on the observation value to obtain a second inspection value and a second inspection mark; filling the second inspection value based on the second inspection mark, combining a local weighted regression scatter smoothing algorithm and an STL decomposition and reconstruction algorithm, and updating the second inspection value and the second inspection mark to complete quality control. The application can realize reasonable filling of missing values, accurate detection of abnormal values and data quality traceability.

The invention belongs to the technical field of global navigation satellite system integrity monitoring, and discloses a Beidou CNAV telegraph text space signal integrity evaluation method and system. The method comprises the following steps: firstly, respectively screening obtained Beidou CNAV messages and post precise ephemeris to obtain available ephemeris; taking a precision satellite orbit and a precision satellite clock after the event as a reference, resolving a CNAV satellite clock error and a CNAV satellite orbit error in an observation period, and further calculating to obtain a worst user distance measurement error of a Beidou CNAV telegraph text space signal; and estimating the fault rate and the average fault duration of the Beidou CNAV telegraph text space signal by using the worst user ranging error large sample data set of the CNAV telegraph text space signal and adopting a confidence test iteration method. The method can provide key technical support for integrity evaluation of the Beidou system.

The invention relates to a total station self-learning measurement adjustment method and system based on rolling prediction. Comprising a gateway control module, a data acquisition and processing module and a platform management module. The gateway control module is respectively connected with the data acquisition and processing module and the platform management module and is responsible for monitoring equipment control and data transfer storage; the data acquisition and processing module is responsible for acquiring monitoring data and taking coordinates predicted based on a Transform time sequence prediction model as initial coordinates of indirect adjustment so as to reduce the number of adjustment iterations to obtain monitoring coordinates; and the platform management module realizes equipment remote management and data visualization. According to the method, prediction coordinates closer to a real value are used as adjustment initial approximate coordinates, and adjustment iteration convergence can be accelerated. Monitoring point coordinates obtained through adjustment are used for updating a rolling prediction model of the next observation period and also used for updating a learning file; a self-learning closed loop which is accurate in model prediction, fast in total station aiming and stable in adjustment result is formed, and continuous optimization of system performance is achieved.

The embodiment of the invention provides a method for evaluating the performance of an electrocardio analysis algorithm and an application product, and the method comprises the steps: obtaining a reference detection result, the reference detection result being a standard analysis result of a first electrocardio record, and the first electrocardio record being obtained by recording an electrocardio signal in an observation time period; obtaining a to-be-evaluated detection result; if the first electrocardio record records a persistent event, performing multi-stage comparison on the to-be-evaluated detection result and the reference detection result to obtain a first comparison result; and / or, if the first electrocardio record records an instantaneous event, comparing the to-be-evaluated detection result with the reference detection result based on a dynamic time window matching strategy to obtain a second comparison result; and providing an evaluation result obtained according to the first comparison result and / or the second comparison result. According to the embodiment of the invention, high-precision, multi-dimensional and automatic evaluation can be carried out on the performance of the electrocardio analysis algorithm.

The application relates to the technical field of network flow security detection methods, and discloses an AI driving-based PCDN flow anomaly detection method, which comprises the following steps: acquiring original NetFlow data and on-off logs, and forming account association flow data; performing aggregation statistics on the account association flow data according to a time window to form flow statistical data of an account to be detected; combining baseline flow data of a similar user group to perform deviation degree representation on the flow statistical data to form behavior deviation feature data; inputting the behavior deviation feature data into a pre-trained PCDN detection model to obtain daily detection results corresponding to each detection day of the account to be detected; continuously determining the daily detection results in a continuous observation period, and marking the account to be detected as a PCDN account when the determination result meets preset conditions. Through baseline correction of the similar user group and continuous observation period determination, the accuracy and stability of PCDN account detection are improved.

The application provides a kind of based on data traceability's subsidence monitoring method and system, first obtain the time series subsidence observation data of each monitoring point in subsidence monitoring area in continuous observation period, including subsidence and corresponding observation time environmental disturbance information, then build subsidence data traceability association network, associate time series subsidence observation data with monitoring point geological basement information, observation equipment traceability identifier and data processing node information, then trace the generation link of each monitoring point subsidence data, extract equipment calibration record etc., generate traceability link feature, then coupling analysis traceability link feature and subsidence, determine traceability reliability index, finally according to index screening data, generate dynamic subsidence trend model and subsidence monitoring analysis report containing traceability link feature in combination with monitoring point spatial distribution characteristics, improve the accuracy and reliability of subsidence monitoring.

The utility model provides a novel settlement observation point device for a booster station, which relates to the technical field of settlement observation, and is characterized in that a settlement observation piece is arranged on an embedded box body, the embedded box body comprises an inner side box body and an outer side box body, the outer side box body is communicated with the inner side box body, the settlement observation piece comprises a first rod body, a second rod body and a measuring head, the first rod body is connected with the inner side box body in an inserted mode, the second rod body is rotationally connected with the first rod body, and the second rod body can be folded in the inner area of the outer side box body. The second rod body is connected with the first rod body, and the second rod body and the first rod body are sectionally arranged, so that the second rod body can be folded at a position which does not exceed the surface of a wall body, and the second rod body can be rotated to be parallel to the first rod body only when settlement observation is needed; in this way, the situation that the protruding part of the wall body is prone to being collided in the non-observation period can be avoided, and therefore observation data are not prone to being affected; meanwhile, the risk that pedestrians are stumbled by the protruding positions is reduced, and personal safety is protected.

The invention relates to the technical field of foundation pit observation, in particular to a foundation pit support crack deformation observation system and method. According to the technical scheme, the device comprises a first gasket arranged at the top of a left ground nail, the top of the first gasket is rotationally connected with a reading ruler, the top of a right ground nail is provided with a second gasket, the top of the second gasket is rotationally connected with a positioning ruler, and the reading ruler and the positioning ruler are staggered up and down. The height of the first gasket is changed by rotating the first lifting bolt, the height of the reading ruler is changed by rotating the second lifting bolt, so that the reading ruler and the positioning ruler are close to each other, the reading error is reduced, after the foundation pit crack is enlarged, the reading ruler and the positioning ruler move relatively, the exposed scale is the width of the foundation pit crack, a mobile phone is used for shooting, and the detection accuracy is improved. The device can amplify observation readings on a computer, store reading pictures, regularly shoot and observe crack changes in a period, is simple in structure, can be used by non-professional measurement personnel, and is high in usability.

The invention provides a hospital consumable inventory prediction management method based on big data analysis, and relates to the technical field of hospital consumable inventory prediction.The method comprises the steps that according to consumption and use characteristics of consumables, repeated consumables are converted into basic consumption units, all the consumables are stored according to a fixed field sequence, and a consumable resource pool is formed; according to the type and the number of consumables used by each patient in the diagnosis information, constructing a mapping relationship based on the types and the number of consumables in the patients and the consumable resource pool, and counting the types and the number of consumables having the mapping relationship with the patients in the historical diagnosis information; and calculating an optimization target loss rate of each consumable type along with the time length based on the time length, setting an observation period of the current moment, counting the use quantity of each consumable type in the observation period, calculating the use quantity according to the diagnosis information, calculating an early warning value according to the target loss rate, and giving an alarm according to the early warning value.

This invention relates to a method, system, and storage medium for detecting slow failures in storage devices in cloud storage systems. The method includes: S1, real-time acquisition of read / write latency and throughput time series of the storage device within an observation period; S2, outlier filtering of the read / write latency and throughput time series; S3, construction of a mapping relationship between throughput and latency, and establishment of a regression model; S4, calculation of the slowness time series based on the regression model; S5, traversal of the slowness event series using a sliding window to form slow events; S6, quantification of slow events into risk scores according to scoring rules, and accumulation based on the time series; S7, determination of whether the accumulated risk score is greater than a pre-configured risk threshold. If so, the corresponding storage device is marked as a slow failure device and isolated offline; otherwise, return to step S1. Compared with existing technologies, this invention has advantages such as accurate location of slow failure devices and strong versatility.

The invention relates to a disturbance observation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps of determining a current actual input torque of a current disturbance observation period based on a previous estimated disturbance of a previous disturbance observation period under the condition that the current disturbance observation period is not the first disturbance observation period; based on the current actual input torque, outputting the current displacement error amount of the current disturbance observation period through an admittance control model; based on the current displacement error amount, determining a current estimated torque through an inverse model of an admittance control model; a current estimated perturbation for the current perturbation observation period is determined based on the current actual input torque and the current estimated torque. Therefore, the accuracy of disturbance observation is improved.

Activity level and fall detection using accelerometer data is described. In one or more implementations, measurements of a user generated by a wearable monitoring device during an observation period are obtained, the measurements including accelerometer data. Physical steps taken by the user are detected based on peaks in the accelerometer data above a threshold, and a step count is generated based on the detected physical steps within predetermined time epochs. Activity level classifications of the user for the predetermined time epochs are generated based on the step count. The step count and the activity level classification for the predetermined time epochs may then be output, such as a notification or in a user interface. Fall predictions may be generated and output by processing the accelerometer data using machine learning models trained to correlate patterns in accelerometer data to fall events.

The invention discloses a user inactive state triggering-based uncertain carbon Popinion integral excitation method, which comprises the following steps of: firstly, acquiring low-carbon behavior data of users in a carbon Popinion platform, constructing an activeness index, evaluating the low-carbon behavior condition of each user, and calculating the activeness index of each user according to the low-carbon behavior condition of each user; when it is judged that a certain user is in an inactive state, at least two incentive schemes are actively sent to the user, in one scheme, the user churn behavior is intervened in advance through the uncertainty of rewards and an activeness index, and the probability that the user participates again is remarkably improved; and continuously performing low-carbon behavior monitoring on the user in a preset observation period after user excitation, and dynamically adjusting reward parameters in the uncertainty reward mode through the constructed feedback adjustment model according to the low-carbon behavior change of the user in the observation period. Therefore, the activeness and long-term effectiveness of the users in the carbon general benefit platform are kept, fine configuration of the excitation resources is finally realized, and technical support is provided for low-carbon behavior excitation of large-scale user groups.

To provide a monitoring device, monitoring method, and program that can recognize objects existing on a background spread across a three-dimensional space. [Solution] The monitoring device according to this disclosure comprises: a division unit that divides a three-dimensional space representing point cloud data into at least one divided space; an observation unit that observes the feature quantities of a first point cloud data included in the divided space during a reference period and observes the feature quantities of a second point cloud data included in the divided space during an observation period; and a determination unit that determines a first divided space to be in an abnormal state if the time during which the difference between the feature quantities of the second point cloud data and the feature quantities of the first point cloud data exceeds a first threshold is longer than the time during which the difference exceeds a first threshold, wherein the first divided space is included in at least one divided space.

The application discloses a kind of based on charging pile intelligent operation and management system and method, belong to the technical field of fault early warning, by collecting the multi-source state data generated in the operation process of charging pile, construct unified time sequence operation data matrix;Extract key features to construct feature vectors, and use multi-classification neural network model to evaluate health status;Combined with the health trend in the continuous observation period, a micro-degradation evolution path model is constructed to generate a fault prediction curve;The prediction result is matched with the similarity of the fault prior curve library, the risk weight coefficient is calculated, the potential fault node is identified and the early warning list is output;Based on high-risk pile site, a regional task scheduling map is constructed, the geographic location, power level and residual life information are fused, and the operation and maintenance path and resource allocation scheme are optimized;The application can significantly improve the fault prediction accuracy and operation and maintenance efficiency of charging pile, realize intelligent operation and maintenance and response optimization.

The present application relates to the field of track measurement technology, and more particularly to a high-precision measurement system for track construction based on a total station, which comprises an acquisition module, an overrun determination module, an analysis module, an identification module, a calculation module, an adjustment module and an execution module. The present application uses multi-dimensional parameter joint analysis to identify preliminary overruns using an allowable deviation threshold, and then merges sections with relevant geometric features into measurement and control units through an analysis model to improve the aggregation of abnormal positioning. Subsequently, combined with the mutation characteristics, it distinguishes between linear anomalies and local deformation, and further through a threshold feedback mechanism that can be dynamically adjusted with the observation period, it realizes the adaptive evaluation transition of abnormal detection from static threshold judgment to the evolution of the line state, making the measurement results more accurately reflect the real deformation characteristics of the track, and effectively solving the problem of the difficulty in timely and accurate identification of small deviations in the track caused by sensor cumulative error and data response lag.

The application discloses an online warehouse allocation method and device considering order sequence and warehouse capacity constraints, wherein the method comprises the following steps: acquiring warehouse capacity occupied by each order on average; acquiring a current order which needs to be responded currently, and calculating a modified profit of each warehouse responding to the order based on the current order, and determining a warehouse with the maximum modified profit; and modifying a shadow price of the warehouse capacity according to a warehouse allocation result obtained by the warehouse with the maximum modified profit until all orders are modified in a preset observation period, and determining a final warehouse allocation result. Thus, the technical problem that the global operation capacity of the store, the global inventory of the warehouse and the randomness of future orders are ignored in the related art, the calculation result cannot meet the global demand, and the algorithm solving efficiency is low is solved.

The application provides a mega-constellation transit tracking method and device, a storage medium and electronic equipment, comprising: generating a transit display map of a mega-constellation; based on ground station information and basic information of the mega-constellation, marking the satellite track of the observable mega-constellation in the observation period in the transit display map; based on priority evaluation rules, generating a satellite observation list with priority; according to the priority in the satellite observation list, allocating resources for each mega-constellation in the satellite observation list to generate a satellite tracking task list corresponding to the observation period. By applying the method provided by the application, the track of the mega-constellation can be observed intuitively by generating a transit display map, and the user's decision can be determined in a man-machine interactive manner, and reasonable resource planning is performed for each mega-constellation in the observation period in combination with the user's decision and the priority evaluation rules in the system, so that optimal tracking is realized under limited resources.

This invention relates to the technical field of network traffic security detection methods, and discloses an AI-driven PCDN traffic anomaly detection method. The method includes acquiring raw NetFlow data and online / offline logs to form account-related traffic data; aggregating and statistically analyzing the account-related traffic data by time window to form traffic statistics for the account to be detected; combining baseline traffic data from similar user groups to characterize the deviation of the traffic statistics, forming behavioral deviation feature data; inputting the behavioral deviation feature data into a pre-trained PCDN detection model to obtain daily detection results for the account to be detected on each detection day; continuously judging the daily detection results within a continuous observation period, and marking the account to be detected as a PCDN account when the judgment result meets preset conditions. Through baseline correction for similar user groups and continuous observation period judgment, the accuracy and stability of PCDN account detection are improved.

The present invention discloses a GNSS-R sea surface height inversion quality control method, device, and terminal device. The quality control method includes: step 1, selecting a plurality of target height data from a preset sea surface effective reflection height dataset; step 2, calculating the average height data and its standard deviation for each observation period as a fitting dataset; step 3, fitting the fitting formula to obtain model parameters and the standard deviation of the fitting residual to construct a prediction model; step 4, predicting the predicted height data for the current observation period; step 5, detecting whether the sea surface effective reflection height data for the current observation period is abnormal; step 6, if not, updating the fitting dataset and re-executing steps 3 to 5 based on the next observation period until quality control of each observation period is completed; if so, deleting the sea surface effective reflection height data and re-executing steps 4 to 5 based on the next observation period until quality control of each observation period is completed. The present invention can effectively improve the accuracy of sea surface height inversion results.

The present application belongs to the technical field of image processing, and particularly relates to a method and system for evaluating the air permeability of a seedling substrate. The method comprises: collecting a photo sequence and a pressure difference sequence of a seedling substrate sample under specific humidity conditions at a fixed frequency within an observation period; constructing an air permeability difference index; generating an adjustment threshold value; generating an evaluation grade and an adjustment instruction based on the comparison result; generating a formula adjustment coefficient; obtaining an environmental correction index; generating a proportioning adjustment instruction; and packing the air permeability difference index, the proportioning adjustment instruction, and a structural risk factor into a new training sample to train a model for iterative learning. The present application effectively solves the technical problem of the prior art that the macroscopic soil pore and effective air permeability performance of the seedling substrate structure are confused, resulting in low sensitivity of the system in identifying abnormal events of the seedling substrate structure.

The invention discloses a foreign exchange pending transaction matching method and device, electronic equipment and a storage medium, and relates to the field of financial science and technology or other related technical fields, and the method comprises the steps: obtaining an original pending price and a user pending amount of a target foreign exchange transaction based on a pending transaction request; obtaining a user time point difference based on the user pending amount, and calculating a user time point price based on the original pending price and the user time point difference; acquiring a foreign exchange price sequence of a financial institution based on a set observation period to obtain an initial observation interval; performing secondary cutting on the initial observation interval based on the time point difference of the user to obtain an observation interval; and matching the time point price of the user with the observation interval, obtaining a target observation interval under the condition that the time point price of the user is successfully matched with the observation interval, and carrying out foreign exchange pending order transaction matching based on the target observation interval. According to the method and the device, the technical problem of relatively low efficiency of a foreign exchange transaction matchmaking mode of comparing data one by one in related technologies is solved.