185 results about "Baseline model" patented technology

A business methodology for optimizing energy procurement energy demand (usage) and energy supply for a facility or complex. After ascertaining a baseline model, energy consumption is monitored and adjusted to reflect dynamic economic factors of operations. With the accumulation of data, contract negotiations with energy suppliers may provide further energy savings. Over time and patterns of usage determined further savings and optimization in energy, equipment, staffing and other operational areas are obtained. Remote access provides efficient monitoring of established systems.

The invention discloses a method and a system for detecting a network security threat based on trusted business flow. The method comprises the following steps of establishing a blacklist and a white list of network flow and constructing a baseline model, wherein the white list is the trusted business flow and is a feature contour library of normal network behavior and host behavior; comparing real-time monitoring flow data and the baseline model; when the real-time data is matched with the blacklist, outputting an abnormal flow alarm; when the real-time data is matched with the white list and a deviation exceeds a preset threshold, outputting a threat flow alarm; and when the real-time data is mismatched with the blacklist and the white list, treating as a gray list and outputting an unknown flow alarm. According to the method and the system, the network security threat can be comprehensively and effectively detected at a low false alarm rate and high anti-virus efficiency, and the method and the system can adapt to a more granular network attack and defense confrontation environment.

A monitoring system including a baseline model that automatically captures and models normal system behavior, a correlation model that employs multivariate autoregression analysis to detect abnormal system behavior, and an alarm service that weights and scores a variety of alerts to determine an alarm status and implement appropriate response actions. The baseline model decomposes the input variables into a number of components representing relatively predictable behaviors so that the erratic component e(t) may be isolated for further processing. These components include a global trend component, a cyclical component, and a seasonal component. Modeling and continually updating these components separately permits a more accurate identification of the erratic component of the input variable, which typically reflects abnormal patterns when they occur.

There is provided a method of automatically identifying and characterizing spectral peaks of a spectrum generated by an analytical apparatus and reporting information relating to the spectral peaks to a user, comprising the steps of: receiving the spectrum generated by the analytical apparatus; automatically subtracting a baseline from the spectrum so as to generate a baseline-corrected spectrum; automatically detecting and characterizing the spectral peaks in the baseline-corrected spectrum; and reporting at least one item of information relating to each detected and characterized spectral peak to a user. In embodiments, baseline model curve parameters or peak model curve parameters are neither input by nor exposed to the user prior to the reporting step.

Systems and methods are provided for detecting an anomaly in a computer that is part of a population of networked computers. Snapshots are received from a plurality of computers within the population of computers, where individual snapshots include a state of assets and runtime processes of a respective computer. An asset normalization model is generated from the snapshots and serves as a baseline model for detecting an anomaly in the state of assets and runtime processes of a respective computer. A snapshot from at least one of the computers is compared to the asset normalization model in order to determine whether an anomaly is present in a state of static assets and runtime processes of the at least one of the computers.

A method of classifying and organizing digital images utilizing optical metadata (captured using multiple sensors on the camera) may define semantically coherent image classes or annotations. The method defines optical parameters based on the physics of vision and operation of a camera to cluster related images for future search and retrieval. An image database constructed using photos taken by at least thirty different users over a six year period on four different continents was tested using algorithms to construct a hierarchal clustering model to cluster related images. Additionally, a survey about the most frequent image classes shot by common people forms a baseline model for automatic annotation of images for search and retrieval by query keyword.

System, method and computer-readable medium for baseline modeling a product or process. A service database contains process data. A preprocessor processes the data into a predetermined format. A baseline modeling component builds a baseline model from the preprocessed data, wherein the baseline model relates process performance variables as a function of process operating conditions.

Customization systems and methods are described in relation to performance devices and apparatus using a closed-loop feedback process for exploration, design and improvement of devices including performance devices. Crowdsourced knowledge bases, iterative improvements and centralized statistical, semantical and other technical analyses are employed to modify a baseline design. A baseline model defining modifications to a base device necessary to obtain a master device including modifications calculated to improve performance of the master device over the base device. Manufacturing instructions including CNC code can be produced and used to causes the production machine to manufacture the customized device. A design-to-order system is described as having of a simplified user interface, a queuing system, CAD / CAM capability and a repository of stored baseline models.

The present disclosure is directed to systems and methods for validating and / or identifying wind farm performance measurements so as to optimize wind farm performance. The method includes measuring operating data from one or more wind turbines of the farm. Another step includes generating a plurality of baseline models of performance of the wind farm from at least a portion of the operating data. Thus, each of the baseline models of performance is developed from a different portion of operating data so as to provide comparable models. The method also includes selecting an optimal baseline model and comparing the optimal baseline model with actual performance of the wind farm. In a particular embodiment, the actual performance of the wind farm is determined after one or more wind turbines of the wind farm is modified by one or more upgrades.

The invention discloses a model updating method based on strain modal shape correlation. The specific steps are as follows: step 1), establishing a finite element model of a structure and analyzing the finite element model; step 2), performing experimental design and analysis; step 3), extracting a finite element simulation strain mode; step 4), performing correlation analysis: adopting a model confidence factor, and analyzing the correlation between the finite element model and the strain modal shape of an experimental test; step 5), selecting a mode to be modified; step 6), selecting a parameter to be identified; step 7), constructing a modification target; and step 8), performing modified iteration. According to the model updating method based on the strain modal shape correlation provided by the invention, by selecting an appropriate unit type, the obtained finite element model of the structure provides a reference model for strain response calculation; by selecting an appropriate strain mode to be modified, parameters to be modified and an optimum design method, the modified finite element model can better reflect the strain response of the structure; and the accurate finite element model is beneficial to the subsequent structural dynamic optimization design based on the finite element model, and the development of structural health monitoring and structural response prediction and so on.

The invention discloses an automatic baseline correction method, which includes the following steps: converting raw data into spectrogram data; utilizing continuous wavelet transform to calculate the numerical derivative of a spectrogram; respectively utilizing the sliding window method and the iterative threshold method to identify narrow peak signals and broad peak signals in the spectrogram; carrying out contour fitting on the broad peak signals to identify the edges of the broad peak signals; calculating a baseline model; correcting the spectrogram data according to the baseline model; searching distorted signals in the corrected spectrogram data, and recalculating the baseline model with the most severely distorted points as baseline points until distorted signals do not exist, so that a final baseline model and final corrected spectrogram data can be obtained. Compared with the prior art, the automatic baseline correction method can eliminate the inaccuracy of noise calculation caused by baseline distortion, accurately identify the narrow peak signals and the broad peak signals in the spectrogram, correct the spectrogram with the severely distorted baseline and prevent spectral peak distortion introduced into the complex spectrogram by baseline correction.

The invention discloses a system failure early warning method based on a baseline model and the Bayesian factor. The system failure early warning method based on the baseline model and the Bayesian factor can be applied to a dynamic complex system and particularly applied to early detection and warning of aerospace vehicle equipment failure. The method includes: using the multivariate state estimation technique to excavate the system 'individuation' baseline model on the basis of system state monitoring data; and using an actual measurement value of a system state parameter to subtract a baseline value so as to obtain a deviation value of a state parameter, analyzing a deviation value sequence by the aid of a Bayesian factor method, monitoring abnormalities of a sequence structure, and timely giving out failure early warning. The system failure early warning method based on the baseline model and the Bayesian factor solves the problems that dynamic complex system performance parameter deviation value dispersion is large, failure early characteristics are easily buried in noise, warning can be given out at the early stage of system failure, and sensitivity and robustness of the monitoring system are improved. Besides, the baseline model is easy to implement from the perspective of engineering, and the method is high in university and capable of meeting the requirements of the dynamic system on real-time monitoring.

Detecting fraud in the health care industry includes selecting a given focus scenario (e.g., prescription rate in a certain drug therapeutic class) for audit analysis, and constructing baseline models with the appropriate normalizations to describe the expected behavior within the focus area. These baseline models are then used, in conjunction with statistical hypothesis testing, to identify entities whose behavior diverges significantly from their expected behavior according to the baseline models. A Likelihood Ratio (LR) score over the relevant claims with respect to the baseline model is obtained for each entity, and the p-value significance of this score is evaluated to ensure that the abnormal behavior can be identified at the specified level of statistical significance. The approach may be used as part of a preliminary computer-aided audit process in which the relevant entities with the abnormal behavior are identified with high selectivity for a subsequent human-intensive audit investigation.

A method for generating an auto-adaptive baseline model for profiling the individual and collective behavior of a plurality of users in a network is provided. The method comprises the steps of creating a model, defining a plurality of members and a plurality of collective variables, each member corresponding to a user, and including a plurality of individual variables, defining conditions for each collective variable and individual variable, upon detecting an activity by a user, updating corresponding individual variables and collective variables, comparing updated individual variables and collective variables against corresponding conditions. If condition is met, an alert event is issued to notify designated personnel; otherwise, returning to the step of upon detecting activity. Finally, upon receiving an alert event, the designated personnel decides whether to manually redefine the conditions or to ignore the alert event. If the alert event is ignored, said conditions are automatically redefined in accordance with system defined mechanism.

A method of detecting alterations in the design of a software system over the development and maintenance life cycle of the software system is disclosed. A baseline model of the software system is generated. A current model of the software system is also generated. The current model is automatically compared with the baseline model, and any alteration in the design of the software system is identified.

A method and apparatus for specifying and generating an alteration in a model representative of a system composed of a plurality of components, the method comprising the steps of associating configuration non-specific representations with corresponding functions operable on the model, each of the configuration non-specific representations includes representations of objects and attributes of, and / or relationships among system components, associating configuration non-specific representations of desired changes to the model for enabling alterations to the objects, attributes, relationships and / or functions of the objects, attributes and relationships subject to constraints imposed by attributes of the system, and applying the configuration non-specific represented changes to the configuration of a baseline model representation of the system to generate a new configuration non-specific model where the changes have been applied.

The current application comprises four major processors for determining network abnormalities. The major difference between the current invention and all other existing systems that are being used by the network operators is that the current invention detects abnormalities by comparing with a baseline statistical model. This baseline model represents typical network traffic characteristics. When a traffic characteristic exceeds or falls outside of the baseline model, an abnormality is identified.

The invention relates to a portable failure detection system and method for an aero-engine gas path component. According to a principle that the tail gas electrostatic-charge level keeps in a certain level range under the condition that an aero-engine is failure-free, and the tail gas electrostatic-charge level can change once the gas path component has a failure, the system and the method detect the health status of the aero-engine gas path component. The system comprises a tail gas static-electricity monitoring sensor subsystem, a signal conditioning, collecting and transmitting subsystem and an industrial PC (Personal Computer). First, a baseline model of engine tail gas electrostatic signals is established, and the health status of an engine is judged through deviation of the analyzed and actually-measured characteristic parameters of the tail gas electrostatic signals and a baseline value. Aiming at the problems of false dismissal and early warning lag in the traditional engine status monitoring method, the invention designs the portable failure detection system for the gas path components. The system is used for engine ground run, can increase the safety of the engine, is low in cost, and is easily realized in engineering.

The invention discloses a civil aircraft air-conditioning system online health monitoring data acquisition and analysis method, and is applied to real-time monitoring and early monitoring and early warning of failure of a large civil aircraft air-conditioning system. As for online health monitoring of the modern large civil aircraft air-conditioning system, firstly the health baseline model of the air-conditioning system is established by using the air-conditioning system monitoring parameters under the failure-free condition and the baseline model mining technology; then the deviation value of the monitoring parameters of the key position is determined to act as the monitoring object, and the baseline value is subtracted from the actual observation value of the monitoring parameters of the air-conditioning system so that the deviation value sequence of the monitoring parameters is obtained; and then the deviation value sequence is monitored and early warning of the failure of the air-conditioning system is emitted according to the abnormal deviation value sequence. The problems that the direct monitoring performance parameters of the civil aircraft air-conditioning system have high discrete degree and the early failure characteristics are not obvious can be solved, and early warning can be timely emitted at the early stage of the system failure by monitoring the deviation value sequence so that the real-time monitoring requirement of the large civil aircraft air-conditioning system can be met.

A method of evaluating the performance of a family of target date funds. A model of the family is constructed using realized rates of return (“returns”) for an evaluation period for each of the funds. The returns for each fund are assigned to the periods of the model corresponding to the periods during which the return was realized by the fund. An estimate of the wealth generated by the family may be determined as a function of the returns assigned to the model and a pattern of contributions. A benchmark portfolio may be selected and used to construct a model of the benchmark portfolio having the same number of periods as the model of the family. The model of the benchmark and the pattern of contributions may be used to determine wealth generated by the benchmark portfolio, which may be compared to the estimate of wealth generated by the family.

System, method and computer product for baseline modeling a product or process. A service database contains process data. A preprocessor processes the data into a predetermined format. A baseline modeling component builds a baseline model from the preprocessed data, wherein the baseline model relates process performance variables as a function of process operating conditions.

In a method for evaluating data of a magnetic resonance signal, a baseline is modeled and at least one metabolite signal is modeled. The modeling of the baseline ensues based on magnetic resonance data in the time domain, and using only data that begins with the magnetic resonance signal and having a length that is shorter than the duration of the overall magnetic resonance signal.

A computerized system for recursively detecting anomalies in monitored behavior of entities. The system comprises a storage unit to store monitored events, event deviations and parameters related to each event and to each event deviation. The system comprises a processing unit configured to receive a plurality of input events, construct a plurality of baseline models, receive an input event that occurred during an analyzed timeframe, compare parameters of the received input event to a corresponding baseline model in order to detect an event deviation, and associate an event deviation score to the detected event deviation. Using the detected event deviation as an input event, said operations are repeated until a predetermined condition is satisfied, and an alert is generated, indicating suspicious activity has been detected. A viewer application configured to receive and display alerts relating to the detected event deviation is provided.

Creation of new edges in a network may be used as an indication of a potential attack on the network. Historical data of a frequency with which nodes in a network create and receive new edges may be analyzed. Baseline models of behavior among the edges in the network may be established based on the analysis of the historical data. A new edge that deviates from a respective baseline model by more than a predetermined threshold during a time window may be detected. The new edge may be flagged as potentially anomalous when the deviation from the respective baseline model is detected. Probabilities for both new and existing edges may be obtained for all edges in a path or other subgraph. The probabilities may then be combined to obtain a score for the path or other subgraph. A threshold may be obtained by calculating an empirical distribution of the scores under historical conditions.

The invention discloses an implicit intention identification and classification method based on eye movement tracking. The method comprises the following steps that a product image set is selected to act as the source of visual stimulation; areas of image (AOI) are set, the areas of image include two types of an area of image having no target and an area of image having the explicit target, and the fixation duration, the number of fixations and the pupil size of the user are acquired; a baseline model for measuring the change of the pupil size is established, the influence of the intensity change on pupil stimulation is separated by using the baseline model based on the image intensity, a reference model based on the pupil size information of the user is accordingly established and the baseline threshold is determined and two implicit intentions are distinguished: the unobjective intention and the objective intention; and the implicit intention of the user is classified into the unobjective intention and the objective intention by using a classifier based on three parameters of the fixation duration, the number of fixations and the change of the pupil size. The implicit intention identification and classification method based on eye movement tracking has the characteristics that the implicit intention of the user can be identified under different intention conditions and the identification accuracy is high.

A dependency parsing field self-adaption method based on web search comprises the following steps: establishing a reorder model according to K-Best candidate dependency parsing trees, and generating a group of candidate dependency parsing trees by a reference model, wherein each candidate dependency parsing tree corresponds to a probability to define initial sequences of dependency parsing results, and on the basis of the initial sequences, the reorder model tries to use the syntax characteristics newly increased in a target field to improve the initial syntax tree sequence; evaluating the dependent relationship between words of the candidate dependency parsing trees based on the web search: at first, splitting a dependency parsing tree into the collection of a group of word relations, wherein the intensity of each word semantic relation in the collection is calculated on the basis of the web search respectively, and then the evaluation level is determined according to the weight of the word relations. Experiment results show that a maltparser trained on a penn Chinese treebank shows favorable performance in biomedicine data.

The disclosed embodiments relate to a technique for training a prognostic pattern-recognition system to detect incipient anomalies that arise during execution of a computer system. During operation, the system gathers and stores telemetry data obtained from n sensors in the computer system during operation of the computer system. Next, the system uses the telemetry data gathered from the n sensors to train a baseline model for the prognostic pattern-recognition system. The prognostic pattern-recognition system then uses the baseline model in a surveillance mode to detect incipient anomalies that arise during execution of the computer system. The system also uses the stored telemetry data to train a set of additional models, wherein each additional model is trained to operate with one or more missing sensors. Finally, the system stores the additional models to be used in place of the baseline model when one or more sensors fail in the computer system.