16 results about "Autoregressive integrated moving average" patented technology

This disclosure relates generally to method and system for generating key performance indicator prediction model for multi-cloud applications. The disclosed method determines an optimized resource model and a predictive cost structure for one or more multi-cloud applications. The method receives a composite usage request to obtain a current resource consumption metrics and a cost structure for each cloud application identifier (ID). Further, a set of cloud provider API endpoints are invoked to obtain a plurality of usage tracking metrics. Further, a plurality of views are generated for each cloud application ID by processing every record associated with each API response file with allocated resource data. Then, a KPI prediction model is generated by leveraging autoregressive integrated moving average on the KPI time series data to determine an optimized resource model and a cost structure.

The present application relates to the technical field of building automation, in particular to an outdoor louver canopy automatic control method and system, comprising the following steps: based on historical and real-time environmental data, using an autoregressive integrated moving average model, analyzing the trend and seasonal variation of illumination, temperature and wind speed data, using a random forest algorithm to process the nonlinear relationship of the data, constructing a model to predict environmental changes, analyzing and predicting the future trend of illumination and temperature parameters, and generating environmental prediction data.In the present application, the autoregressive integrated moving average and random forest algorithm accurately predict environmental changes, the gradient boosting machine algorithm reduces prediction error, the long short-term memory network timely identifies structural damage and functional failure, enhances reliability and safety, the fuzzy logic control handles input uncertainty, improves control accuracy, the PID algorithm adjusts the control amount according to real-time data, realizes accurate adjustment, real-time data stream analysis monitors the state, and ensures effective execution of the strategy.

The invention discloses a networking delay dynamic sensing and intelligent compensation method for a distributed test system. According to the method, a technical system of'high-precision perception-intelligent prediction-dynamic compensation-robust guarantee 'is constructed aiming at the networking delay and time variation problem caused by fixed delay, variable delay and system-level errors in a distributed test system; a subnanosecond clock synchronization reference is established through an IEEE 1588 (PTP) precision time protocol; unidirectional delay accurate measurement is realized based on a PTP delay request-response mechanism; modeling and predicting the delay time sequence by adopting an ARIMA (Autoregressive Integrated Moving Average) model, and sending an instruction in advance through dynamic forward compensation so as to achieve a quasi-synchronization test; and the test continuity and the precision controllability when the network state dramatically changes are ensured by combining the self-adaptive model updating and abnormal backspacing mechanism. According to the method, the time sequence synchronization precision and the environmental adaptability of the distributed test system are remarkably improved, and the technical problem that a traditional method is difficult to deal with delay jitter and dynamic change is solved.

A system can maintain a group of data processing units and a storage array that comprises a group of sub-logical unit numbers of storage. The system can collect, by a central processing unit, first data indicative of input and output events for the storage array. The system can process, by respective data processing units, respective autoregressive integrated moving average models for respective sub-logical unit numbers of the group of sub-logical unit numbers with the first data, to generate respective statuses that indicate respective frequencies of access of the respective sub-logical unit numbers. The system can determine, by the central processing unit, respective classifications for respective sub-logical unit numbers of the group of sub-logical unit numbers of storage based on the respective statuses. The system can compress, by a compression engine, second data stored in at least some of the respective sub-logical unit numbers based on the respective classifications.

The invention discloses a long-term load prediction system and method for a data center, and the system achieves the explicit modeling of trend, seasonal and random components in a load sequence through the optimization and application of a seasonal differential autoregressive moving average model (SARIMA), especially the parameterized description of a long-term seasonal rule. Therefore, the prediction precision at seasonal transition points and periodic peaks and valleys is obviously improved. According to the method, the characteristic that the SARIMA model is simple in structure is utilized, a reliable prediction model which is relatively low in historical data size requirement and not prone to overfitting is constructed, and stable output in practical application is ensured. According to the method, a set of standardized and reproducible prediction process is provided, so that each step of operation has a clear statistical basis, and the interpretability of a prediction result and the credibility of management personnel are enhanced. According to the method, a solution with excellent balance between prediction precision and calculation resource consumption is realized, so that the dual requirements of practical engineering application on timeliness and economy are met.

The invention discloses a landslide displacement prediction method based on MSCNN-Attention-BiLSTM and ARIMA (autoregressive integrated moving average). The landslide displacement prediction method comprises the following steps: S1, data decomposition; s2, precipitation time lag analysis; s3, trend term prediction; s4, predicting a periodic term; s5, performing overall fusion; according to the method, the precise modeling of the multi-scale deformation characteristics of the landslide is realized by constructing the landslide displacement prediction framework fusing CEEMDAN decomposition, deep learning and time sequence modeling, so that the precision and stability of the total displacement prediction of the landslide are remarkably improved; according to the method, a CEEMDAN algorithm is introduced to decompose landslide displacement data, and an MSCNN-Attention-BiLSTM model and an ARIMA model are combined to model a trend term and a period term respectively, so that structured extraction, attention weighting and modular prediction of multi-source time sequence features are realized, and the interpretability and robustness of displacement prediction are enhanced.

The invention discloses a fire-fighting system alarm method based on multi-sensor coupling data support utility index prediction, and relates to the technical field of fire-fighting system alarm, and the method comprises the steps: synchronously collecting real-time measurement data, and forming respective time sequences; carrying out normalization processing on various measurement data by adopting a minimum-maximum normalization method to obtain a normalized time sequence; an autoregressive integrated moving average model algorithm is adopted for the normalized time sequence, and normalized values of various parameters at the next moment in the future are predicted; and calculating a multi-sensor coupling data support utility index according to the predicted normalized values of various parameters at the next moment in the future, evaluating the coupling utility of the sensors, and triggering a fire behavior judgment strategy. According to the method, comprehensive identification of early characteristics of the fire is realized, so that the system can process abnormal data in a complex environment more stably, technical support is provided for upgrading of an intelligent fire-fighting system, and the method has engineering application value and popularization significance.

The application discloses a high-voltage cable grounding loop resistance prediction method based on data driving, collects historical grounding loop resistance detection records of high-voltage cable sections with different static attribute labels under the same voltage level; adopts iterative threshold variational mode decomposition (ITVMD) to decouple irregular interval multi-loop resistance time series into a plurality of intrinsic mode components with different frequency characteristics; splices frequency domain characteristics of loop resistance mode components with static attributes to construct a comprehensive feature vector, and clusters the comprehensive feature vector through a K-means algorithm; combines a two-way delay embedding transformation (TDT) algorithm, and based on Tucker decomposition, adopts a block Hankel tensor autoregressive integrated moving average (BHT-ARIMA) multi-sequence prediction model to predict the high-voltage cable grounding loop resistance. Compared with the prior art, the application effectively solves the technical problems of low prediction accuracy caused by sparse grounding loop resistance data, complex coupling and missing values, and realizes accurate prediction of the loop resistance.

The invention relates to the technical field of online diagnosis, in particular to an equipment abnormal state online diagnosis method. The method comprises the following steps: S1, acquiring the temperature, vibration, pressure and current of equipment in real time through a sensor and monitoring equipment; s2, accurately judging whether the equipment has a fault or not based on the collected equipment data through a local anomaly factor LOF, and giving an alarm through a threshold setting algorithm; s3, automatically analyzing the acquired equipment data by adopting an autoregressive moving average (ARIMA) model and predicting the future operation state, performance index and fault risk of the data; and S4, analyzing a fault reason through a support vector machine (SVM), and providing a detailed diagnosis report and a repair suggestion. According to the design, the definition of the LOF algorithm on the anomaly can be automatically adjusted to adapt to data distribution of different density regions, and points which are remarkably different from other sample points in a local region can be identified, even if the points are in a global range, the points do not appear to be abnormal.

The invention discloses a remote anti-electricity-stealing monitoring and alarming method based on artificial intelligence, and particularly relates to the field of power monitoring, which comprises the following steps: S1, multi-source data acquisition: acquiring user electricity consumption data and power grid operation data based on a preset frequency, S2, primary data processing: performing abnormal value processing, noise data processing and missing data processing on the acquired data, and S3, carrying out secondary data processing on the acquired data. S3, carrying out secondary data processing, carrying out standardization processing on numeric data after primary processing, and carrying out data coding on non-numeric data; S4, carrying out electricity consumption abnormity marking based on time sequence analysis, constructing an autoregression integral moving average model, predicting future electricity consumption power and comparing the future electricity consumption power with real-time power; S5, carrying out electricity stealing identification, and establishing a theoretical calculation model to calculate power deviation. And extracting harmonic characteristics to calculate a distortion rate, and comparing with a preset threshold to judge an electricity stealing behavior. Through multi-dimensional data acquisition and intelligent analysis, the anti-electricity-stealing identification precision is improved, and stable operation of a power system is effectively guaranteed.

The invention relates to the technical field of mechanical equipment fault diagnosis, in particular to a gear wear state prediction method based on a time sequence ARIMA (autoregressive integrated moving average) model, which is characterized by comprising the following steps: step S10, constructing a prediction model based on ARIMA (p, d, q): step S11, constructing an abrasive particle group characteristic time sequence; step S12, carrying out stability inspection; step S13, determining the order of the model; step S14, estimating parameters; step S15, carrying out model inspection; and step S20, analysis and verification of a prediction result based on the ARIMA model. According to the method, the wear state is predicted by establishing the time sequence ARIMA prediction model for the constructed abrasive particle group feature time sequence, the wear severity can be analyzed and judged in advance, and the failure rate of mechanical equipment is effectively reduced.

The invention belongs to the technical field of corrosion protection and prediction of ocean engineering structures, and particularly relates to a method for predicting the cathode protection potential of an FPSO stand pipe supporting structure. According to the prediction method, the advantages of the seasonal autoregression integral moving average model and the long short-term memory neural network are combined, collaborative modeling and high-precision prediction of linear and nonlinear characteristics in the potential data are achieved, the prediction precision is high, and the prediction result robustness is high. The method for predicting the cathode protection potential of the FPSO riser support structure comprises the following steps: collecting historical potential time sequence data of the FPSO riser support structure under the condition of external cathode protection potential, and preprocessing the historical potential time sequence data; constructing and training a seasonal autoregressive integral moving average model; constructing and training a long-short-term memory neural network model; and performing combined prediction on the target time period to obtain a prediction result of the target time period of the cathode protection potential of the FPSO riser support structure.

The application discloses a DC fault arc detection method based on synchronous compression wavelet transform and ARIMA algorithm, utilizes electromagnetic radiation signals of a DC fault arc obtained by a Hilbert antenna, obtains time-frequency information of the electromagnetic signals through synchronous compression wavelet transform, adopts an autoregressive integrated moving average (ARIMA) model to perform fault detection, and if the fault detection result is fault in three continuous time windows, outputs a fault arc cutting signal, so that accurate prediction of the DC fault arc is realized by analyzing the characteristics of the electromagnetic radiation signals. Through synchronous compression wavelet transform, the application mines characteristic information of the electromagnetic radiation signals of the DC fault arc under different current levels, different electrode materials, different antenna measurement distances and angles, and more accurately and quickly cuts off the DC fault arc, and the universality of the fault arc detection method and the safe and stable operation ability of the system are improved.

The application provides a power transaction business security risk prediction method and device based on time sequence detection, the method comprises the following steps: preliminarily dividing linear time sequence data components from the time sequence data; using an autoregressive integrated moving average model to evaluate the linear fitting significance of the divided linear time sequence data components, and then generating a linear business security risk prediction result; performing residual error processing on the to-be-observed time sequence data and the time sequence data with linear fitting significance exceeding a preset threshold in the linear time sequence data components, to obtain nonlinear time sequence data components; using a multilayer stacked long short-term memory network to extract features from the nonlinear time sequence data components, and performing nonlinear prediction to generate a nonlinear business security risk prediction result; and comprehensively combining the linear business security risk prediction result and the nonlinear business security risk prediction result to obtain a final business security risk prediction result through an activation function.

A system for detecting knock using an autoregressive integrated moving average (ARIMA) model. The system includes an electronic processor. The electronic processor is configured to receive a signal from a knock sensor and determine components of the signal using an ARIMA model, where the components include a residual component. The electronic processor is further configured to determine whether combustion is knock or non-knock combustion based on a magnitude of the residual component.