229 results about "Prediction interval" patented technology

The invention provides a fault predicting and diagnosing method suitable for a dynamic complex system. The method can be applied in the field of fault prediction and diagnosis of dynamic complex systems of spacecrafts and the like. The method comprises the following steps of: performing failure mode and effect analysis (FMEA) on the dynamic complex system to obtain a main fault mode and corresponding performance detection parameters, dividing the performance detection parameters into slowly variable data and fast variable data, pre-processing the performance detection parameters, establishingan autoregressive moving average model (ARMA) aiming at the slowly variable data to perform time sequence prediction, establishing a multi-resolution wavelet neural network aiming at the fast variable data to perform time sequence prediction, performing fault early warning on the time sequence prediction results by establishing a prediction interval model, and performing fault diagnosis by establishing a D-S (Dempster-Shafer) evidence theory-based multi-signal fusion model. The method can be used for predicting and diagnosing the faults of the dynamic complex system with high precision, and has strong universality.

The invention provides a resource self-adaptive adjusting system and method of multiple virtual machines under a single physical machine. The system is achieved in the single physical machine and comprises a data collecting module, a preprocessing module, a prediction module, a resource adjustment strategy generating module, a resource adjustment strategy gain calculating module, a monitoring module, a strategy evaluation module and a historical database. The method comprises the steps that the historical data of a server are collected and stored in the historical database; the historical data of the server is preprocessed; the concurrent user request amount at the next moment is predicted, and then the demand amount prediction intervals of virtual machine resources is obtained by the predicted value of the concurrent user request amount; an optimal resource adjusting strategy is determined; CPU resource adjustment and internal storage resource adjustment are conducted; the optimal resource adjusting strategy is evaluated; the current optimal resource adjustment strategy and the evaluation value of the current optimal resource adjustment strategy are stored into the historical database. According to the resource self-adaptive adjusting system and method of the multiple virtual machines under the single physical machine, the resource amounts of all the virtual machines on the single physical machine can be adjusted to be adaptive to dynamically-changed resource demands, and therefore the resources of the single physical machine can obtain the biggest benefits.

The present invention belongs to the field of power prediction of wind power generation and particularly relates to a method for predicting a wind power interval based on a particle swarm optimization nucleus limit learning machine model. The method comprises: carrying out data preprocessing, i.e. preprocessing historical data in SCADA according to correlation between a wind speed and power; initializing a KELM model parameter and carrying out calculation to obtain an initial output weight betaint; initializing a particle swarm parameter; constructing an optimization criterion F according to an evaluation index and carrying out particle swarm optimization searching to obtain a model optimal output weight betabest; and bringing test data into a KELM model formed by betabest to obtain a wind power prediction interval and evaluating each index of the prediction interval. The method is easy for engineering realization; a good prediction result can be obtained; not only can a future wind power possible fluctuation range be described, but also reliability of the prediction interval is effectively evaluated, possible fluctuation intervals of wind power at different confidence levels are given out and reference is better provided for a power system decision maker.

A method and system is used to construct a forecast error confidence interval. The predication interval provides a range of error for a current forecast value to any desired confidence level. The method and system involve running a forecast method on a set of historical data. For each historical period, a forecast is obtained at each time point in the period. The forecasts are compared to the target value of interest in each period. The comparison of target values to forecast values is used to build an error series for each historical period. The error values within each error series are pooled to form an error distribution series. The error distribution series can be used to provide a confidence interval for the current forecast.

The invention provides a computer cluster performance index detection method, electronic equipment and a storage medium. The computer cluster performance index detection method includes the steps of extracting performance time sequence data with periodic forms in a certain period of time from a historical database; conducting modelling on the performance time sequence data, and determining a timesequence model; calculating a fitting error of a preset initial step length of historical data according to the time sequence model; predicting a threshold interval of a preset future step length according to the fitting error and preset reliability; detecting whether or not an actual value corresponding to the preset future step length of a computer cluster performance index is located in the threshold interval, determining that the index is normal if yes, and determining that the index is abnormal if not. According to the computer cluster performance index detection method, the electronic equipment and the storage medium, the fitting error of the corresponding step length of the historical data is automatically calculated according to the predicted step length, then the prediction interval is determined according to the error, a more reasonable threshold range is conveniently designed for the prediction interval, and the missing alarm rate or false alarm rate of abnormality detectionis reduced.

The invention discloses a statistical analysis method of accelerated life tests based on a fuzzy theory. The statistical analysis method comprises a first step of reasonably fuzzifying failure terminated censoring of constant-stress accelerated life tests, thereby obtaining fuzzy failure data, a second step of building a fuzzy statistical model of the accelerated life tests by use of a maximum likelihood method, and a third step of performing estimation on model parameters and fuzzy prediction on lifetime and reliability. The statistical analysis method is capable of providing the fuzzy estimation values of the model parameters according to the built model and further providing the fuzzy lifetime prediction interval and the fuzzy reliability interval of products; compared with point estimation values provided by a traditional statistical analysis method, the results of the statistical analysis method are more reasonable and have better reference value.

The invention discloses a method and a device for monitoring energy consumption abnormity. The method comprises the following steps: monitoring an energy consumption parameter of a current period of a vulcanization process in a tyre manufacturing process; according to the value of the energy consumption parameter of the current period obtained through monitoring, obtaining an energy efficiency monitoring value and an energy efficiency predicting value of the current period of the vulcanization process in the tyre manufacturing process; according to the obtained energy efficiency predicting value and a preset confidence coefficient, obtaining the energy efficiency prediction interval of the current period; when the energy efficiency monitoring value is beyond the energy efficiency prediction interval, determining that abnormity occurs in the energy consumption of the current period of the vulcanization process in the tyre manufacturing process. The method and the device can provide reasonable monitoring for complex and variable energy consumption situations in the industrial production process, missing detection and false detection are reduced, and the precision of energy consumption abnormity monitoring is improved.

The invention provides a real-time agricultural Internet of Things data stream abnormality detection and processing method and a real-time agricultural Internet of Things data stream abnormality detection and processing device. In the method, the size of a sliding window is first determined according to a characteristic cycle of acquired data and a time interval between acquisitions, and a measured value acquired by a sensor at the current moment and a prediction interval are then predicted according to historical acquired data in the sliding window; an actual measured value which is actually acquired at the current moment is then compared with the precision interval, if the actual measured value does not fall into the prediction interval, then data which are actually acquired at the current moment are determined as abnormal data, and thereby the detection of abnormal data is fulfilled. In addition, after the method provided by the invention determines that the data at the current moment are abnormal data, the obtained predicted value is utilized to replace the abnormal data, and thereby the processing of the abnormal data can be fulfilled. The accuracy of acquired data streams is effectively increased, and powerful data support is provided for the automatic control of equipment and effective data analysis.

The embodiments herein disclose a system and method for particle swarm optimization and quantile regression-based rule mining for analyzing data sets involving only continuous explanatory variables. The system discloses an architecture for PSO based quantile regression rule mining for determining the prediction intervals (PIs). The system generates ‘if-then’ rules that yield PIs while solving a multiple regression problem having only continuous explanatory variables. The system performs an ensembling process to reduce the size of the rule base to a manageable number based on the quality metrics of prediction intervals. The system comprises a data set, and a rule miner designed to divide the data into deciles based on the descending order of the target attribute variable. PSO is invoked to derive a set of rules for each decile and capture the heteroscedasticity of the distribution of the data with the help of quantile regression, in a non-traditional way.

The invention discloses a distribution network scheduling method comprehensively considering photovoltaic output and load demand prediction intervals. The distribution network scheduling method comprises the steps of firstly building a photovoltaic output prediction interval model based on the illumination intensity which approximates to Beta distribution, secondly building a load demand prediction interval model by combining an empirical mode decomposition method and a sparse Bayesian learning method on the basis of a power consumption load layering probability prediction method, and finallyputting forward a scheduling model considering the distribution network operating reliability and economy based on the photovoltaic output prediction interval and the load demand prediction interval.The method provided by the invention not only solves a problem of power supply unreliability caused by uncertainty of photovoltaic power generation output, but also reduces the distribution network economic operation cost caused by difficult prediction for the photovoltaic power generation and load. The distribution network scheduling method can be widely applied to distribution network schedulingof a power network.

The invention relates to a prediction method for a residual discharging energy of a battery based on prediction of future operation condition, and belongs to the technical field of battery management. Operation condition data of the battery is collected, and a further output power and a further temperature change rate of the battery are predicted; an internal resistance parameter is identified from a battery equivalent circuit model, and a curve that the internal resistance parameter changes along with the state of charge (SOC) is updated; a prediction interval of the SOC of the battery is determined, and a future SOC sequence of the battery is calculated; a future voltage sequence, a future current sequence and a future temperature sequence of the battery are predicted; and the residual discharging energy of the battery is calculated. According to the method, influence of the future operation condition on the residual discharging energy of the battery is considered, the residual discharging energy of the battery can be predicted in real time, and higher prediction precision can be ensured in different operation conditions.

The invention discloses a repair decision-making method of health state of an airplane system. The method includes steps of 1), calculating preventive repair threshold value uO and a predicting interval h of a target unit of an airplane system; 2), calculating the average rest life of the target unit at kh moment, wherein k is the frequency of preventive repair of the target unit within a certaintime scale; 3), when the average rest life of the target unit at kh moment is less than the preventive repair threshold value uO, performing preventive repair on the part; otherwise, not performing the preventive repair. The method can scientifically and reasonably confirm the preventive repair threshold value and the predicting interval; on the basis of guaranteeing the availability of the systemparts, the repair decision-making method can reduce the average repair cost of the system part, effectively solve the 'excessive repair' or 'under-repair' situation easily caused by preventive repairat a fixed time, and avoid the waste of repair resource.

The invention discloses a method and a device for identifying abnormal data of monitoring index data. The method comprises the steps of: obtaining index data at a current moment, determining the prediction interval of the index data according to the historical index data in a preset period, and determining the fluctuation value of the index data according to the prediction interval and the index data at the current moment; when the fluctuation value does not accord with a fluctuation threshold, determining that the index data is problem data, the fluctuation threshold being obtained by comparing the index data with historical index data in a first time period before the current moment; and if the index data at the current moment is stable, determining the problem data as abnormal data. Theindex data at the current moment is identified according to the historical index data, and the alarm accuracy is improved; and the problem data is determined as abnormal data according to the stability of the index data at the current moment, so that the accuracy of identifying the abnormal data is improved, and false alarms caused by data increment or data cutting are reduced.

A receiver configured for: a) receiving (410) a first OFDM symbol and generating a plurality of demodulated symbols for the first OFDM symbol; b) generating (420) decoder output code symbols corresponding to a subset of the plurality of demodulated symbols; c) determining (430) that a set of the decoder output code symbols make up a set of reference symbols corresponding to at least a portion of the subset of the plurality of demodulated symbols; d) generating (440) the set of reference symbols; e) generating (450) a set of channel estimates based on the set of reference symbols and the at least a portion of the subset of the plurality of demodulated symbols, for use in decoding a current OFDM symbol; and f) repeating steps b-e until a channel estimate for each demodulated symbol corresponding to the first OFDM symbol has been generated.

The invention discloses a method and device for determining the size of a congestion window, relating to the technical field of mobile Internet. The invention aims to solve a problem that a congestion window size can not be accurately adjusted. The method comprises a step of obtaining the sampling data including multiple ACK data packets with a first preset time length, a step of dividing the first preset time length into multiple candidate prediction intervals according to time, a step of determining the target transmission delay corresponding to each ACK data packet, a step of calculating the correlation coefficient between the average bandwidth of each candidate prediction interval and the average bandwidth of each candidate prediction interval after each target transmission delay, a step of determining a target prediction interval and a prediction time period according to the correlation coefficient, and a step of establishing a TCP connection, periodically predicting the average bandwidth of a next transmission delay according to the prediction time period starting from the reception of a first ACK data packet, and thus periodically determining the size of the congestion window. The method and the device are applied to the process of adjusting the congestion window in mobile Internet.

The invention discloses a distributed control method for large-scale heating and ventilation air-conditioning systems in university campus buildings. Coordinated operation of the heating and ventilation air-conditioning systems in all regions in a plurality of buildings is achieved by adopting a model predictive control and alternating direction multiplier method, and the sum of the energy cost and the heat energy non-comfort cost is minimized under the constraint that the total input power and the acceptable temperature range of all the regions are not violated. The distributed control methodparticularly comprises the following working procedures that (1) a local controller of the heating and ventilation air-conditioning system in each region is used for predicting the electricity pricein a prediction interval, the external temperature and the user comfort degree preference temperature; (2) optimal input power decisions for a plurality of time slots in the future are given by all the local controllers according to a control algorithm; (3) the input power decision of the first time slot is made to act on actual operation of the heating and ventilation air-conditioning system; and(4) the prediction interval rolls back to one time slot, and the above three steps are repeatedly conducted. The method has the advantages of being high in expandability, high in flexibility, capableof protecting the privacy of a user and the like.

The invention provides a wind-water-fire joint robust unit combination method. Firstly, an uncertain set is established for the uncertainty of wind power output, and the uncertainty of the wind poweroutput is described in three aspects of a wind power output prediction interval, a time smoothing effect and a spatial cluster effect. Then, a mixed integer linear programming model for water and power dispatching is built by using a linearization method, and the model is integrated in a robust unit combination model considering wind and power dispatching to obtain a wind-water-fire joint robust dispatching model. Finally, a robust optimization model in two phases is solved by adopting a C&CG method. Compared with a traditional robust unit combination model, the wind-water-fire joint robust unit combination model has the advantages that the operation cost is reduced, and the starting number and the running time of thermal power units are reduced, so that the carbon emission is reduced, andthe environmental benefit is achieved; the wind power consumption capacity of a system is increased due to the addition of water power; and the problem of a unit combination containing cascade hydropower stations can be solved within a reasonable time range.

The invention relates to a wind power station wind speed prediction method based on wavelet analysis and a system thereof. The method comprises the following steps: according to a specific prediction time interval, determining an input and an output variable of a prediction model; reading a historical wind speed value and correcting an incomplete point in the historical wind speed value so as to acquire a training sample value sequence of a wind speed prediction model; carrying out rapid wavelet decomposition to the training sample value sequence so as to acquire an approximation detail component value sequence; establishing the wind speed prediction model according to the approximation detail component value sequence so as to carry out the wind speed prediction. According to the wind power station wind speed prediction method based on the wavelet analysis and the system of the invention, through the wavelet decomposition, the training sample value sequence is decomposed into different layers according to a scale so that a trend term, a period term and a random term are separated. Each layer is individually analyzed and predicted and finally the corresponding prediction value can be obtained through reconstruction. By using the method, any prediction interval can be selected according to different demands. The wind speed prediction which is many steps ahead and has high precision can be performed.

A movement predicting method is disclosed. The method utilizes at least one phone communicating in and moving between neighboring cells of base stations to predict population movement in a prediction interval. Firstly, obtain the traffic volumes in the cells generated by the phone and the handover information generated by the phone moving between the cells in a day. Next, calculate the traffic volumes and the handover information to obtain a movement probability for the population moving between the cells and an average residence time that the population stays in the region of each cell in the different periods of the day. Finally, according to the data obtained, predict an appearance probability that the population appears in each region at the end point of the prediction interval.

The invention discloses an accelerated degradation experiment modeling method based on a fuzzy theory. The accelerated degradation experiment modeling method comprises the specific steps of 1 utilizing the fuzzy theory to enable degradation data to be reasonably fuzzified so as to obtain fuzzy degradation data, 2 utilizing fuzzy degradation regression to establish an accelerated degradation test fuzzy linear degradation model and 3 performing model parameter evaluation and reliability degree prediction. According to the established fuzzy linear degradation model, a fuzzy evaluation value of a model parameter is given, and a fuzzy life prediction interval and a fuzzy reliability interval of a product are further given. Compared with a point estimated value given by means of a traditional statistical analysis method, the result obtained by means of the accelerated degradation experiment modeling method is reasonable and has reference value.

The invention provides an indoor temperature prediction method for an air conditioner. The indoor temperature prediction method comprises the following steps that a prediction model is established, wherein a plurality of outdoor environment temperatures and a plurality of indoor environment temperatures are sampled according to the same first sampling frequency; a sampling temperature difference of each corresponding sampling outdoor environment temperature and each sampling indoor environment temperature is calculated; the sampling temperature differences are sorted, and prediction temperature differences are determined; the temperature differences in a sampling chamber are calculated; prediction parameters are calculated; the indoor temperatures are predicted, and the prediction of the indoor temperature comprises the following steps of setting a first prediction interval time, obtaining the outdoor prediction environment temperature at the end of the first prediction interval time,and sampling the indoor temperature in the current state; and the indoor environment temperatures are calculated at the end of the first prediction interval time. The invention further discloses the air conditioner. According to the method, the variation trends of the indoor environment temperatures can be predicted in advance, and data basis is provided for intelligent control of a subsequent airconditioner. A user uses the change trends of the indoor environment temperatures to accurately control the air conditioner to start and stop, and the indoor temperature prediction method has the advantage of being high in intelligent degree.

The invention relates to a method for predicting fault of an electromechanical device based on a combined prediction model. The method comprises the steps of: (1) acquiring monitoring data of an operating state of an industrial filed from an industrial filed monitoring system, and extracting fault sensitive characterization factors to be taken as a prediction time sequence; (2) primarily selecting uniterm predicting models, and predicting various signal data in the prediction time sequence respectively at one prediction interval by utilizing the primarily selected uniterm predicting models; (3) determining a proper prediction accuracy evaluation index in the conventional evaluation index according to experiment, so as to carry out containment detection on the primarily selected uniterm predicting models for determining that whether the models are selected in a combined prediction model bank; (4) calculating comentropy value of the uniterm prediction models at the ith moment for the prediction time sequences, and determining weight coefficient of the uniterm prediction models at the ith moment; and (5) predicting the (i+1)th moment according to the weight coefficient omega j(i) by utilizing the combined prediction value fj(i+1) of the jth prediction method at the (i+1)th moment, so as to obtain the prediction value at the (i+1)th moment. The method is widely applied in various large electromechanical devices.

The invention relates to a model prediction control method of stabilizing wind power fluctuation with hybrid energy storage, and belongs to the technical field of new energy power generation control. The method comprises the steps of determining a topological structure of a control object, namely a system structure comprising wind power, battery energy storage and supercapacitor energy storage, predicting an output value of the wind power within a control prediction interval by utilizing an average value method, determining an objective function and three constraint conditions of optimization control, solving an optimization problem at each moment, only obtaining optimization control quantity at the current moment, inputting the optimization control quantity into a battery and a supercapacitor for control respectively, summing power output by the battery and the supercapacitor and actual output power of the wind power, and then obtaining grid-connected power of an overall system after power fluctuation stabilization. The control method can achieve the power fluctuation stabilization of the wind power more economically and reliably, is simple in calculation, and facilitates engineering realization.

The invention discloses a combined prediction method for a wind power output interval, and aims at solving the problems that wind power generation is relatively undetermined due to influence of natural wind speed and the prediction precision is decreased due to advanced prediction time. The range of wind power output is predicted, the predicted wind power output range is set in a combined manner based on three methods, namely the wind speed change ratio, the predicted value change rate and the practical power optimized value, and the optimal power wind output prediction interval in each period is selected according to historical wind power output data. According to the different parameters including the wind speed change ratio, the predicted value change rate and the practical power optimized value are integrated, the optimal prediction interval is selected, and the wind power output can be predicted more accurately.

The invention discloses an annual maximum load prediction method based on an engineering consultation industry expansion and temperature reduction model. The method comprises the following steps of: a historical data processing step: obtaining an annual maximum load of each year in the history and a maximum temperature at a maximum load day, carrying out accumulative temperature correction on the maximum temperature at the maximum load day of each year in the history, calculating an annual basic load of each year in the history, and calculating a temperature sensitive coefficient at each temperature; a load reduction step: determining a maximum reference temperature, solving an adjustment coefficient, and calculating an annual reduction maximum load of each year in the history; a load extrapolation prediction and result adjustment step: establishing a regression model of engineering consultation industry expansion and temperature reduction, carrying out extrapolation prediction on an annual reduction maximum load of a target year, determining a prediction interval of an annual maximum temperature of the target year, and determining an annual maximum load prediction interval according to the prediction interval of the annual maximum temperature of the target year.