9 results about "Period length" patented technology

The invention discloses a time sequence prediction method and system based on adaptive period segmentation and parallel decoding, and the method comprises the following steps: judging the period length of time sequence data, segmenting the time sequence data according to the period length, and obtaining time sequence segmentation data; adaptively adjusting the hiding dimension of the embedded layer according to the period length, and mapping the time sequence segmentation data into the hiding dimension to obtain a time sequence representation; the obtained time sequence representation is input into a Transform encoder to carry out feature extraction; the decoder input is initialized according to the Transform encoder output, the initialized decoder input and the encoder output are subjected to feature interaction in the decoder based on a cross attention mechanism, and finally prediction output is obtained. According to the method, a more flexible period segmentation scheme and a decoding mechanism are adopted, the reasoning efficiency is improved, accumulative errors are reduced, and the time sequence basic model can achieve a better prediction effect with fewer parameters.

The application relates to the technical field of data processing, and discloses a time series-based data processing method, device, equipment and storage medium, the method comprises the following steps: determining a period length, when current data does not have periodicity, the period length is infinite, based on the data length of batch processing corresponding to a target time window, data processing delay and data generation interval, adjusting the memory size to realize time series-based data processing; when the current data has periodicity, based on the period length, adjusting the memory size to realize time series-based data processing. In the foregoing manner, the application adaptively identifies the period of the current data; the hash value of the current data is calculated; based on the hash value of the current data and the hash values of the previous n windows, the period length of the data is determined, and when the data does not have periodic characteristics, the period is infinite. Through pre-allocation of the memory required for data processing, the frequent creation of new data and the memory recovery of garbage data are reduced, and the performance and reliability of data processing in the Internet of Things are improved.

A prediction device according to the present invention comprises a prediction unit that calculates a first fraction for a first period and a second fraction for a second period for each phase of a plating layer on the basis of how much of each of a plurality of components is included in the plating layer, calculates the period length of the first period, which is until a portion of the corrosion of the plating layer reaches a base steel, on the basis of a corrosion rank for each phase, the first fraction for each phase, and a first corrosion speed for each phase, calculates the period length of the second period, which follows the first period, on the basis of the corrosion rank for each phase, the second fraction for each phase, and a second corrosion speed for each phase, and calculates the total of the period length of the first period and the period length of the second period as prediction results for the corrosion resistance of the plating layer.

A periodic detection method based on a time sequence comprises the following steps: acquiring the time sequence, preprocessing and encrypting the time sequence, and respectively setting a noise threshold value and a trend component proportion threshold value; carrying out self-correlation function calculation on the originally obtained time sequence; calculating the similarity difference of each sub-sequence of the autocorrelation sequence A; obtaining a period length T, a noise degree and a trend component proportion based on a minimum value of the similarity difference sequence; the noise degree and the trend component proportion of the sequence are calculated, and encryption is carried out after correction is carried out based on a preprocessing result; by means of the method, the problems that the period detection efficiency and accuracy of the time sequence period with the low signal-to-noise ratio and the trend component are low can be solved, and the operation and maintenance information processing efficiency and accuracy are improved.

The invention discloses a database management method and device and a storage medium, and the method comprises the steps: determining the cycle length of a first index based on a data life cycle feature corresponding to the first index; the number of fragments of the first index in a target period is determined, and the target period is a current period or a future period; and creating the first index corresponding to the target period in the database according to the fragment number and the period length. In this way, the resource storage efficiency of the database can be improved.

The embodiment of the invention discloses a prediction method and a communication device, relates to the technical field of communication, and can reduce the overhead of reporting prediction information of a predicted future moment or time period by terminal equipment. The method comprises the following steps: receiving first indication information, wherein the first indication information is used for indicating execution of a first prediction task; determining a first time unit, the first time unit being one of the at least two candidate time units; and / or determining a first time period length, the first time period length being one of the at least two candidate time period lengths; determining the number N of prediction information contained in each report in at least one report corresponding to the first prediction task, wherein N is a positive integer; a first report corresponding to the first prediction task is determined according to at least two of the first time unit, the first time period length or N, the first report is one of the at least one report, and the first report comprises N pieces of prediction information; and sending the first report. The embodiment of the invention is used in a prediction information reporting process.

The invention provides a resource exclusion method during sidelink transmission resource selection, which is applied to a terminal and comprises the following steps: aiming at each period supported by a preset resource pool, determining all historical sidelink control information SCI corresponding to the period in a sensing window according to the period length of the period; and according to all historical SCIs corresponding to each period, performing resource exclusion on the sensing window to obtain a candidate resource set. The invention further provides a terminal, a computer readable medium and a computer program product.

A signal transmission method and an apparatus. A first time domain resource is determined based on a first frequency domain resource, a second frequency domain resource, a period length, and a period index, where the first frequency domain resource is selected from a plurality of candidate frequency domain resources, and the second frequency domain resource is a frequency domain resource for sending a reference signal in a first period or a frequency domain resource determined based on a configuration parameter. The reference signal is sent on the first frequency domain resource and the first time domain resource. In this way, a plurality of candidate frequency domain resources can be provided for reference signals to be periodically sent, to increase a possibility of sending the reference signal, and improve sensing performance.

The invention provides an anomaly detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining long time series data which comprises a plurality of first data points divided according to a preset initial time granularity and the cycle length information of the first data points; the period length information is adopted to convert the first data points of the long time sequence data, new time sequence data is obtained, and the new time sequence data comprises a plurality of second data points divided with the period length as the granularity; obtaining a target data point corresponding to the second data point according to a preset correlation condition, and obtaining a similarity value between the second data point and the corresponding target data point; when it is detected that the similarity value is smaller than a preset threshold value, it is determined that the second data point is abnormal, and whether the abnormity of the second data point meets the alarm noise reduction condition or not is judged; and if yes, performing alarm and noise reduction on the abnormity of the second data point. According to the embodiment of the invention, the processing efficiency and prediction result accuracy of intelligent operation and maintenance are improved.