5 results about "Correlation ratio" patented technology

Various embodiments are directed to configuring or training deep neural network (DNN) machine learning models comprising one or more hidden layers and an output layer. Various embodiments provide technical advantages in training DNN machine learning models, including improved computational efficiency and guaranteed optimality. In one embodiment, an example method includes identifying a nonlinear-model-based representation for each hidden layer, which may be a Bank of Wiener Models, a nonlinear units of the hidden layer, and / or the like. The method further includes individually and sequentially configuring the hidden layers, each configured by determining a correlation measure (e.g., a correlation ratio) between the layer output and a target signal. Parameters of the particular hidden layer are modified by maximizing the correlation measure to yield maximal correlation over the space of functions. The method further includes performing automated tasks using the DNN machine learning model after configuring its parameters on a training set.

This invention discloses an intelligent detection system and method for rotor bar breakage faults in asynchronous motors, belonging to the field of intelligent motor fault diagnosis technology. By deploying multi-dimensional sensors on the motor and aligning them over time, the system divides the continuous data stream into fixed-duration analysis windows. It calculates the characteristic average values ​​of each physical quantity within each window, using the average current intensity as a benchmark, and calculates the correlation ratios with torque, noise, temperature rise, and vibration. This quantifies the energy conversion efficiency and the response characteristics of each physical quantity to current, and monitors short-term fluctuations in each correlation ratio between adjacent windows in real time. When these fluctuations exceed a threshold, an instability warning is triggered. Simultaneously, the real-time correlation ratios are compared with pre-stored baseline data on bar breakage status to calculate single-dimensional differences, which are then fused to generate a comprehensive health index, enabling a quantitative assessment of the rotor's mechanical integrity degradation trajectory. Based on the comprehensive health index classification, a closed-loop hierarchical management system is completed, from early identification to severe alarms.

The embodiment of the present application provides a satellite signal carrier-to-noise ratio estimation method and device, which comprises: processing a satellite signal to obtain a digital intermediate frequency signal; acquiring a capture detection statistic quantity by performing pseudo code capture on the corresponding digital intermediate frequency signal according to a satellite number; if the capture detection statistic quantity is greater than a preset threshold, acquiring the digital intermediate frequency signal of a preset segment number and a preset data length; for each segment of the digital intermediate frequency signal, acquiring a capture correlation value by performing pseudo code capture, calculating a segmented correlation ratio value according to the capture correlation value, and obtaining an average correlation ratio value; calculating an absolute value of an autocorrelation function of a pseudo code sequence of the satellite signal, calculating a theoretical correlation ratio value based on a maximum value and a second maximum value, and acquiring a number of second maximum correlation values relative to the second maximum value; calculating an equivalent parameter according to the average correlation ratio value, the theoretical correlation ratio value and the number of second maximum correlation values; and calculating a carrier-to-noise ratio estimation value according to the equivalent parameter. The embodiment of the present application realizes fast and high-precision estimation of the carrier-to-noise ratio based on a small amount of data.

The present invention relates to the field of enterprise evaluation, and in particular to an enterprise evaluation method based on big data analysis, comprising: obtaining target enterprise data of a target evaluation enterprise; performing data segmentation on the target enterprise data of the target evaluation enterprise to obtain target evaluation subsets; determining a subset processing strategy according to the shard execution status of each target evaluation subset; in a first confidence analysis mode, determining a shard quality parameter of a type of evaluation subset according to the subset execution efficiency and a reference correlation index, determining an efficiency adjustment mode according to the shard quality parameter, and determining a confidence reference coefficient of each relevant analysis text; in a second confidence analysis mode, determining a subset execution coefficient of a second type of evaluation subset according to the reference correlation index and the topic correlation ratio, and determining a set setting mode and an effective evaluation strategy according to the subset execution coefficient; and determining a final evaluation score for each target evaluation data. The present invention improves the accuracy of the effectiveness evaluation of the acquired web page information.

The invention belongs to the technical field of data processing, is applied to the field of financial science and technology and the field of medical health, and discloses a data processing method, device and equipment and a storage medium, and the method comprises the steps: obtaining original business data, and storing the original business data in a preset distributed database; extracting basic data of the original business data and corresponding association proportion parameters from the preset distributed database based on a preset big data platform; based on a preset proportional relation, carrying out parallel calculation on the basic data and the associated proportional parameter to generate a data processing result; and writing the data processing result back to a specified data table of the preset distributed database. The invention provides a data processing method and device, equipment and a storage medium. The problems that in the prior art, a traditional processing method adopted during data processing is low in processing efficiency and the accuracy is difficult to guarantee are solved.