5 results about "Recurrent neural network" patented technology

The invention provides an employee incentive management method, system and device and a storage medium, and the method comprises the steps: screening out brokers meeting a preset condition from a database; obtaining broker features according to brokers; obtaining first broker features, obtaining a total score of the first broker features, and obtaining a plurality of first brokers according to the total score; obtaining multiple second broker features, constructing a recurrent neural network time sequence model according to the logistic regression model and a recurrent neural network algorithm, obtaining the predicted filing trading volume of the brokers, and determining multiple second brokers according to the predicted filing trading volume; obtaining third broker characteristics, sorting brokers according to the third broker characteristics, selecting preset brokers, carrying out sorting according to the filing trading volume of the preset brokers, and obtaining the third brokers according to the sorting result; and respectively exciting the first broker, the second broker and the third broker according to a set rule. The working enthusiasm of employees can be continuously stimulated.

The invention provides a commercial building demand response method and system based on a dynamic game, and relates to the technical field of power demand response. According to the dynamic game-based commercial building demand response strategy provided by the invention, the commercial building load can be flexibly and effectively managed, and the effect of peak load shifting is achieved, so that the balance of power supply and demand is promoted, the safe and reliable operation of a power system is ensured, and the power consumption cost of the commercial building is reduced. The deep recurrent neural network is used for predicting the load and the photovoltaic output power of the commercial building, the uncertainty of the power supply side and the power demand side is overcome, and the obtained demand response strategy is more accurate and effective.

The invention discloses a zinc flotation working condition identification method based on long-time-history depth features, and the method comprises the following steps: firstly, employing a 3D convolutional network as a basic network, and simulating an optical flow network through a knowledge distillation method by employing a part of structure of an RGB flow network, so that the RGB flow network can learn the motion information of the optical flow without extracting the optical flow during testing; segmenting a single video, performing frame-level feature extraction on each segment by using a trained RGB flow network, and inputting the extracted frame-level features of each segment into an LSTM network to further extract video-level global spatial-temporal features; and supplementing a 2D convolutional network for the network to extract supplemented appearance features, splicing the global spatial-temporal features and the enhanced appearance features together, and inputting the spliced features into a multi-layer perceptron to carry out final working condition identification. According to the invention, the advantages of the convolutional neural network and the recurrent neural network are combined, and the zinc flotation working condition can be quickly and accurately identified, so that dosing is effectively guided.