CNN-LSTM optimization model for power equipment monitoring and its visualization implementation

Aiming at the problem that traditional power equipment monitoring is difficult to meet the needs of modern power grid, this paper proposes a CNN-LSTM optimization model with attention mechanism, and constructs a situation awareness visualization system. CNN is used to extract local features of data, LSTM captures time series dependence, and attention mechanism weights important features. The hardware module is used to sample and transmit data, and the software module is developed to process visualization. Experiments show that the minimum voltage perception deviation of the model is 0.08 V, the perception accuracy is 96.52 %, the system data reception rate is over 98.0 %, the loss rate is 1.10 %, and the power perception deviation is 522 kW lower than that of a single model, which can provide support for the operation and maintenance of power grid equipment.