Method and system for optimizing power consumption information collection terminal fault prediction model based on gru
A fault prediction and acquisition terminal technology, applied in biological neural network models, data processing applications, character and pattern recognition, etc., can solve problems such as obstacles to normal power work, low efficiency, labor and material resources, and achieve rapid fault diagnosis Effects of predicting, saving costs, and reducing the impact of terminal failures
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0035] This embodiment provides a method based on GRU to optimize the terminal fault prediction model for power consumption information collection to solve the problem of negative losses when terminal faults occur during the construction of smart grids. By using big data analysis technology, an accurate and effective The advanced terminal failure prediction method conducts regular quantitative analysis and early warning of the status and failure possibility of the terminal in operation, and makes corresponding processing measures in advance to meet normal business needs and avoid field personnel having to quickly rush to each failure when the terminal fails Maintenance at the terminal location consumes a lot of manpower and material resources. This embodiment is based on the Bayesian network model and uses time-series historical information and related features to complete the fault prediction service for all terminals in operation in real time. Such as figure 1 with 9 As sh...
Embodiment 2
[0097] The purpose of this embodiment is to provide a power user credit evaluation system based on the dynamic combination of time-varying weights based on the method described in Embodiment 1, including:
[0098] The data acquisition module receives sample data of terminal failures and extracts attribute features related to terminal failures;
[0099] The data stable state prediction module uses the GRU gating mechanism to predict the stable state according to the attribute characteristics related to the stable state of the terminal data collection and the stable state of the transmission network data among the attribute characteristics;
[0100] The prediction model construction module, based on the attribute characteristics and the steady state prediction results, adopts the method of score search to construct and iteratively optimize the Bayesian network topology; based on the Bayesian network topology, perform parameter learning on the Bayesian network structure, Training...
Embodiment 3
[0103] The purpose of this embodiment is to provide a computing device based on the method described in Embodiment 1, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the program, accomplish:
[0104] Receive sample data of terminal failures, and extract attribute features related to terminal failures;
[0105] According to the attribute characteristics related to the stable state of the terminal data collection and the stable state of the transmission network data among the attribute characteristics, the GRU gating mechanism is used to predict the stable state;
[0106] Based on the attribute characteristics and steady state prediction results, the Bayesian network topology structure is constructed and iteratively optimized by using the scoring search method;
[0107] Based on the Bayesian network topology, learn the parameters of the Bayesian network structure and train the terminal fault...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More