Grid State Prediction Using Neural Network and Analytical Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the state of power transmission grids, especially with increased distributed energy production and weather dependency, require high processing power and are inefficient, often resulting in delayed and costly solutions such as massive expansions of computing capacity.
Innovation Solution
Combining an artificial neural network with an analytical optimization method to speed up computation, allowing for timely predictions without the need for new hardware or complex software, using a central computer arrangement to process measurement data and control commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If distributed energy production and weather-dependent energy sources are integrated into the power transmission grid, then energy production diversity increases, but grid state prediction difficulty increases
Solution Approach 1:
The system performs preliminary actions by generating multiple possible grid state scenarios in advance, each representing different weather conditions and energy production levels. These pre-calculated scenarios are stored and can be quickly selected based on actual weather forecasts, avoiding the need for complex real-time predictions when the grid state must be determined.
Solution Approach 2:
The system dynamically adapts to changing conditions by selecting from pre-calculated scenarios based on current weather forecasts and actual grid measurements. Rather than using a static prediction model, the system flexibly chooses the most appropriate scenario and adjusts control measures accordingly, handling the variability introduced by weather-dependent energy sources.
2Measurement precision
If traditional analytical algorithms are used for grid state prediction, then prediction accuracy is maintained, but processing time increases
Solution Approach 1:
The system performs computationally intensive analytical calculations in advance, generating multiple grid state scenarios with different weather conditions and energy production levels. These pre-calculated scenarios store the results of complex DSPF computations, so when prediction is needed, the system simply selects from pre-computed results rather than performing new calculations, dramatically reducing processing time while maintaining accuracy.
Solution Approach 2:
The system creates copies of the grid model and performs parallel calculations for multiple scenarios. Each scenario is a copy of the base grid model with different weather and production parameters, allowing simultaneous pre-computation of multiple possible states without sequential processing delays.
3Productivity
If computing capacity is massively expanded to handle real-time predictions, then prediction speed increases, but cost increases
Solution Approach 1:
The system performs computationally intensive calculations in advance during off-peak times when computing resources are more readily available and less costly. By pre-generating multiple grid state scenarios before they are needed, the system avoids the need for massive real-time computing capacity, achieving fast predictions using minimal processing power at the moment of need.
Solution Approach 2:
The system uses periodic batch processing to pre-calculate scenarios at scheduled intervals rather than requiring continuous high-speed computation. This allows using standard computing resources periodically to generate scenarios that serve multiple prediction needs, reducing the overall computing resource requirements compared to continuous real-time processing.
Data Source
AI summary
A method predicts a grid state of an electrical power distribution grid, in which a central computer arrangement is used to receive measured values from measuring devices. A state estimation device is used to predict a future grid state, wherein the prediction of the future grid state is taken as a basis for ascertaining measures to guarantee stability of the power distribution grid. The prediction is made for multiple times within a predefined time window. A first prediction device is used to ascertain a prediction for a first portion of the multiple times on the basis of a voltage var control method, and in that a second prediction device is used to ascertain a prediction for a second portion of the multiple times on the basis of a neural network method.