Load Signal Gap Filling for Disruption-Resilient Demand Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electricity demand forecasting techniques for utility systems are adversely affected by network disruption events, such as transformer failures or scheduled maintenance, which lead to inaccurate short-term and long-term demand forecasts due to inconsistent archived data.
Innovation Solution
A system that preprocesses load signals by applying a first difference function to identify gaps caused by network disruptions, followed by spike-detection to pinpoint these gaps, and then fills them with projected load values using localized loadshape forecasting based on preceding continuous load values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If archived load signals are used for forecasting, then forecast coverage is achieved, but forecast accuracy deteriorates due to network disruption events
Solution Approach 1:
The patent applies preliminary action by preprocessing the load signals before they are used for forecasting. The system identifies gaps caused by network disruptions and fills them with projected values in advance, so that the forecasting models receive clean, continuous data without the adverse effects of disruptions.
Solution Approach 2:
The patent extracts and removes the harmful portions of the data by identifying gaps associated with network disruption events using spike-detection on difference signals. These problematic segments are separated from the continuous load values and replaced with synthesized projections, effectively taking out the harmful factors before forecasting.
2Measurement precision
If gap filling is performed using simple methods, then processing speed is improved, but forecast precision deteriorates
Solution Approach 1:
The patent applies dynamics by using different gap-filling strategies depending on the context. The system performs spike-detection to identify gaps, then uses localized loadshape forecasting that adapts to the specific characteristics of each gap and its surrounding data, rather than applying a single static filling method.
Solution Approach 2:
The system performs preliminary gap identification and filling before the main forecasting process. By detecting spikes in difference signals and filling gaps with projected values in advance, the preprocessing prepares high-precision data for subsequent forecasting operations without adding complexity during the actual forecasting phase.
Data Source
Figure 1A~1C
Figure 2A~2B
Figure 3
AI summary
The system receives a set of load signals from an archive that contains historic load information gathered at various locations throughout an electrical grid, which distributes electrical power for the utility system. Next, the system applies a first difference function to the set of load signals to produce a set of difference signals. The system then performs a spike-detection operation on the set of difference signals to identify pairs of positive-negative and negative-positive spikes, which identify gaps in the set of load signals associated with periods of network disruption. Next, the system modifies the set of load signals by filling in each identified gap with projected load values determined by performing a localized loadshape forecasting operation based on the continuous load values immediately preceding the identified gap. Finally, the system forecasts electricity demand for the utility system based on the modified set of load signals.