Building Power Forecasting with Energy Drift and Hybrid Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power consumption forecasting systems face challenges in accurately predicting long-term building energy consumption due to issues with handling outliers, missing values, and dynamic input features, which affect the efficiency of energy management systems.
Innovation Solution
A method and system that preprocesses historical data to remove outliers and impute missing values, calculates energy-temperature correlation and drift, and employs a hybrid model combining auto-regressive features with support vector regression and artificial neural networks for forecasting power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If conventional forecasting methods are used for long-term building energy consumption, then the forecast horizon can be extended, but the prediction accuracy deteriorates due to outliers and missing values
Solution Approach 1:
The patent applies preliminary action by preprocessing the historical power consumption data before forecasting. This includes removing outliers, imputing missing values, and normalizing data in advance. By preparing the data beforehand, the system maintains high prediction accuracy even when extending the forecast horizon to long-term predictions.
Solution Approach 2:
The patent replaces conventional mechanical forecasting approaches with a hybrid intelligent system combining Support Vector Regression (SVR) and Artificial Neural Networks (ANN). This substitution enables the system to handle non-linear relationships and complex patterns in energy consumption data, maintaining accuracy across extended forecast horizons where traditional methods fail.
2Device complexity
If simple forecasting models are used, then the system complexity is reduced, but the ability to handle dynamic input features and correlations deteriorates
Solution Approach 1:
The patent applies the composite principle by creating a hybrid forecasting model that combines SVR and ANN algorithms. This composite approach leverages the strengths of both methods: SVR provides robustness against outliers and good generalization, while ANN captures complex non-linear patterns. The hybrid model achieves high adaptability to dynamic input features without requiring excessive system complexity.
Solution Approach 2:
The patent dynamically adjusts model parameters based on the characteristics of the input data. The system calculates energy-temperature correlation and energy drift, then selects and configures model parameters accordingly. This parameter adaptation enables the system to handle diverse and changing input features effectively.
3Device complexity
If energy drift and temperature correlation are not considered, then the model selection is simplified, but the forecasting accuracy for different scenarios deteriorates
Solution Approach 1:
The patent introduces dynamics by making the model selection process adaptive rather than static. The system calculates energy drift (change in average power consumption over time) and energy-temperature correlation, then dynamically selects the most appropriate forecasting model based on these characteristics. This dynamic approach ensures high forecasting accuracy across different scenarios without requiring a single complex model for all cases.
Data Source
AI summary
Power consumption forecasting plays a key role in the efficient operation of a building energy management system to assess energy demands of building, and at the same time, help electrical utilities in planning their supply operations. However, no state-of-the-arts are available for forecasting medium-term or long-term power consumption of the buildings. This disclosure relates to a method and system for forecasting a power consumption of buildings for a scalable forecast horizon. The system is configured to pre-process to deal with outliers/missing values, followed by synchronization of smart meter data with other sensory data. An energy-temperature correlation is calculated to estimate an energy drift using historical power consumptions. Further, in feature derivation stage, additional features necessary for the forecast are derived. The system is to be employed for modeling the building load consumption that depends on the time horizon of forecasting and the granularity of the data.