Solar Irradiance Forecasting Using Accessible Weather Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current solar irradiance forecasting methods, including statistical and NWP models, face challenges in providing reliable forecasts due to the untrustworthiness of predictor variables and limited accessibility of NWP data, which affects the accuracy of energy generation predictions for photovoltaic systems.
Innovation Solution
A method utilizing a sequence-to-sequence neural network with encoder and decoder networks to forecast solar irradiance by performing auto-regression and regression analyses on freely accessible weather data, including measured and forecasted values of weather parameters, to estimate energy generation in a predetermined geographic area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If NWP models are used for solar irradiance forecasting, then forecast accuracy can be improved, but data accessibility deteriorates
Solution Approach 1:
The patent uses statistical models as intermediary tools that process freely accessible weather data (temperature, humidity, pressure) to generate solar irradiance forecasts. These statistical models act as mediators between readily available weather information and the need for accurate solar forecasts, avoiding direct dependence on restricted NWP data while maintaining forecast quality
Solution Approach 2:
The patent creates simplified copies of NWP model functionality using statistical relationships derived from historical weather data. Instead of directly using complex NWP models, the system replicates their forecasting capability through statistical regression models that use easily accessible weather parameters, making the forecasting process both accurate and accessible
2Ease of operation
If statistical models are used for solar irradiance forecasting, then data accessibility is improved, but forecast reliability deteriorates
Solution Approach 1:
The patent merges multiple statistical approaches (auto-regressive models, regression analysis, time series analysis) into a hybrid forecasting system. By combining different statistical techniques and integrating multiple weather parameters (temperature, humidity, pressure, cloud cover), the system achieves reliable forecasts using only freely accessible data, overcoming the limitations of individual statistical methods
Solution Approach 2:
The patent transforms the forecasting approach by changing from using direct solar irradiance predictors (which are unreliable) to using correlated weather parameters (temperature, humidity, pressure) as proxy variables. This parameter substitution maintains forecast reliability while using only accessible data sources
3Productivity
If auto-regression is applied to sequential solar irradiance data, then forecast capability is improved, but predictor reliability deteriorates
Solution Approach 1:
The patent inverts the traditional forecasting approach by not using solar irradiance data itself as the predictor variable. Instead, it uses other weather parameters (temperature, humidity, pressure) that have reliable forecast capability and are correlated with solar irradiance. This inversion solves the problem of unreliable solar irradiance predictors while maintaining strong forecast capability
Data Source
AI summary
Estimating energy generated by a solar system in a predetermined geographic area comprises, at each predetermined time instant: retrieving measured values of at least one weather parameter and of solar irradiance in the geographic area, the values related to a time slot before the predetermined time instant; performing auto-regression analysis of the measured values; estimating, based on the auto-regression analysis, a relationship between the at least one weather parameter and the solar irradiance; retrieving forecasted values of the at least one weather parameter in the geographic area, the forecasted values being forecasted for a second time slot after the predetermined time instant; performing regression analysis of the relationship between the at least one weather parameter and the solar irradiance of the forecasted values; forecasting solar irradiance in the second time slot based on the regression analysis, and estimating energy generated by the solar system in the second time slot.