Full-Field DNI Prediction for Spatial Heliostat Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DNI prediction methods for tower solar thermal power stations often result in significant unnecessary operations due to uniform heliostat control, affecting power generation efficiency, as they predict average DNI rather than individual heliostat field locations.
Innovation Solution
A full-field refined DNI prediction method using at least two all-sky imagers to determine cloud location, calculate cloud velocity and thickness, and predict DNI values by combining machine learning with image processing algorithms to accurately identify and track cloud movement, enabling targeted heliostat operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If uniform heliostat control is implemented based on average DNI prediction, then heat absorber protection is improved, but power generation efficiency deteriorates due to unnecessary operations
Solution Approach 1:
The patent divides the heliostat field into multiple spatial zones (e.g., first heliostat region and second heliostat region) based on cloud coverage predictions. Different control strategies are applied to different zones: heliostats in cloud-affected zones are stopped to protect the heat absorber, while heliostats in clear zones continue operating to maintain power generation. This spatial segmentation resolves the contradiction by protecting the heat absorber without unnecessarily stopping all heliostats.
Solution Approach 2:
The patent implements localized control strategies for different regions of the heliostat field based on predicted cloud coverage and DNI changes at specific locations. Instead of uniform control, the system adjusts heliostat operation according to local conditions - stopping heliostats only in areas where clouds are predicted to arrive, while maintaining operation in areas without cloud coverage. This localizes the protective action to where it is needed most.
2Device complexity
If average DNI prediction is used for full field control, then system simplicity is maintained, but measurement precision deteriorates for individual heliostat locations
Solution Approach 1:
The patent segments the prediction domain from the control domain. The all-sky imager captures full-field cloud coverage information, which is then processed to predict DNI changes at specific heliostat locations. This segmentation allows the system to maintain relatively simple imaging hardware while achieving precise location-specific predictions through computational methods.
Solution Approach 2:
The patent introduces an intermediary processing layer between the all-sky imager and the heliostat control system. This intermediary uses machine learning models and spatial mapping algorithms to translate general cloud coverage images into specific DNI predictions for individual heliostat locations. The intermediary layer adds the necessary precision without requiring complex direct measurement systems at each heliostat.
3Reliability
If early heliostat shutdown is implemented to prevent cloud damage, then heat absorber protection is improved, but energy loss increases due to premature stopping
Solution Approach 1:
The patent uses all-sky imagers to capture cloud movement information in advance and predicts future cloud coverage and DNI changes at heliostat locations. This preliminary prediction allows the system to take protective action only when and where clouds are actually expected to arrive, rather than shutting down early based on premature or overly conservative assumptions. The preliminary action is based on accurate predictive modeling rather than precautionary shutdown.
Solution Approach 2:
The patent implements a feedback loop where real-time cloud coverage data from all-sky imagers is continuously processed to update DNI predictions. The system monitors actual cloud movement and adjusts heliostat control decisions based on the predicted trajectory and intensity of cloud coverage. This feedback mechanism ensures that heliostats are stopped only when necessary, preventing both premature shutdown and insufficient protective action.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enhances power generation efficiency by allowing precise control of heliostats based on actual DNI changes, reducing unnecessary operations and protecting heat absorbers, thus improving overall power output.
Implementation Method 1
cloud identification: accurately recognizing a cloud cluster in an image of the all-sky imagers; cloud's image velocity calculation: calculating a velocity and direction of each cloud pixel point using Farneback algorithm
Implementation Method 2
shade location prediction: predicting the shade location after a period of time by calculating changes in coordinates of a shade point at different time periods
Data Source
AI summary
The present invention discloses a full-field refined DNI prediction method. At least two total-sky imagers are used to determine the actual position of a cloud, and then a shadow position is determined on the basis of a solar angle; and the thickness of the cloud is determined by means of the imaging brightness of the cloud, and then a DNI value is predicted. The method specifically comprises the following steps: performing cloud identification, cloud image speed calculation, cloud actual-position calculation, cloud/shadow actual-speed calculation, shadow position prediction, cloud thickness extraction, DNI mapping, and DNI prediction. In the method, at least two total-sky imagers or pinhole cameras are used to perform a DNI prediction operation, and a DNI change at each specific position in a heliostat field can be accurately predicted, such that the power generation efficiency is improved.