Hierarchical Model Predictive Voltage Control for DER Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High levels of distributed energy resources (DER) in power distribution systems cause voltage rises and rapid voltage variations, which traditional voltage regulation techniques are inadequate to handle, due to intermittent and uncertain power output from DER systems, and incomplete information about the distribution system topology.
Innovation Solution
The implementation of model predictive voltage and VAR controls, coordinated with autonomous reactive power elements such as smart inverters and switched capacitor banks, using a hierarchical control architecture that constructs a linearized model of the power distribution system and optimizes control commands through mixed integer programming to manage voltage and reactive power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional voltage regulation techniques are used, then the system is simple to operate, but they are inadequate to handle rapid voltage variations caused by DER power output changes
Solution Approach 1:
The patent implements a dynamic model predictive control system that continuously updates control decisions based on real-time system state and predicted DER power output. The controller adapts to rapid voltage variations by recalculating optimal control actions at each time step, transforming the static traditional regulation approach into a dynamic responsive system that can handle intermittent DER power changes effectively.
Solution Approach 2:
The control system performs preliminary predictions of DER power output and voltage variations using forecast models before actual changes occur. By anticipating future system states and pre-calculating optimal control actions, the system proactively mitigates voltage deviations rather than reacting after problems arise, improving adaptability while maintaining manageable complexity through structured prediction frameworks.
2Manufacturing precision
If detailed power flow models and power flow solutions are used, then voltage control precision is improved, but the system requires complete information about distribution system topology which is often unknown
Solution Approach 1:
The patent introduces a linearized equivalent circuit model as an intermediary representation that captures essential voltage-control-relevant characteristics of the distribution system without requiring complete topology information. This simplified model acts as a mediator between the available limited measurements and the voltage control objectives, enabling precise control decisions to be made based on partial system information through the surrogate linearized model.
Solution Approach 2:
The system transforms the complex non-linear power flow equations into a linearized model with simplified parameters that are sufficient for voltage control purposes. By changing the mathematical representation from detailed non-linear power flow models to linearized approximations, the system achieves adequate voltage control precision while operating with incomplete topology information, as the linearized parameters can be estimated from available measurements.
3Reliability
If model predictive control with multiple predictions and mixed integer programming solvers is implemented, then voltage and VAR regulation effectiveness is improved, but computational complexity increases
Solution Approach 1:
The patent segments the control problem into hierarchical layers and temporal stages, separating fast-acting voltage control decisions from slower reactive power optimization. The model predictive controller divides the prediction horizon into multiple stages, solving simplified sub-problems at each stage rather than optimizing all variables simultaneously. This segmentation reduces computational complexity while maintaining regulation effectiveness by addressing different control objectives at appropriate time scales and detail levels.
Solution Approach 2:
The system dynamically adjusts the complexity of optimization based on system conditions, changing parameters such as prediction horizon length, number of scenarios considered, and optimization variables included. Under normal conditions, simplified models with fewer variables are used for faster computation, while during critical voltage events, more comprehensive models are activated. This adaptive parameter changing allows the system to maintain high regulation effectiveness while managing computational complexity through conditional model selection.
Data Source
AI summary
Apparatuses, method and systems featuring model predictive voltage and VAR controls with coordination with and optional optimization of autonomous reactive power control such as autonomous distributed energy resource and/or autonomous switched capacitor banks One embodiment includes an electronic control system structured to construct a linearized model of the power distribution system including a plurality of predetermined nodes, operate a model predictive controller to identify optimized control commands using an objective function defined over a plurality of future scenarios over a look ahead time horizon and a plurality of constraints, and transmit the identified control commands to control operation of at least the voltage regulators and the switched capacitor banks.