Microgrid MPC Framework Using Object-Oriented Graph Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Model Predictive Control (MPC) problems in microgrid operation are notoriously difficult due to error-prone logging, updating, and bootstrapping processes, requiring significant overhead work and being challenging to manage effectively in complex power systems.
Innovation Solution
A framework using object-oriented graph models with class-inheritance principles dynamically constructs MPC problems, defining constraints that represent the dynamical models of power and energy assets, such as generators and storage systems, to optimize power dispatch and reserve allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional MPC problems are implemented in microgrid operation, then power dispatch optimization can be achieved, but the system becomes error-prone and requires significant overhead work due to complex logging, updating, and bootstrapping processes
Solution Approach 1:
The patent creates a simplified copy or representation of the complex MPC problem through a standardized framework that captures essential dynamics without requiring full implementation complexity. This framework copy enables optimization while avoiding the error-prone aspects of traditional MPC logging and bootstrapping processes.
Solution Approach 2:
The patent develops a universal framework that can handle multiple microgrid operations and scenarios through standardized classes and interfaces. This multi-functional approach allows the same framework to manage different asset types, control strategies, and operational modes without requiring separate complex implementations for each case.
2Manufacturing precision
If detailed dynamical models of power assets are constructed for accurate control, then control precision improves, but the system complexity and maintenance difficulty increase significantly
Solution Approach 1:
The patent segments the complex dynamical modeling task into modular, reusable components represented as software classes. Each class encapsulates specific asset behaviors and constraints, allowing precise control modeling to be built from standardized building blocks rather than constructed entirely from scratch, thereby reducing maintenance complexity while preserving control precision.
Solution Approach 2:
The patent enables precise control by allowing dynamic parameter changes within the standardized framework. The system can adjust model parameters, constraints, and operational characteristics without changing the underlying framework structure, maintaining control precision while simplifying the complexity of model construction and maintenance.
3Reliability
If comprehensive constraints are formulated to define dynamical models of assets, then reliability of power system operation improves, but the overhead work and difficulty of managing MPC problems increases
Solution Approach 1:
The patent implements self-service mechanisms where the standardized framework automatically handles constraint formulation, validation, and enforcement. The system serves itself by providing built-in methods for managing operational constraints, reducing the manual overhead work required to maintain reliability while improving ease of operation through automated constraint handling.
Data Source
AI summary
A simulator models an energy and power system. The simulator allows a user to manipulate digital representations of nodes, which represent one or more components of power and energy assets. The simulator also allows a user to manipulate digital representations of edges, which connect the nodes, to form a power and energy network. A plurality of object classes correspond to the nodes and edges. The object classes comprise class inheritance structures so that constraints of a parent class are retained by one or more child classes. The interface on the simulator allows a user to model a power and energy system by allowing the user to connect the nodes with edges to construct a power and energy system, wherein the object classes for the nodes, when executed by the simulator, implement one or more dynamical models to model the power and energy assets.