Time-Variant Inverter Controller Using Dynamic Gain Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Time-varying systems, such as switching power electronic converter systems, pose challenges in stability control due to their time-dependent system matrices and sudden changes in eigenvalues, leading to unstable or borderline stable states, which existing control methods fail to address effectively.
Innovation Solution
A controller method that actively compensates for destabilizing time-variant system properties by incorporating variable system parameters into the controller law, ensuring stability and improved dynamics through state or output feedback mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a robust controller design is used for time-variant systems, then the system can be stabilized around an operating point, but the dynamics of the controller are reduced and control behavior deteriorates
Solution Approach 1:
The controller uses time-varying gain matrices K(t) that are updated in real-time based on the current switching state of the power electronic converter. This dynamic adaptation allows the controller to maintain optimal performance across different operating conditions without sacrificing stability or dynamics.
Solution Approach 2:
The controller parameters (gain matrices) are changed dynamically according to the switching state variables. By adjusting the controller parameters based on the current operating state, the system achieves both stability and fast response characteristics that a fixed robust controller cannot provide.
2Adaptability or versatility
If switching elements are used in power electronic systems, then system functionality is improved, but time-varying system properties cause sudden eigenvalue changes leading to instability
Solution Approach 1:
The controller implements state feedback using the current switching state variables to dynamically adjust the control action. This feedback mechanism ensures that the controller responds appropriately to switching transitions and maintains system stability despite the time-varying nature of the system.
Solution Approach 2:
The controller pre-calculates and applies appropriate control actions based on the current switching state before transitions occur. By anticipating switching events and preparing control responses in advance, the system prevents instability that would otherwise result from sudden eigenvalue changes.
3Ease of manufacture
If conventional control methods are used for time-variant systems, then implementation is simple, but they fail to address time-varying instabilities effectively
Solution Approach 1:
The controller automatically adapts to time-varying system conditions by using real-time switching state information to adjust its control parameters. This self-adjusting capability eliminates the need for complex external tuning or manual intervention while effectively addressing instability issues.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The invention relates to a method for controlling a time-variant system (1), in particular a switching power electronic inverter system having semiconductor switches, which can be described by a differential equation in the form of a state space model according to dx/dt = A x(t) + B u(t), in which x(t) is a vector having one or more state variables (Xi), A is the system matrix, B is the control matrix and u(t) is a vector having at least one input variable (Ui) of the system. At least one of the state variables (Xi) in the system (1) is multiplicatively linked with a variable system parameter (p(t)). At least one of the state variables or one of the output variables of the system (1) is coupled back negatively via a controller (2) having a control matrix K to at least one input variable (Ui), wherein at least one of the state variables (Xi) or output variables (yi) coupled back via the controller (2) is multiplied by the variable system parameter (p(t)) contained in an element of the controller matrix K or a value of the variable system parameter (p(t)) averaged in sections. Furthermore, the invention relates to a data processing program for carrying out the method, to a data storage medium having such a data processing program and to a data processing system, on which such data processing program is loaded.