ANN Control for DC Microgrid Voltage Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current control strategies for DC microgrids lack an effective artificial neural network (ANN) based control for DC/DC converters, which is essential for maintaining voltage stability and power sharing among distributed generation units in standalone operations.
Innovation Solution
Integration of ANN control with the droop mechanism for DC/DC converters to regulate voltage stability and power sharing in standalone DC microgrids, utilizing a state-space model and training the ANN controller with approximate dynamic programming to optimize control actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If traditional PI control is used for DC/DC converters, then the control structure is simple, but the response time is slow and voltage stability is poor under dynamic conditions
Solution Approach 1:
The patent replaces the traditional PI control mechanism with an artificial neural network (ANN) based control system. The ANN controller uses trained weight matrices and bias vectors to compute control signals, substituting the conventional proportional-integral calculation mechanism with a neural network inference mechanism that provides faster response and better adaptability to dynamic conditions
Solution Approach 2:
The patent transforms the control approach by changing from fixed gain parameters in PI control to adaptive parameters learned through training data. The ANN controller's weights and biases are optimized through training processes to achieve superior voltage regulation performance under varying load and voltage conditions
2Reliability
If droop method is used for power sharing control, then parallel DG units can share power, but bus voltage stability deteriorates
Solution Approach 1:
The patent merges the droop control mechanism with ANN-based voltage regulation control. The droop control handles power sharing among parallel DG units by adjusting output based on local measurements, while the ANN controller simultaneously maintains bus voltage stability by compensating for voltage deviations. This combined approach resolves the conflict between power sharing and voltage stability
Solution Approach 2:
The ANN controller acts as an intermediary layer between the droop control and the DC/DC converter. It receives the control signal from droop control and adds voltage regulation compensation, mediating between the power sharing requirement and voltage stability requirement to achieve both objectives simultaneously
3Stability of the object's composition
If ANN control is applied to DC/DC converters, then voltage stability and response time improve, but control system complexity increases
Solution Approach 1:
The patent applies preliminary action by training the ANN controller offline using historical or simulated data before actual operation. The training process pre-optimizes the weight matrices and bias vectors for various operating conditions, so that during real-time control, the ANN can directly apply pre-learned parameters without requiring complex online computation or adaptation, reducing real-time system complexity
Data Source
AI summary
An example method for controlling a DC/DC converter or a standalone DC microgrid comprises an artificial neural network (ANN) based control method integrated with droop control. The ANN is trained to implement optimal control based on approximate dynamic programming. In one example, Levenberg-Marquardt (LM) algorithm is used to train the ANN, where the Jacobian matrix needed by LM algorithm is calculated via a Forward Accumulation Through Time algorithm. The ANN performance is evaluated by using power converter average and switching models. Performance evaluation shows that a well-trained ANN controller has a strong ability to maintain voltage stability of a standalone DC microgrid and manage the power sharing among the parallel distributed generation units. Even in dynamic and power converter switching environments, the ANN controller shows an ability to trace rapidly changing reference commands and tolerate system disturbances, and operate the DC/DC converter or the microgrid in standalone conditions.