Electric Motor FPGA Simulation With Matrix Routing Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing real-time simulation methods for electric motors using FPGAs suffer from poor resource utilization and high latency due to fixed configurations and inefficient matrix operation handling, particularly when dealing with varying motor characteristics and different discretization methods.
Innovation Solution
Implementing matrix operations in separate matrix routing and calculation blocks, with alternating input signal provision and output storage in routing blocks, and generating a bitstream from a graphical model, allowing flexible configuration and efficient use of FPGA resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a fixed predefined configuration is used for FPGA simulation, then resource utilization is poor and latency is relatively long, but implementation is simpler
Solution Approach 1:
The patent segments the simulation configuration into modular components: a library of parameter sets corresponding to different motor types, and a template-based structural framework. This allows the system to select and instantiate only the necessary parameters for each specific simulation task, improving resource utilization while maintaining manageable complexity through systematic organization.
Solution Approach 2:
The patent implements dynamic configurability by allowing the FPGA configuration to be adapted at runtime based on the selected motor type and discretization method. The system transitions from a static fixed configuration to a dynamic one where parameters can be changed without redesigning the entire system, enabling fast reconfiguration through parameter instantiation rather than structural redesign.
2Adaptability or versatility
If model-based bitstream generation is used for specific motors, then adaptability improves, but resource utilization decreases and latency increases
Solution Approach 1:
The patent creates a universal simulation platform that can handle multiple motor types and discretization methods through a common template structure. Instead of creating separate model-based configurations for each motor type, the system uses a single adaptable template that instantiates different parameters, achieving multi-functionality while maintaining efficient resource usage and low latency.
Solution Approach 2:
The patent achieves adaptability through parameter instantiation rather than structural modification. By changing numerical parameters and configuration values within the fixed template framework, the system can adapt to different motor characteristics and discretization methods without triggering the resource utilization and latency penalties associated with full model-based regeneration.
3Productivity
If matrix operations are implemented with separate routing and calculation blocks, then resource utilization improves and latency reduces, but implementation complexity increases
Solution Approach 1:
The patent segments matrix operations into distinct routing blocks and calculation blocks, allowing independent optimization of each function. The routing blocks handle signal distribution and timing, while calculation blocks perform computational operations. This segmentation improves resource utilization by preventing bottlenecks and reduces latency through parallel processing, while the modular nature keeps implementation complexity manageable.
Solution Approach 2:
The patent ensures continuous useful action by designing the routing blocks to continuously supply input signals to calculation blocks without idle waiting periods. The alternating provision and storage mechanism eliminates gaps in the processing pipeline, keeping FPGA resources continuously utilized and maximizing calculation efficiency without requiring overly complex synchronization mechanisms.
Data Source
AI summary
A method for real-time simulation of an electric motor includes: modeling, by a computer system, the electrical motor in a graphical model including at least two blocks connected by signals. The graphical model includes at least two matrix operations which produce a multidimensional output signal from at least one multidimensional input signal. The computer system includes an operator computer and a real-time computer connected thereto, wherein the operator computer executes a graphical modeling environment, and the real-time computer has at least one programmable logic device.