AI Circuit Netlist Generation with Closed-Loop Simulation Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional circuit design processes are lengthy, resource-intensive, and lack intelligent automation, particularly for complex designs, leading to slow development cycles and high costs.

Innovation Solution

An AI-based automated circuit generation method using a multi-layer Transformer structure to capture dependency relationships, simulate, and correct circuit netlists, incorporating a natural language parsing module and closed-loop simulation correction mechanism to improve design efficiency and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual circuit design with EDA tools is used, then design flexibility and control are maintained, but design cycle length and resource consumption increase significantly

Engineering Contradiction:
Improvedesign efficiencyVSAvoiddesign cycle
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical design process with an AI-based automated system. The Transformer model automatically generates circuit netlists from natural language descriptions, substituting the engineer's manual EDA tool operations with intelligent algorithms that capture design intent and directly produce implementable circuit configurations.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service circuit design by allowing the AI model to autonomously interpret design requirements, generate netlists, perform simulations, and correct errors without continuous human intervention. The closed-loop simulation-correction mechanism allows the system to self-validate and self-improve the generated circuits.

Inventive Principle:
Principle #25Self-service

2Extent of automation

If automated EDA tools are used, then some design tasks are accelerated, but complex analog circuit design remains difficult and lacks intelligent automation

Engineering Contradiction:
Improveautomation levelVSAvoiddesign flexibility
Core Design Contradiction:
Extent of automationVSAdaptability or versatility

Solution Approach 1:

The patent introduces natural language as an intermediary between the designer's intent and the automated EDA tools. The Transformer model translates human-readable design descriptions into structured netlists, bridging the gap between high-level design concepts and low-level implementation details that traditional automated tools cannot handle flexibly.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the input parameter format from traditional schematic capture or HDL code to natural language descriptions. This parameter transformation allows the AI model to understand design intent more naturally and generate circuits that better reflect the designer's goals while maintaining full automation capability.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If multiple simulation iterations are performed for error checking, then circuit reliability improves, but computing resource overhead and time consumption increase

Engineering Contradiction:
Improvecircuit correctnessVSAvoidcomputing resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent performs preliminary error detection and correction by integrating simulation validation into the netlist generation process itself. The system proactively identifies potential errors in the generated netlists and corrects them through automated refinement iterations before final deployment, preventing the need for extensive post-generation debugging.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a closed-loop feedback mechanism where simulation results are automatically fed back to the Transformer model. The model learns from simulation outcomes and adjusts subsequent netlist generations accordingly, progressively improving circuit correctness while reducing the number of iterations needed through intelligent adaptation.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS12462083B1AI-based automated circuit generation method
Publication Date: 2025.11.04 AICHIPSKY MICROELECTRONICS CO LTD
  • US12462083B1 patent drawing

AI summary

Disclosed is an AI-based automated circuit generation method. The method includes obtaining design metrics; capturing a dependency relationship among design metrics by using an attention mechanism, extracting features by using a multi-layer Transformer structure to obtain an overall structure and a high-level feature representation of a circuit, and matching the high-level feature representation with a standard circuit netlist template based on the overall structure of the circuit to obtain a current circuit netlist; simulating the current circuit netlist to obtain a simulation report, and extracting error information from the simulation report; and correcting the current circuit netlist based on the error information, simulating a corrected circuit netlist again, looping until there are no errors in the simulation report, and outputting a final circuit netlist. The present invention solves the problems that the existing method has limitations and is difficult to quickly, automatically and flexibly design a complex circuit.