Design Optimisation/Simulation TRIZ Cases
- Source Mask Lens Optimization for Lithography
Segmented optimization modules accelerate source mask lens convergence, expanding process windows beyond traditional limits.
- Photonic Neural Network Reconfigurable Waveguide Modulation
Programmable electrodes modulate light in a planar waveguide layer, resolving chip-scale integration limits while delivering 10,000 TOPS at under 5 watts.
- Antenna Offset Learning for Perception-Aided Wireless Communication
Dynamic segmentation mask and trajectory tracking predict antenna position, resolving measurement precision versus device complexity contradictions.
- Simulation Program Synchronization via Integral Error Monitoring
A synchronization method approximates variables between simulation programs operating at different frequencies to maintain correct physical coupling.
- Discrete Dislocation Dynamics Model for Titanium Alloy Grain Refinement Prediction
A discrete dislocation dynamics model predicts grain refinement in titanium alloy machined surfaces during ultra-precision cutting.
- Thermally Aware Design Automation for Semiconductor Chips
A thermally aware design automation suite modifies semiconductor chip structures to equalize temperature variations.
- ML-Based EDA Layout Design Optimization
Convolutional neural networks and reinforcement learning adapt design flows to meet stringent specifications at miniaturized technology nodes.
- Flexible Cable Simulation via Control Point Segmentation
Segmenting the external cable into fixed control points reduces computational time and memory usage for robot retraction systems.
- Transaction Specific Stage-Wise Log Data Capture for Verification
A prediction unit generates expected output packets to compare against actual results from a design under verification.
- Deep Reservoir Fracture Prediction Using Thickness Optimization
A reservoir fracture prediction method calculates structural fracture linear density using a thickness-based optimization formula.
- Adaptive Filter Optimization for Electromagnetic Device Design
Adaptive filter algorithms optimize electromagnetic device structural parameters through iterative physics simulation feedback.
- Automated Drawing Information Mapping for Semantic Digital Twins
A named entity recognition model maps extracted drawing information to industrial standard classes.
- Acoustic Modeling via Finite Element Partitioning
Partition finite element model elements into sets to calculate spatial distances between composite centroids for acoustic modeling.
- Predicting DRC Violations in Placement Layouts
ML models analyze placement data to detect routing congestion risks, reducing chip area and turnaround time.
- Multi-Phase Simulation Initialization for Autonomous Vehicles
Segmented warmup and blending phases minimize simulation drift from actual road geometry.
- Apparel Patternmaking Software Anchor Point Resizing
Patternmaking software adjusts digital sewing patterns using anchor points and connecting elements on a grid to create customized designs.
- Anisotropic Viscosity Modeling for Fiber-Reinforced Thermoplastic Molding
A processing module generates an anisotropic viscosity distribution based on fiber orientation to control composite molding resin flow.
- GNSS Navigation Data Generation via Configurable Templates
A GNSS simulation system generates navigation data using configurable configuration files and programming scripts to define message formats.
- Just-in-time Compilation for Discrete Event Simulation Speed
A just-in-time compiler generates high-speed code blocks within discrete event simulations.
- Work Plan Verification Device for Crane Load Movement
A work plan verification device calculates load factors and displays movement feasibility data on a monitor to guide crane operators.
- Topology Optimization for Additive Manufacturing Residual Stress
Iterative topology optimization accounts for additive manufacturing residual stresses and deformations, reducing warping and post-processing costs.
- Semiconductor Design Optimization via AI Physical Property Prediction
A semiconductor design optimization system predicts physical properties using image data from design drawings to generate optimized layouts.
- Spatial Estimation Model Training via Wavefront Plane Integration
Arithmetic system updates spatial estimation models using wavefront plane integration to sample emission waves.
- Hierarchical Block Diagram Model for Multi-Domain Systems
Hierarchical block diagram model reduces graphical network complexity by connecting multi-domain components through intermediary signal routing blocks.
- AI Force Field for Molecular Dynamics Computation
A hybrid force field combines machine learning with coarse-grained models to reproduce global structural changes efficiently.
- AI Neural Network for Water Quality Biotoxicity Prediction
An artificial intelligence neural network predicts water quality biotoxicity by analyzing simulated water and fish response parameters.
- Discrete Fracture Network Realization Selection for Reservoir Simulation
Computer method selects discrete fracture network realizations using geological parameters to generate reservoir models.
- Physics Engine Momentum Conservation for Articulated Body Stability
Solver separates external and internal forces to correct momentum drifts, stabilizing articulated body simulations.
- Hybrid TPFA-MFD Projection Embedded Discrete Fracture Model
A hybrid TPFA-MFD projection embedded discrete fracture model calculates numerical fluxes across K-orthogonal and non-K-orthogonal grids.
- Fabricability Verification of Segmented Designs Using Paintbrush Patterns
A design verification system processes segmented designs using paintbrush patterns to calculate positive and negative loss values.
- Machine Learning Stiffness Matrix Estimation
A graph convolutional neural network estimates element stiffness matrices from node feature vectors.
- Sheet-Metal-Forming Simulation Parameter Compliance Display
A simulation software interface displays active and standard parameter values with visual compliance indicators.
- Genetic Algorithm Cell-Fuse Placement for Battery Balance
A computer system determines optimal cell-fuse pair placement within battery modules using genetic algorithm breeding operations.
- Welding Simulator Virtual Environment Training
A welding simulator generates virtual environments using CAD data to track weld paths and display simulated results.
- Reinforcement Learning for Resin Pellet State Optimization
A machine learning device determines optimal resin kneading conditions by calculating rewards based on state variables.
- Navier-Stokes Simulation via Constraint Decoupling
Decoupling constraints from Navier-Stokes equations via a commuting matrix eliminates inverse calculations, reducing design time while maintaining accuracy.
- Burst Clock Control for Independent Core Testing in Mesh Architecture
Clock gating devices segment the mesh architecture into independent domains, resolving the trade-off between uniform distribution and asynchronous core testing.
- Simulation Primitives Dependency Management
Unified primitives and a data structure represent dependencies between simulations, reducing iterative process time and errors.
- Reinforcement Learning for Automated Object Placement Optimization
A reinforcement learning apparatus configures a simulation environment to determine optimal target object positions based on design data.
- Springback Cause Identification in Press-Formed Parts
Numerical analysis of physical property values identifies springback causes, reducing trial-and-error development time.
- SPICE Resistor Model Segmentation for Parasitic Resistance
Segmenting bulk, terminal, and contact resistors with temperature coefficients corrects simulation accuracy errors from unaccounted parasitic resistance.
- Incremental NoC Topology Synthesis for Deadlock-Free Routing
A tool generates network-on-chip topologies incrementally using machine learning models and mathematical optimization.
- Recursive Hierarchical Particle Swarm Optimization for Global Optima
Recursive hierarchical particle swarm optimization escapes local optima by segmenting solution spaces and focusing on promising branches.
- Single Loop Hybrid Search for Hydrant Fire Flow Design
A single-loop algorithm determines hydrant fire flow capacity by coupling search logic within the hydraulic solver.
- Neural Network Conservation Law Discovery via Contrastive Learning
Contrastive learning automates conservation law discovery, replacing expensive symbolic regression with scalable neural network training.
- Digital Twin Sensor Validation via Virtual-Physical Deviation
Virtual sensors compare readings against physical measurements to detect deviations and trigger recalibration for reliable process control.
- Layout-Dependent Anisotropic Modeling for BEOL Stress
Replaces isotropic smear methods with layout-dependent anisotropic modeling to resolve unreliable stress estimation in integrated circuit design.
- Hybrid-Equivalence Partition Computing for Power Electronics
A hybrid-equivalence method partitions power electronics circuits into sub-circuits to enable parallel computing.
- Twin-Screw Extruder Devolatilization Prediction via Flow Segmentation
Segmenting flow paths into three profiles resolves Latinen's model inaccuracy by accounting for dynamic surface renewal and changing fluid behavior.