How to Control Air Injection Timing for NOx Reduction

Overview of Technical Issues:

The injection timing control mechanism insufficiently regulates when secondary air enters the exhaust stream, failing to synchronize with the optimal temperature and chemical composition window in the exhaust gas flow, which directly reduces NOx conversion efficiency and prevents the system from meeting emission reduction targets.

Solution directions generated for this problem

Problem Direction 1 :

ImproveInjection timing control precision
VS
ConstraintControl system complexity

Inspiration 1 : Cross-domain reference

Application Principle: #26 Copying
Cross-domain applicability Assess applicability
Compact aero-thermo model based engine material temperature control
Innovative Solution Refine solution

Virtual exhaust state model for predictive injection timing control

Virtual exhaust model predicts optimal timing window using minimal sensors
How to solve :
  • Build a real-time virtual exhaust model using existing engine control unit with RPM, throttle position, and one fast O2 sensor (λ-sensor, 30ms response) to predict exhaust temperature and NOx concentration profiles without adding physical sensors
  • Train a neural network surrogate model offline using 500+ engine operating cycles across idle/cruise/acceleration modes, mapping engine parameters to exhaust composition with ±8°C temperature accuracy and ±12ms timing prediction
  • Implement adaptive correction algorithm that updates model coefficients every 50 cycles based on downstream NOx sensor feedback, maintaining ±10ms timing precision across 90% of operating conditions while using only 9 total control modules
Expected Effect : Timing precision ±10ms achieved; module count held at 9 vs 15+; NOx reduction efficiency +28%
Risk Control :
  • neural network training data insufficiency
  • model drift under extreme transient conditions
  • O2 sensor aging affects input accuracy

Problem Direction 2 :

ImproveExhaust parameter detection response speed
VS
ConstraintSensor manufacturing precision requirement

Inspiration 1 : Cross-domain reference

Application Principle: #28 Mechanics substitution
Cross-domain applicability Assess applicability
Decoding approaches for protein identification
Innovative Solution Refine solution

Infrared pyrometry with dual-wavelength ratio thermometry for fast exhaust sensing

Replace contact sensors with optical measurement
How to solve :
  • Install dual-wavelength infrared pyrometer (measuring at 1.5μm and 2.2μm wavelengths) 15cm upstream of injection point, achieving <15ms temperature response without physical contact
  • Apply ratio thermometry algorithm (T = C·ln(I₁/I₂)) to eliminate emissivity variation effects, maintaining ±8°C accuracy across 300-800°C range using standard optical components (sapphire window, InGaAs photodiodes)
  • Integrate optical O2 sensing via tunable diode laser absorption spectroscopy at 760nm, co-axial with pyrometer beam, providing simultaneous temperature and composition data in single 25mm diameter probe assembly
Expected Effect : Response time <15ms (5× faster), manufacturing cost -60%, accuracy ±8°C with standard optics
Risk Control :
  • optical window fouling by soot deposits
  • ambient vibration affecting beam alignment
  • calibration drift from window degradation

Problem Direction 3 :

ImproveControl synchronization accuracy
VS
ConstraintControl system complexity

Inspiration 1 : Cross-domain reference

Application Principle: #35 Parameter changes
Cross-domain applicability Assess applicability
Method for processing received PDCP PDUs for D2D communication system and device therefor
Innovative Solution Refine solution

Adaptive injection timing map with self-calibrating feedback loop

Self-learning timing maps adjust injection windows based on emission outcomes
How to solve :
  • Implement adaptive timing lookup tables storing 20 pre-calculated injection windows indexed by engine RPM (500-6000 rpm in 250 rpm steps), load (0-100% in 10% steps), and coolant temperature (60-110°C in 10°C bands) — tables reside in existing ECU memory requiring only 8 KB additional storage
  • Install single downstream NOx sensor (zirconia-based, ±15 ppm accuracy, 200 ms response) measuring actual NOx conversion efficiency every 5 engine cycles, comparing against 85% target threshold — when below target for 3 consecutive cycles, shift timing map entry by ±3 ms and re-evaluate
  • Execute closed-loop calibration algorithm running at 10 Hz background priority that updates timing table entries based on weighted moving average of last 50 cycles — convergence tolerance ±2 ms, quality gate requires 90% cycles within ±10 ms of optimal window over 1000-cycle validation period
Expected Effect : Synchronization accuracy 92%, system modules remain at 9, NOx sensor cost $45
Risk Control :
  • timing map convergence instability during rapid transients
  • NOx sensor drift after 80,000 km requiring recalibration
  • memory corruption risk in lookup table storage

Problem Direction 4 :

ImproveControl synchronization accuracy
VS
ConstraintSensor manufacturing precision requirement

Inspiration 1 : Cross-domain reference

Application Principle: #11 Beforehand cushioning
Cross-domain applicability Assess applicability
Radiographic apparatus, system, control method and computer-readable storage medium
Innovative Solution Refine solution

Self-calibrating injection timing via embedded reference thermal mass

Embed reference element to enable runtime calibration without tight sensor tolerances
How to solve :
  • Install a ceramic reference thermal mass (known heat capacity 0.8 kJ/kg·K, 5g weight) 10cm upstream of each exhaust sensor, creating a predictable temperature signature during transient events
  • Program control unit to measure sensor response to reference element thermal lag during engine startup and every 500 operating cycles, calculating real-time calibration correction factors (±0.5-2.0 multiplier range) that compensate for ±10°C sensor manufacturing variation
  • Apply stored correction factors to all runtime temperature readings, achieving effective ±6°C system accuracy and <30ms corrected response time using standard ±15°C sensors with relaxed manufacturing tolerances (cost reduction 60%)
Expected Effect : Synchronization accuracy 92%, sensor tolerance ±15°C, cost -60%
Risk Control :
  • reference element fouling over time
  • calibration algorithm convergence failure
  • thermal mass detachment under vibration
Patsnap Eureka Solution