7.7.8. Blower Imbalance
Detect blade imbalance in HVAC blowers using 3-phase motor current analysis.
7.7.8.1. Overview
Blade imbalance in HVAC blowers causes increased vibration, noise, and premature bearing wear. This example uses motor current signature analysis to detect imbalance conditions before they cause equipment failure.
Application: HVAC systems, industrial fans, predictive maintenance
Task Type: Time Series Classification
Data Type: Multivariate (3-phase motor currents)
7.7.8.2. Configuration
common:
target_module: 'timeseries'
task_type: 'generic_timeseries_classification'
target_device: 'F28P55'
dataset:
dataset_name: 'blower_imbalance'
training:
model_name: 'FanImbalance_model_1_t'
training_epochs: 50
batch_size: 32
testing: {}
compilation: {}
7.7.8.3. Running the Example
cd tinyml-modelzoo
./run_tinyml_modelzoo.sh examples/blower_imbalance/config.yaml
cd tinyml-modelzoo
run_tinyml_modelzoo.bat examples\blower_imbalance\config.yaml
7.7.8.4. Dataset Details
Input Variables:
Phase A current
Phase B current
Phase C current
Classes:
Normal operation
Blade imbalance detected
7.7.8.5. Recommended Models
Important
The FanImbalance_model_* models listed below are only available in
TI’s Edge AI Studio (GUI) and are not included in Tensorlab.
Use the generic CLS_*_NPU models (e.g., CLS_1k_NPU,
CLS_4k_NPU) as equivalent alternatives in Tensorlab.
Edge AI Studio models:
Model |
Parameters |
Use Case |
|---|---|---|
|
Varies |
Baseline detection |
|
Varies |
Improved accuracy |
|
Varies |
Maximum accuracy |
Tensorlab alternatives: Use CLS_1k_NPU, CLS_2k_NPU, or
CLS_4k_NPU for equivalent performance.
7.7.8.6. See Also
Fan Blade Fault Classification - Accelerometer-based fan fault detection