ParticleTensorAI® PTA4

ParticleTensorAI® PTA4 is the four-channel configuration of the ParticleTensorAI® measurement platform. It combines the LSS4 optical probe with the CLEON control unit for AI-based analysis of synchronized time-resolved light-scattering signals.

PTA4 records four optical detector signals simultaneously. Instead of reducing the signals immediately to individual droplet or particle parameters, the synchronized measurement data are retained over a defined acquisition interval and transformed into a multidimensional representation for ParticleTensorAI® analysis.

The four detector channels provide the most extensive standard optical signal representation within the ParticleTensorAI® platform. The AI model can analyze signal structures within each detector channel as well as temporal and optical relationships between all four channels.

PTA4 consists of the LSS4 optical measurement probe, the CLEON multi-channel control and acquisition unit, and the ParticleTensorAI® analysis software.

Component	PTA4 Configuration	Function
Optical probe	LSS4	Generates the measurement region and detects time-resolved scattered light
Detector channels	4	Records four synchronized optical signal streams
Control unit	CLEON	Multi-channel signal acquisition, buffering and processing
Evaluation	ParticleTensorAI®	AI-based analysis of buffered multi-channel signal data

The LSS4 optical probe generates a defined illumination field in the measurement region. Droplets or particles passing through this field produce time-dependent scattered-light signals that are detected by four optical channels.

Each detector observes the particle interaction from its defined optical configuration. The resulting synchronized signals contain information about the passage of particles through the measurement region and about their interaction with the optical field.

In PTA4 operation, CLEON retains the four signal streams over a defined acquisition interval and makes them available for ParticleTensorAI® processing.

The defining feature of PTA4 is the simultaneous acquisition of four time-resolved detector channels.

The four channels do not simply provide redundant measurements. They provide multiple optical views of the same particle population and therefore increase the amount of information available to the AI model.

Depending on the optical configuration, ParticleTensorAI® can analyze:

• signal shape within each detector channel
• signal amplitude and intensity distributions
• temporal signal sequences
• time relationships between detector channels
• particle or droplet event density
• overlapping signal structures
• statistical patterns within the acquisition interval
• relationships between multiple optical detector views
• changes in multi-channel signal distributions over time

ParticleTensorAI® preserves larger portions of the original optical measurement data instead of immediately converting every detected signal into a small number of predefined particle parameters.

The four synchronized detector signals are recorded over a defined measurement interval and transformed into a multidimensional data representation. This representation preserves temporal information together with relationships between the optical detector channels.

The exact data representation and AI model can be adapted to the measurement task. This makes PTA4 suitable for research and development of measurement concepts that require more optical information than a single- or two-channel configuration can provide.

PTA4 can analyze complete buffered signal sequences without requiring every droplet or particle event to be identified and separated individually before the AI evaluation.

This can be particularly useful for complex or dense sprays where particle events occur in rapid succession or several light-scattering signals overlap.

Instead of discarding such complex measurement intervals, ParticleTensorAI® can retain the multi-channel signal information and make it available to the AI model.

PTA4 is designed exclusively for AI-based evaluation. The relationship between the measured four-channel optical signal structure and the required output is learned from suitable training and validation data.

The four detector channels allow the AI model to use both information contained within each individual signal stream and relationships between different detector channels.

Depending on the application, this information can be used to develop AI models for particle, spray, material, or process characterization.

The available results depend on the optical configuration, training data, reference information, and AI model developed for the application.

• spray condition classification
• particle or droplet classification
• detection of process changes
• material-related particle characterization
• analysis of changes in optical particle properties
• comparison of different operating conditions
• monitoring of stable and unstable spray states
• analysis of high-concentration sprays
• AI-based quality monitoring
• advanced process-state characterization

The structure of a light-scattering signal depends not only on particle size and velocity. Optical and material-related properties can also influence the measured waveform.

With four synchronized detector channels, PTA4 provides multiple optical views of these interactions. This creates a broad data basis for investigating whether changes in material composition or particle properties produce characteristic multi-channel signal patterns.

PTA4 is therefore particularly suitable for research tasks in which conventional particle size and velocity measurements are not sufficient and additional material-related information is required.

In sprays with high particle or droplet concentrations, individual light-scattering signals can overlap. Classical counting methods may become increasingly difficult when individual events can no longer be separated reliably.

ParticleTensorAI® can retain and analyze complete multi-channel signal sequences. This provides an alternative approach for extracting information from measurement intervals containing many or overlapping particle events.

The combination of four detector channels and buffered AI-based analysis makes PTA4 particularly relevant for research on complex sprays and multiphase particle systems.

PTA4 requires suitable training and validation data for the intended measurement task. During model development, four-channel optical signals are recorded under defined reference conditions and assigned to known classes or reference quantities.

The dataset is divided into training and independent validation data. The AI model learns relationships within the multi-channel signal representation and is subsequently tested using measurements that were not included in model training.

Important factors include:

• quality of the reference measurements
• range of operating conditions
• material and optical properties
• stability of the optical configuration
• relationships between detector channels
• particle or droplet concentration
• number and diversity of measurement cases
• selected AI model architecture
• similarity between training and later application conditions

Customer-specific PTA4 applications may therefore require an individual measurement, training, and validation campaign.

The LSS4 is the four-channel optical measurement probe used by PTA4. It generates the optical measurement region and provides four synchronized time-resolved light-scattering signals.

The LSS4 platform can be configured for different measurement tasks. Optical geometry, working distance, wavelength, and mechanical integration can be adapted to the application.

Parameter	LSS4
Measurement principle	Time-resolved light scattering
Active detector channels	4
Available wavelengths	405 nm, 450 nm
Dimensions	Approx. 200 × 150 × 70 mm
Compatible control unit	CLEON

Technical details for LSS4

The CLEON control unit provides synchronized multi-channel signal acquisition and processing for the LSS4 optical probe.

In PTA4 operation, CLEON is configured to retain and process the four time-resolved detector signals for ParticleTensorAI® analysis.

The same hardware platform can also be used for classical TSTOF measurements. The difference lies in the acquisition workflow, firmware, calculation method, and evaluation software.

Technical details for CLEON

PTA4 and TSTOF use the same LSS4 optical probe and CLEON multi-channel acquisition platform. The difference is how the four detector signals are evaluated.

Parameter	TSTOF	PTA4
Optical probe	LSS4	LSS4
Control unit	CLEON	CLEON
Detector channels	4	4
Primary data unit	Individual particle event	Buffered multi-channel signal sequence
Evaluation concept	Classical physical TSTOF calculation	Tensor-based AI analysis
Individual event separation	Required for classical evaluation	Not always required
Typical focus	Particle size and velocity measurement	Complex particle, material and process characterization

This comparison demonstrates the modular concept of the ai-quanton measurement platform: the same physical optical signals can be used with fundamentally different evaluation strategies.

PTA4 can analyze consecutive measurement intervals during an operating particle or spray process. Each measurement interval can be evaluated by the trained AI model and converted into an application-specific output.

Possible outputs include classifications, predicted parameters, probability values, trend information, or indicators describing changes relative to a defined reference process.

The four-channel configuration provides an extensive optical data basis for detecting subtle changes in particle populations, material properties, or process conditions.

• advanced AI-based spray characterization
• material characterization in sprays
• paint and coating process monitoring
• high-concentration sprays
• complex multiphase particle systems
• detection of process changes
• comparison of atomizer operating conditions
• research on multi-channel optical scattering signatures
• development of new AI-based measurement methods
• industrial process monitoring

PTA4 can be adapted to demanding measurement tasks. The final configuration depends on the spray or particle system, material, optical properties, concentration, process conditions, available installation space, reference data, and required analysis result.

Possible adaptations include:

• optical probe geometry
• working distance
• illumination configuration
• detector arrangement
• wavelength
• mechanical integration
• acquisition interval
• buffer configuration
• tensor representation
• AI model architecture
• output parameters
• process interfaces

The three standard ParticleTensorAI® systems differ mainly in the number of synchronized optical detector channels and therefore in the amount of optical information available for AI analysis.

System	Optical Probe	Control Unit	Detector Channels	Characteristic
PTA1	LSS1	ZEON	1	Compact single-channel configuration
PTA2	LSS2	IMEA	2	Two synchronized detector channels
PTA4	LSS4	CLEON	4	Most extensive standard multi-channel configuration

PTA1 provides the most compact configuration. PTA2 adds detector-to-detector information using two synchronized channels. PTA4 provides four synchronized channels and is intended for applications where the largest amount of optical signal information is required.

PTA4 combines the established multi-channel optical measurement platform of TSTOF with the AI-based ParticleTensorAI® evaluation concept.

Research has investigated the use of time-resolved light-scattering signals for particle characterization, material-related analysis, complex sprays, and high-concentration particle systems. ParticleTensorAI® extends this work by retaining larger portions of the original multi-channel measurement information for AI-based analysis.

[1] Schaefer, W.; Li, L.; Stegmann, P.; Terada, M. Technical Report on the TSTOF Measurement Method: Technical Basics, Historical Development, and Comparison with Other Laser-Based Measurement Methods. Photonics 2026, 13(1), 56. https://doi.org/10.3390/photonics13010056

[2] Schaefer, W.; Fleck, S.; Haas, M.; Jakobs, T. Optical measurement method for monitoring high-mass-concentration slurry sprays: An experimental study. Photonics 2025, 12(7), 673. https://doi.org/10.3390/photonics12070673

[3] Schäfer, W.; Jakobs, T.; Stegmann, P. AI-assisted monitoring of coal particles within droplets in a coal monoethylene glycol slurry spray. 41st International Symposium on Combustion, Kyoto, Japan, 2026. KITopen ID 1000196272. https://doi.org/10.5445/IR/1000196272

• ParticleTensorAI® Overview
• ParticleTensorAI® PTA1
• ParticleTensorAI® PTA2
• LSS4 Optical Probe
• CLEON Control Unit
• TSTOF

For an application-specific PTA4 configuration, contact ai-quanton with information about the spray or particle system, material, operating conditions, concentration, available reference data, and required analysis result.