Chaitanya Krishna Prasad Vallabh
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chaitanya Krishna Prasad Vallabh has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
Cetin Cetinkaya
7 shared publications- Cetin Cetinkaya
Photo-Acoustics Research Laboratory, Department of Mechanical and Aerospace Engineering, Clarkson University, Potsdam, NY 13699-5725, USA
5 shared publications James D. Stephens
5 shared publicationsXiaochi Xu
Intel Corp, USA
4 shared publications- Maxine Fontaine3 shared publications
- Ajaz S. Hussain
Insight, Advice & Solutions LLC, USA
2 shared publications
Latest Publications
- 21 August 2026
- Published by American Society for Engineering Education
The Role of Geometry Skills in Statics Problem Solving
- 21 August 2026
- Published by American Society for Engineering Education
The Role of Geometry Skills in Statics Problem Solving
- Open Access
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
Sim-to-Real Transfer Learning for Real-Time Melt Pool Geometry Prediction in Directed Energy Deposition
In Directed Energy Deposition (DED), melt pool geometry governs part quality, but its depth lies below the surface and cannot be measured directly in real time, a long-standing barrier to closed-loop control. This paper presents a Sim-to-Real transfer learning framework that predicts melt pool width and depth in real time from infrared (IR) imagery and process parameters, bridging physics-based thermal simulation with a small set of experimental data. A multi-input hybrid CNN is pretrained on a large, simulated corpus of surface temperature maps paired with laser power, scan speed, and spot size, then fine-tuned, with all layers unfrozen, on a limited set of labeled IR frames from single-track builds. Against a baseline trained on experiments alone, the fine-tuned model is more accurate and converges faster and more stably, with the clearest gains for depth in the low-data regime (for small training sets it reaches R² > 0.7 where the baseline stays below 0.2). On six previously unseen single-track builds, frame-by-frame real-time prediction yields a mean absolute percentage error (MAPE) of 7.41% for width and 22.45% for depth, versus 29.35% and 74.66% for the baseline. An input ablation confirms that pairing thermal imagery with process parameters gives the best accuracy. Together these results show the approach is practical for sensor-driven monitoring and points toward closed-loop control in metal additive manufacturing.
- 12 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSim-to-Real Transfer Learning for Real-Time Melt Pool Geometry Prediction in Directed Energy Deposition
In Directed Energy Deposition (DED), melt pool geometry governs part quality, but its depth lies below the surface and cannot be measured directly in real time, a long-standing barrier to closed-loop control. This paper presents a Sim-to-Real transfer learning framework that predicts melt pool width and depth in real time from infrared (IR) imagery and process parameters, bridging physics-based thermal simulation with a small set of experimental data. A multi-input hybrid CNN is pretrained on a large, simulated corpus of surface temperature maps paired with laser power, scan speed, and spot size, then fine-tuned, with all layers unfrozen, on a limited set of labeled IR frames from single-track builds. Against a baseline trained on experiments alone, the fine-tuned model is more accurate and converges faster and more stably, with the clearest gains for depth in the low-data regime (for small training sets it reaches R² > 0.7 where the baseline stays below 0.2). On six previously unseen single-track builds, frame-by-frame real-time prediction yields a mean absolute percentage error (MAPE) of 7.41% for width and 22.45% for depth, versus 29.35% and 74.66% for the baseline. An input ablation confirms that pairing thermal imagery with process parameters gives the best accuracy. Together these results show the approach is practical for sensor-driven monitoring and points toward closed-loop control in metal additive manufacturing.
- 27 April 2026
- Published by Elsevier in International Journal of Pharmaceutics
- Vol. 698, 126925
A review of the State-of-the-Art: progress in ultrasonic and acoustic techniques for quality assessment in the development and manufacturing of oral solid dosage forms − Part II: Applications and emerging directions
- 27 April 2026
- Published by Elsevier in International Journal of Pharmaceutics
- Vol. 698, 126925
A review of the State-of-the-Art: progress in ultrasonic and acoustic techniques for quality assessment in the development and manufacturing of oral solid dosage forms − Part II: Applications and emerging directions
- 25 November 2025
- Published by Elsevier in International Journal of Pharmaceutics
- Vol. 687, 126429
A review of the state-of-the-art: progress in ultrasonic and acoustic techniques for quality assessment in the development and manufacturing of oral solid dosage forms − Part I: theoretical foundations and principles
- 25 November 2025
- Published by Elsevier in International Journal of Pharmaceutics
- Vol. 687, 126429
A review of the state-of-the-art: progress in ultrasonic and acoustic techniques for quality assessment in the development and manufacturing of oral solid dosage forms − Part I: theoretical foundations and principles
- 21 August 2025
- Published by American Society for Engineering Education
Effects of Spatial Ability on the Mastery of Statics Concepts
- 21 August 2025
- Published by American Society for Engineering Education
Effects of Spatial Ability on the Mastery of Statics Concepts
- Open Access
- 21 November 2024
- Published by Springer Nature in Research Square
Using Acoustic Emission Signal Analysis and Machine Learning Algorithms for Predicting Mechanical Hardness in Laser Directed Energy Deposition Parts
Despite recent advancements in Laser Directed Energy Deposition (LDED) Additive Manufacturing (AM) process, LDED fabricated parts frequently suffer from predictable and optimized mechanical performance and print quality. To address this, efforts are focused on predicting mechanical properties of fabricated parts using real-time monitoring data. This study aims to develop a non-destructive method coupled with machine learning (ML) models for predicting the mechanical hardness of samples printed by LDED process using acoustic emission (AE). The influence of key printing parameters, namely overlap ratio, dwell time, and number of layers, on the relationship between AE signal characteristics and the resulting mechanical hardness are investigated. AE features are extracted through signal processing from the raw signals acquired during the printing process. The mechanical hardness of the fabricated samples are then evaluated using Rockwell Hardness (HRB) measurements. For establishing a predictive relationship between the obtained AE features and the mechanical hardness, we develop and test three ML models, namely, Support Vector Regression (SVR), Gradient Boosting Regression (GBR), and Gaussian Process Regression (GPR). The performance of these models is assessed using two key metrics: Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Among the tested models, GBR demonstrated the highest accuracy, achieving an MAE of 1.41 and a MAPE of 1.88% in predicting the hardness of fabricated samples. These results highlight the potential of integrating AE sensors with advanced machine learning algorithms for real-time, non-destructive prediction of mechanical properties in LDED AM process.
- 21 November 2024
- Published by Springer Nature in Research Square
Open AccessUsing Acoustic Emission Signal Analysis and Machine Learning Algorithms for Predicting Mechanical Hardness in Laser Directed Energy Deposition Parts
Despite recent advancements in Laser Directed Energy Deposition (LDED) Additive Manufacturing (AM) process, LDED fabricated parts frequently suffer from predictable and optimized mechanical performance and print quality. To address this, efforts are focused on predicting mechanical properties of fabricated parts using real-time monitoring data. This study aims to develop a non-destructive method coupled with machine learning (ML) models for predicting the mechanical hardness of samples printed by LDED process using acoustic emission (AE). The influence of key printing parameters, namely overlap ratio, dwell time, and number of layers, on the relationship between AE signal characteristics and the resulting mechanical hardness are investigated. AE features are extracted through signal processing from the raw signals acquired during the printing process. The mechanical hardness of the fabricated samples are then evaluated using Rockwell Hardness (HRB) measurements. For establishing a predictive relationship between the obtained AE features and the mechanical hardness, we develop and test three ML models, namely, Support Vector Regression (SVR), Gradient Boosting Regression (GBR), and Gaussian Process Regression (GPR). The performance of these models is assessed using two key metrics: Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Among the tested models, GBR demonstrated the highest accuracy, achieving an MAE of 1.41 and a MAPE of 1.88% in predicting the hardness of fabricated samples. These results highlight the potential of integrating AE sensors with advanced machine learning algorithms for real-time, non-destructive prediction of mechanical properties in LDED AM process.
- 12 January 2024
- Published by ASME International in Journal of Manufacturing Science and Engineering
- Vol. 146 (3) , 1-17
An Additive Manufacturing Testbed to Evaluate Machine Learning-Based Autonomous Manufacturing
This paper details the design and operation of a testbed to evaluate the concept of autonomous manufacturing to achieve a desired manufactured part performance specification. This testbed, the autonomous manufacturing system for phononic crystals (AMSPnC), is composed of additive manufacturing, material transport, ultrasonic testing, and cognition subsystems. Critically, the AMSPnC exhibits common manufacturing deficiencies such as process operating window limits, process uncertainty, and probabilistic failure. A case study illustrates the AMSPnC function using a standard supervised learning model trained by printing and testing an array of 48 unique designs that span the allowable design space. Using this model, three separate performance specifications are defined and an optimization algorithm is applied to autonomously select three corresponding design sets to achieve the specified performance. Validation manufacturing and testing confirms that two of the three optimal designs, as defined by an objective function, achieve the desired performance, with the third being outside the design window in which a distinct bandpass is achieved in phononic crystals (PnCs). Furthermore, across all samples, there is a marked difference between the observed bandpass characteristics and predictions from finite elements method computation, highlighting the importance of autonomous manufacturing for complex manufacturing objectives.
- 12 January 2024
- Published by ASME International in Journal of Manufacturing Science and Engineering
- Vol. 146 (3) , 1-17
An Additive Manufacturing Testbed to Evaluate Machine Learning-Based Autonomous Manufacturing
This paper details the design and operation of a testbed to evaluate the concept of autonomous manufacturing to achieve a desired manufactured part performance specification. This testbed, the autonomous manufacturing system for phononic crystals (AMSPnC), is composed of additive manufacturing, material transport, ultrasonic testing, and cognition subsystems. Critically, the AMSPnC exhibits common manufacturing deficiencies such as process operating window limits, process uncertainty, and probabilistic failure. A case study illustrates the AMSPnC function using a standard supervised learning model trained by printing and testing an array of 48 unique designs that span the allowable design space. Using this model, three separate performance specifications are defined and an optimization algorithm is applied to autonomously select three corresponding design sets to achieve the specified performance. Validation manufacturing and testing confirms that two of the three optimal designs, as defined by an objective function, achieve the desired performance, with the third being outside the design window in which a distinct bandpass is achieved in phononic crystals (PnCs). Furthermore, across all samples, there is a marked difference between the observed bandpass characteristics and predictions from finite elements method computation, highlighting the importance of autonomous manufacturing for complex manufacturing objectives.