Hybride Modellierung für die Vorhersage und Überwachung des Wachstums von Mikroalgen

Tehreem Syed

doi:10.51202/9783186482204

Keywords

Search publication

Bibliographic data

Publication year: 2025
Publication date: 28.05.2025
ISBN-Print: 978-3-18-348220-7
ISBN-Online: 978-3-18-648220-4
Publisher: VDI Verlag, Düsseldorf
Series: Rechnerunterstützte Verfahren
Volume: 482
Language: German
Pages: 136
Product type: Book Titles

Titelei/Inhaltsverzeichnis No access Pages I - XVIII
1. Motivation No access
2. Objectives No access
3. Contributions No access
4. Overview of the dissertation No access
1. 1. Light No access
  2. Nutrients No access
  3. Temperature No access
  4. pH No access
  5. Salinity No access
2. Types of microalgae cultivation system No access
3. 1. Types of Open Systems No access
  2. Racewayponds No access
  3. Circular ponds No access
4. 1. Types of Closed Systems No access
  2. FlatpanelPBRs No access
  3. Tubular PBRs No access
  4. Verticle column PBRs No access
5. 1. Models accounting for light intensity effect No access
  2. Models accounting for light intensity and temperature effect No access
  3. Models accounting for light intensity and substrate effect No access
6. 1. Prediction of microalgae growth or productivity using neural- network-based and non-neural-network-based approaches . No access
7. 1. Hybrid modeling application in biotechnological processes No access
  2. Hybrid modeling application in chemical engineering . . . No access
1. 1. Challenges in Data for Microalgae Cultivation No access
  2. Challenges for modeling of microalgae cultivation No access
2. 1. Potential solution for the dataset enhancement No access
  2. Potential solutions for modeling rigorousness enhancement No access
3. Research Questions, Hypothesis and Methodology No access
1. 1. Description of cultivation No access
  2. Datapreprcoessing No access
  3. Training and Test Dataset No access
2. 1. Average light intensity No access
  2. Monod and Haldane model No access
3. 1. Support vector regression No access
  2. LSTM No access
  3. Analysis of the trained machine learning models No access
4. 1. Biomasssoftsensor No access
  2. Harvest strategy No access
1. 1. Description of Dataset No access
  2. Data Preprocessing No access
  3. Training and Test Dataset No access
2. 1. Monod Model No access
  2. LSTMModel No access
  3. Biomass prediction using Runge-Kutta No access
3. 1. Hybrid Model Approach 1 No access
  2. Theintegratorcell No access
  3. Hybrid Model Approach 2 No access
1. 1. Comparison of machine learning models No access
  2. Light acclimation impact on specific growth rate No access
  3. Applications of the models No access
2. 1. Evaluation of Sequence Lengths in LSTM No access
  2. LSTM varying light sequence length and train-test batch ratios No access
  3. LSTM-based softsensor for the prediction of microalgal biomass No access
3. 1. Performance evaluation of LSTM as a residual predictor of specificgrowth No access
  2. Evaluation of Hybrid model performance across varying training and test batches for LSTM residual predictor and biomassprediction No access
  3. Hyperparameter optimization No access
1. 1. Comparison between machine learning and traditional models No access
  2. Interpretation of Machine Learning Models’ Performance . No access
  3. Interpretation of Light Acclimation and Respiration in Mi¬croalgae Growth No access
  4. Implications of Model Applications No access
  5. Interpretation of Sequence Lengths in LSTM No access
  6. Interpretation of LSTM Model Performance with Varying Light Sequence Lengths and Train-Test Batch Ratios . . . No access
2. Hybrid Model Sensitivity to Training and Testing Data Variability No access
3. 1. Integration with ML models No access
  2. Vulnerability of Hybrid Models and LSTM Models to Data, Normalization, and Scaling No access
4. Comparative analysis of LSTM and Hybrid model No access
1. Summary No access
2. 1. 1. LightAttenuationModel No access
    2. Evaluation of Hybrid model performance across varying training and test batches for LSTM residual predictor No access
Bibliography No access Pages 109 - 130
Declaration No access Pages 131 - 136