Design and intelligent control of a tropical produce storage system /
Availability and price of produces especially during their off-season are influenced depending on their storage systems. Produces like tubers are staple food in many countries in the tropical region. However, there exists no scientific storage systems for such produces, thus shelf life of these prod...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur :
Kulliyyah of Engineering, International Islamic University Malaysia,
2016
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Availability and price of produces especially during their off-season are influenced depending on their storage systems. Produces like tubers are staple food in many countries in the tropical region. However, there exists no scientific storage systems for such produces, thus shelf life of these produces is significantly low. Even in general, limited uses of intelligent control are found in postharvest storage process. This study proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS) approach for tropical produce because of its combined advantages of fuzzy inference and neural networks. Thermal properties of the taro tuber (Colocasiaesculenta) as affected by tuber temperature and moisture content were determined experimentally using transient methods. Models to predict thermal properties as a function of tuber temperature and moisture content were developed. Mathematical modeling of the storage process followed, where computational fluid dynamics (CFD) was used to determine the air velocity and pressure distribution within the product-filled storage unit. An algorithm was developed to analyze the heat and mass transfer occurring within the storage volume. The new algorithm served as the plant model used in subsequent simulation studies. A tubular storage system prototype was developed and integrated with a cooling system with the provision of air flow and temperature control. Intelligent controllers - fuzzy logic and ANFIS controllers – for storage temperature and humidity control were developed. The feed-back Temperature Fuzzy Logic Controller (TFLC) had as inputs error and change in error while the output was inlet air temperature. The feed-forward Humidity Fuzzy Logic Controller (HFLC) had as inputs inlet air temperature (from the TFLC) and inlet air moisture deficit whereas the output was the duration of moisture injection. Different FLC configurations (type and number of membership functions) were investigated for both controllers. For the TFLC, a 3×3 Triangular MFs with Trapezoidal shoulders was found to perform the best while for humidity control, a 5×5 Gaussian MFs structure was found to have the best performance. Input/output data from the optimally performing TFLC and HFLC were used to develop the ANFIS Temperature Controller (ATC) and ANFIS Humidity Controller (AHC) respectively. With respect to MSE, the ATC and AHC performed better than the corresponding TFLC and HFLC, while in the case of PMTD, TFLC recorded better performance in comparison to the ATC whereas the AHC maintained the edge over the HFLC. Real-time operation of the storage process involved temperature control only and was split into two stages. The first stage involved water-filled Polythylene Terephthalate (PET) bottles, after which the scaling factor for the controllers were adjusted for the actual storage of the taro tubers. It was found that the ATC performed better than the TFLC even in the face of disturbances. MSE values for water-filled PET bottles storage using the TFLC and ATC were 0.5554 and 0.388 respectively, whereas real-time storage by the same controllers yielded values of 0.8361 and 1.0286 respectively. ANFIS control of the postharvest storage process comes highly recommended by the results of this study. |
|---|---|
| Physical Description: | xxvi, 207 leaves : ill. ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 190-201). |