Team:UCAS-China/Model

• Home
• Project
  1. Description
  2. Design
  3. Proof of Concept
  4. Safety
  5. Implementation
  6. Engineering Success
  7. Measurement
  8. Experiments
  9. Parts
  10. Notebook
• Hardware
  1. Hardware Overview
  2. Technical details
• Software
• Model
  1. Model Overview
  2. Model for the Tea Pack
  3. Model for the Loophole
  4. Modelling Methods
• Human Practice
  1. Integrated Human Practice
  2. Communication
  3. Education
  4. Collaboration
  5. Partnership
  6. Excellence In Another Area
• Team
  1. Meet the Team!
  2. Attributions
  3. Contribution
  4. Acknowledgement

• Overview
• Model Overview

Model Overview

For the model part, we firstly analyzed two types of hardware design. For the tea pack, we assumed that diffusion rate is infinite compared to the reaction rate. Therefore, simple Michaelis-Menten kinetics could be used to describe the whole system. When the system is described, we can analyze the performance of caffeine decrease and SAH adsorption. Additionally, as the temperature changes when the reactions process, and this will influence the speed of reaction, therefore, we evaluated the influence of temperature. For this part, we considered caffeine concentrations and SAH leakage. Specially, we found that the enzymes might denature when operating, so we also estimated the operating life of our tea pack.

The other hardware design is a loophole complex. For this part, most of the models were similar, but in order to control the caffeine concentration, a valve was used here to control the flow rate of solution. As we assumed that reaction rate was proportional to contact area with filters and friction of liquid was related to surface tension, we claimed that adjustment of the valve could result in caffeine concentration control. According to those factors, we could estimate the reaction velocity and caffeine concentration. Another device was used to adsorb SAH, and for the device, we also estimated the time consumption for adsorption.

Finally, the influence of caffeine was estimated by a mixed model - that is, a linear metabolism model concatenated with a multi-head neuron network, whose input came from caffeine intake items and users’ feedbacks. As we need a huge quantity of parameters to describe a user, if we attempted to fetch all the parameters, it might fall to failure, so we constructed a neural network to learn these parameters from users’ history and feedbacks, as they were relative easier to access. Therefore, a personalized predictor can be trained for users.

In addition, as for tools, we used PROSS to optimize the thermal stability of CkTcS, therefore, it could be operated in a higher temperature to keep the taste of coffee.

Detaila of our models are illustrated in the following pages:

• Model for the hardware: tea pack
• Model for the hardware: loophole complex
• Modelling Methods