Charged particles in a conductive ratchet, simulation is done using CAVIAR package
Mainly, I would like to reach to the better understanding of the physics of the Complex Systems . What I would like to know is the Emergence and how it occur in nature. However, other phenomena in soft condensed matter maybe interesting, such as Granular Materials and Diffusion of Charged Particles in Conductive Porous Environments. Besides, I am interesting on the application of complex systems, especially in innovative technologies. Recently, I focus on three major subjects;
For more details, please click on each item to get more information.
Transport of the charged particles in an environment with conductive boundaries is of scientific interest, due to its theoretical and practical applications. For example, transport of particles with permanent charges in conductive porous media has a wide ranges of applications, such as super capacitors and new generation of batteries. These systems gain much technologically attention because of its application in future electrical energy storage. These devices, unlike ordinary chemical batteries, have infinite number of charging and discharging (at least in theory)
These systems involve complex interactions such a long range electrical interactions and are hard to studying theoretically. On the other hands, experimental investigation of this systems can not cover all details of charging and discharging processes. Therefore, simulation is an alternative for studying these systems in details. However, missing a suitable method for taking the interactions of charged particles with conductive boundaries into account, make the simulation of such systems hard and computational time consuming.
One of my goal in this field of research is introducing new algorithms for simulating such systems. Here, in our group, we works on developing new algorithm, which is called to Laplace transformation (PLT) algorithm to simulate these systems. Besides, we are developing a simulation package, named as CAVIAR, to enable us simulate any complex geometry. Details of this package can be found in Softwares page, which is available from toolbar.
In addition to developing algorithm for simulation transport of charges particles, I am interesting on systems and geometries which be able to store more electrical charge/energy. We mainly looking for structures which have more electrical energy density. In addition, I would like to understand the role of geometry and structure of the conductive porous electrodes on discharging process. It is important for next generation of electrical storage devices to have a suitable profile of discharging as well, it is expected from an ideal battery to keep the voltage difference between its electrodes to be constant during discharging process. In real world, however, have approximately constant voltage difference for a reasonable time of discharging is what one expected. Up to now, super-capacitors have linear voltage drop with time, which in turn makes lots of technical problems. Indeed, in addition to low density of storage electrical energy and power compare with chemical batteries, linear drop of voltage difference is two major obstacles for playing the electrical storage role in industry.
Another important application of these studies is desalination of see water. In many countries accessing to sustainable resources of fresh water is one of a major concerns of governments. The global warming is intensified this problem. One of the solution is getting fresh water by desalination of see water. There are lots of scientific/technical efforts are done to suggest an economic method for desalination. On of the candidate is using the electrolyse, however, negligible penetration length of the electrical field in ionic water (Debye length) is a big obstacle. To overcome to this problem, using the nano-structures in electrodes is one suggestion. I personally interested to studying these structures and their role to increase the penetration length of the electrical field.
After the digital revolution in the human life, one of the most important issues and concerns is rising the artificial intelligence (AI) . Smart agents which are able to behave as human was subjects of many researches, philosophical discussion and science fiction novels. There are some progress in AI and some algorithms for it have been introduced. These algorithms, however, are implemented numerically as a software in a computer. Indeed robots and other intelligent devices are a mobile computer which run these algorithms in their processors. These algorithms are a digitized version of a fuzzy logic and fuzzy phenomena.
These is a field of research which its main focus is trying to implement the functionality of biological brain using electrical/electronic elements. Indeed this electronic circuits, which is called Neuromorphic circuit, provide hardware for next generation of AI.
One of a major progress Neuromorphic circuits designs took place when Memristor first discovered by D B. Strukov et. al. of Hewlett Packard in 2008. Memristor, as the forth elementary element of electrical circuit, first was predicted by Leon, Chau in 1971. It has lots on interesting features, which the most important of them is its memory; status of the memristor (its resistance) changes depending on applying voltage (current) on it, and will remain unit it receive a reverse voltage (current). Indeed the name of memristor is a combination of two word, memory and resistor. This memory feature make the memristor a suitable candidate for playing the synapse role, since its variable status could model the synaptic plasticity, which is a necessary phenomena in learning.
As it had been mentioned by Chau, since the memristor is one of the elementary elements, its functionality could not modeled using any combination of other elementary electronic elements. Therefore, it is necessary to add its governing equation of states to any electronic simulation package like SPICE.
In this field of research I am interesting to the minimal circuits including memristors to mimic the learning action happens in biological brain. In order to do that, we develop a simulator which able to simulate the behavior of electronic circuits include memristor. It is able to model a simple environment which our virtual creature is in, simulate the action by the environment to the creature, and provide the reactions of it. Our aim is to find and suggest the minimal circuit for learning how to response to a predefined action. In this way, it will be straight forwarded to make a intelligence creature for especial purposes.
Today, exponential growth of emerging technologies are changing and revolutionizing the way we live and work. Smart cities, as one of these novel concepts, are associated with various approaches based on the Internet of Things (IoT) and big data. IoT is the Internet of things which are not obviously computer themselves, however, have computer inside them. Indeed, IoT is the next evolutionary generation of the future information technology. Most of future devices which are utilized in daily life such as vehicles and home appliances will be a part of IoT.
Basically there are two models for IoT; centralized and decentralized IoT. In the centralized version of IoT, there is a control plane (server) which is responsible for managing the structure of the network as well as data flow through it. Besides, all network data will be store in centralized storages like network aria storage (NAS). Designing and managing of the centralized IoT is straightforward, it is the same as the conventional Internet. However, it faces the following problems;
One candidate to address the above mentioned issues is making the IoT decentralized. Here Decentralization means that absence on existence a central controlling center, all nodes based on their local situations as well as their functionalities make connection to their neighbors in the way that whole network be connected. Here, all decision makings, routine procedures, data storage and analyzing, and any other of network tasks are done by collective behavior of nodes. As a classic example of such network, one can refer to the Ad-hoc Mobile Network.
Decentralization of complex network is at the early stages of it development. Learning from the nature, especially collective behavior of herd animals, flocking behavior of birds, shoaling behavior of fish and so on will help us to improve our models for suggesting more efficient algorithms for decentralized IoT. In this context, I personally am interesting on developing models for collective behaviors of spices in herd. Furthermore, learning from these models, I would like to suggest new algorithms for making the IoT networks more decentralized.