Research Experiences

2013.8 - Present Pulse Position Coded PDU (PPCP):

  • A brand-new link layer architecture proposed for energy-efficient wireless sensor networking (MAWSN). The key idea is to encode a PDU in terms of silence duration between two sets of delimiter pulses, whose positions are modulated based on the value of the PDU. This new link architecture offers a suitable networking technique for severely energy-starved networks, and achieves significant energy-savings by using lesser amount of bits/pulse transmissions, and by eliminating long multi-bit preambles, which are normally used in traditional packets.
  • The core research contributions are: 1) develops architectural solutions to a subset of the challenges from transmitting, receiving, idle listening, hidden collision and medium sharing. 2) proposes multi-PPCP PDU-based solutions for a multi-access wireless sensor network, in which PPCP provides various PDU formats for low-energy sensors to operate in either transmitter or receiver mode. 3) a concept of Pulse Slot is used for accommodating pulse shifts and for achieving clock synchronization between transmitter and receiver without overhead of preamble during a PPCP transmission. 4) Flexible Base Digit Separation (FBDS) and the corresponding concrete mathematical principle are developed to mitigate the transmission delay, which uses a variable coding base  to shrink the value. 5) a systematical error detection theory is established based on the proposed PPCP architecture. 6) a hardware platform—Jurassic—is designed for the purpose of PPCP implementation, and this platform has been elegantly operating in Green House of MSU Agricultural Department for more than six months.

 

 

2007.1 - 2009.7 Energy-saving Technology in Electrical Transmission Systems, A Chinese National 863 Key Program

  • To reduce the energy consumption of the electric transmission system, the program developed an electric transmission and control system for high-power permanent magnet synchronous motors (PMSM) by carrying out the off-line simulation, the real-time hardware-in-the-loop simulation and the physical system experiments
  • Adopted the energy-saving PMSM as the actuator
  • Improved the circuit topological structure of the electric transmission system by applying the power factor correction and the soft switching technology to the system design
  • Designed cost functions, optimal control laws, calculated control rate, and developed control strategies by applying vector control, fuzzy control, and Fourier integral-adaptive control techniques to minimize the energy consumption

Responsibilities 

  • Analyzed the technical difficulty of the project by carrying out the engineering analysis (from a technical feasibility point of view), the material collection on fuzzy control and adaptive control, the evaluation of the technical feasibility, and managed the project schedule
  • Studied the limitation of the existing control solutions on energy-saving, and presented the PMSM-oriented vector control algorithm
  • Developed the fuzzy control and Fourier integral adaptive control methods to correct the power factors and to achieve the system steady-state precision and the transient performance
  • Created the target functions and the optimal control methods with the aim to minimize the energy consumption

 

2009.4 - 2010.10 The Fuzzy Control Approach of Nonlinear Random System under Network Environment, a project funded by National Natural Science Foundation of China (No. 60774048)

Responsibilities 

  • Analyzed the characteristics of the specific nonlinear system under network environment. Carried out fuzzy identification for the nonlinear system, ie. approached the target nonlinear system within a compact set by the fuzzy identification method and built the T-S fuzzy model for the non-linear system
  • Carried out the system modeling by treating as an independent Bernoulli process the data packet loss which is introduced from network transmission, and set an upper bound for the maximal packet loss ratio
  • Solved the random disturbance problem by introducing robust control design, and designed a new robust adaptive controller which could achieve asymptotic stability of the closed-loop system and a better transient state performance

 

2011.1 - 2012.3 System Monitoring and Fault Diagnosis of Complex Control Process based on Data-driven, a project funded by National Natural Science Foundation of China (No. 61034005)

Responsibilities 

  • Set up a data set Xm×n (m is the number of sampling points and n is the number of sensors) under normal production conditions, then built a statistical model for the complex control process; Carried out standardized processing on the data matrix, namely standardized the every column vector in Xm×n and got a standardized data matrix X
  • According to the principle of CPV (Cumulative Percent of Variance), carried out space decomposition and dimension reduction on data matrix X, and got the first k-dimension linear independent vector X 1 =[ x 1 , x 2 ,…, x k ] to constitute the principal space or loading vectors, and the last (n-k)-dimension vector X 2 =[ x k+1 , x k+2 ,…, x nto form the residual space
  • Built PCA (Principal Component Analysis) statistical models respectively in the two sub-space in order to implement the variable detection in low-dimensional sub-space
  • When the statistical data of the established model went beyond the control limit, it indicated that there may be some faults or changes in the complex control process. Under these circumstances, system detecting was carried out to further determine whether the system has real faults or not
  • If faults really do exist in system process, implement characteristics extraction aiming at the sub-system where the faults happened, and diagnose fault causes. Designed adaptive controllers to get rid of or, for the most part, minimize the impacts on stability of complex control process and transient performance brought by specific faults