基于改进QGA的T-S模糊控制器设计

Design of Takagi-Sugeno Fuzzy Controller Based on Improved Quantum Genetic Algorithm

利用基于量子位测量的二进制量子遗传算法(QGA)对连续问题进行优化时,频繁的解码运算严重降低了优化效率。针对该问题,提出一种基于量子位相位编码的QGA。该算法直接采用量子位的相位对染色体进行编码,利用量子旋转门实现染色体上相位的更新,通过Pauli-Z门实现染色体的变异,由于优化过程统一在[0,2π]n空间进行,因此对不同尺度空间的优化问题具有良好的适应性。以单级倒立摆T-S模糊控制器参数的优化设计为例进行仿真,证明该算法在搜索能力和优化效率方面的优势。

Frequent decoding operations severely reduce the optimization efficiency when the binary Quantum Genetic Algorithm（QGA） based on qubits measure is applied to the continuous space optimization.Aiming at the problem,an improved QGA based on phase of qubits encoding is proposed.The chromosomes are encoded by the phase of qubits,updated by quantum rotation gates,and mutated by quantum Pauli-Z gates.As the optimization process is performed in,the algorithm has good adaptability for a variety of optimization problems in the different scale space.Taking parameter optimization of Takagi-Sugeno（T-S） fuzzy controller of single level inverted pendulum as example,the simulation results show that the algorithm has advantages in search ability and optimize efficiency.