Optimal Sensor Placement Based On Monkey Algorithm

Optimal configuration of sensors is a typical combinatorial optimization problem. At present, there are many methods to optimize the configuration of sensors, but they all have their own shortcomings. Monkey algorithm is a kind of intelligent bionic algorithm proposed in recent years, which is suitable for solving multi-variable and multi-peak function optimization problems. Using the monkey algorithm for optimal placement of sensor is one of the hot issues of wide attention and research for scholars at home and abroad at present. In thesis, on the premise of summarizing the research status and achievements at home and abroad, monkey algorithm is improved to adapting to the needs of sensor optimal placement. The contents of the thesis are as follows:(1) Introduce the significance of optimal placement of sensors, review the research status and achievements of monkey algorithm at home and abroad, give the direction of improvement, and establish a mathematical model for optimal placement of sensor.(2) In view of the problems of monkey algorithm that is limited in searching local optimal solution because of the random initialization distribution and fixed climbing step, an improved monkey algorithm is proposed. Modal assurance criterion is used as the objective function, normal distribution methods is used to construct initial population to enhance the diversity of monkeys, and adaptive changeable climbing step is used to improve the running speed and solution accuracy of the algorithm.(3) In view of the problems of monkey algorithm that jump interval fixation and information of outstanding characteristic monkey can't be transmitted, weed monkey algorithm is proposed. In order to improve the accuracy of the algorithm, based on the improved monkeys algorithm, adaptive jump process is used, and weed reproductive evolution and competitive survival mechanism based on adaptive degree is introduced to solve the problem of information transmission of outstanding characteristic monkey.(4) In the search for the optimal solution in the vicinity of the region, monkey algorithm can't avoid the existence of the blind spot, which easily lead to some optimal solution hidden in the area covered by the step size and miss the opportunity to reduce the ability of the algorithm to search for the global optimal solution. In view of the problems above, a monkey algorithm based on the behavior of honey bees is proposed. Based on the improved monkey algorithm, honey behavior of bee colony algorithm is introduced. After searching for all regions by using the swarm search mechanism, basic search for preliminary selected monkeys with monkey algorithm is conducted, which improve the search performance of the algorithm.(5) 8 test functions and common algorithm are used to test the 3 algorithms above, which prove that the solution accuracy and convergence rate of the improved monkey algorithm are both improved, and algorithm performance is also improved significantly.(6) Finite element model of gelatinize agency belongs to paper and yarn compounded bag bottom-pasting machine is established, and 3 kinds of algorithms are chosen to optimize the layout of the sensor, and also carry on the lateral comparison on the characteristics of them.