Research And Application Of Scheming The Working Routes Of Sprinklers

In our daily life, the scheme of working routes of sprinklers is usually decided manually, depending on the work experience. This kind of decision isn't scientific, regarding to reducing the driving cost, improving the efficiency, and so on. After investigating and analyzing the material, we can find that the scheme of working routes of sprinklers can be abstracted into a special kind of vehicle routing problem for researching, and we can call it Complicated CARP(CCARP for short). Compared of general CARP, the complexity of CCARP mostly lies in: diversity of roads, where exist two-way roads, one-way roads and slopes. And next is diversity of demand of roads, where exist sprinkling in two-way roads separately or simultaneously or just by one side. Then, it's necessary to consider turning prohibition of vehicles as regard to transport condition. Then, vehicles have different loading capacity of water. Finally, the exact amount of road for sprinkling is decided according to concrete condition.CARP has been proved to be a kind of NP-hard problem, and the exact algorithm is rather hard to be applied for solving the practical large-scale CARP. Nowadays, it is popular that to construct efficient heuristic algorithms to get the approximate solution. In this paper, an improved genetic algorithm which combining the niche technique is adopting for solving the problem presented above. With the assistance of relational department, an application system which can solve the practical problem has been designed and developed. During the researching procedure, the main contribution of this paper lies in the aspects shows as follows:①After the survey of working condition of sprinklers, we have collect plenty of useful data. Under the consideration of the restrictions mentioned above, based on some reasonable premises and hypothesis, we improved and extend the mathematical model of general CARP, and establish the model which meets the problem presented in this paper.②According to the model established, we utilize an improved genetic algorithm to solve the problem. In the algorithm, we adopt a novel constructional mechanism of chromosome to form a population in the initialization of genetic algorithm, and introduce the niche technique to avoid the occurrence of early convergence. Thus each population can evolve respectively with different crossover probability and mutation probability, and also there establish some survival rules for populations to determine their existence or disappearance. In the inner evolution procedure of each population, there adopts local search strategy to accelerate the optimization process.③In this paper, we design and develop an application system for scheming the working routes of sprinklers. The system can provide many functions including modification of roads data, configuration of vehicles information and the turning situation. During the experimentation process, we make a lot of comparative test making use of the real data from the survey, and find a set of optimal configuration of parameters of the algorithm. Finally, we get an optimal scheme utilizing the system. Compared to the manual one, the former gains a saving of nearly 15.85%, and the computation process finish within rather perfect time, which adequately shows the practicability and efficiency of the algorithm.This paper shows a first attempt to research scheming of working routes of sprinklers as the application background of CARP, and finally gains rather good production. The application system presented produces the working scheme with scientific method, thus not only achieves economical route, but also take the rationality of route partition into consideration to some extent. It can help for raising the level of scientific management of department, and has great value in application and potential for replication.