| Emergency response to nuclear accidents is the basis for ensuring the rapid development of nuclear power.In order to reduce the radiation damage to the affected people in the plume emergency plan area after a serious nuclear accident,the evacuation path of the plume emergency plan area is affected by the change of radionuclide distribution and the differentiation of evacuation demand,which leads to the increase of radiation risk in the evacuation path,and combining with the existing evacuation strategy of nuclear emergency response,the whole evacuation path planning process is divided into three phases,namely,the affected people going to evacuation gathering point,the emergency vehicle traveling to evacuation gathering point and the emergency vehicle traveling from evacuation gathering point to resettlement point.The whole evacuation path planning process is divided into three phases: the affected people go to the evacuation assembly point,the emergency vehicles travel to the evacuation assembly point,and the emergency vehicles travel from the evacuation assembly point to the resettlement site.Taking the fuzzy mathematics and the intelligent optimization algorithm as the basis of the evacuation path modeling and solution algorithms,a multi-stage multi-constraint evacuation path optimization model is established,and dynamic path planning algorithms for different evacuation phases are designed,which provide technical support for the evacuation path of nuclear emergency response.The specific research contents are as follows:(1)During the evacuation process,the different departure times of both the affected persons and the emergency vehicles may cause changes in the evacuation time,which is closely related to the accumulated radiation dose in the evacuation path.The existing evacuation path cumulative radiation dose calculation model does not take into account the impact of path passage time.Therefore,in order to address the problem of increasing the cumulative radiation dose of the affected persons during the evacuation process due to the change of evacuation path passage time caused by the different departure moments in the event of a nuclear accident,the fuzzy time-based evacuation path cumulative radiation dose calculation model is established based on the fuzzy theory and the dynamic fuzzy network to provide a theoretical model for the nuclear emergency evacuation path optimization.Based on the fuzzy theory and dynamic fuzzy network,the cumulative radiation dose calculation of the evacuation path with fuzzy time is established to provide a theoretical basis for the optimization model of the evacuation path in nuclear emergency.(2)Aiming at the problem of incomplete information of the evacuation path model and reduced effectiveness of the evacuation path caused by the influence of uncertain factors such as the distribution of radiation dose in the first stage and the number of people in the evacuation assembly point,the multi-constraint evacuation path planning model is established by considering the influence of the change of radionuclide concentration in the environment on the evacuation path with the conditional constraints of the evacuation time and the capacity of the evacuation assembly point,and the minimization of the cumulative radiation dose on the evacuation path as the objective function.A multi-constraint evacuation path planning model was established,and a hybrid path optimization algorithm of elephant colony-ant colony was designed.The algorithm is based on the ant colony algorithm and minimizes the cumulative radiation dose on the evacuation path by introducing the elephant colony algorithm,speed update and "hybridization operator".Finally,the proposed algorithm is combined with the ant colony algorithm,the max-min ant algorithm and the greedy ant colony algorithm to conduct 50 independent repetitive experiments on raster maps with different levels of complexity.The results show that the average cumulative radiation dose of the evacuation path obtained by the proposed algorithm compared with the other three algorithms is reduced by 15%,14% and 15%,the average number of iterative convergence is reduced by 67%,44% and 48%,and the stability is improved by 92%,91% and 88%,respectively,which proves the reliability and feasibility of the proposed algorithm.(3)To address the problem of increased evacuation cost and radiation risk caused by dynamic changes in the number of people waiting at the evacuation rally point and radiation dose distribution during the second-stage evacuation process,a multi-constraint path optimization model for the evacuation of emergency vehicles to the evacuation rally point is established with the number of people waiting to be rescued,the evacuation time,and the capacity of the emergency vehicles as the constraints,and the minimization of the cumulative radiation dose in the evacuation path as the objective function,and a fuzzy genetic algorithm is designed with the environmental At the same time,a fuzzy genetic algorithm is designed to consider the influence of changes in the concentration of radionuclides in the environment on the evacuation path.The algorithm dynamically adjusts the crossover probability according to the fitness value,and improves the convergence performance of the genetic algorithm while reducing the influence of the parameters on the genetic algorithm by introducing strategies such as elephant swarm optimization,roulette wheel selection,and large neighborhood search.Finally,public test cases and road network structures are selected to verify the feasibility and effectiveness of the proposed algorithm.And the proposed algorithm is experimentally compared with six heuristic hybrid algorithms,including adaptive genetic algorithm,genetic-large-scale domain hybrid algorithm,and hybrid genetic algorithm,and the average error rate of the proposed algorithm is about 10-4,that of adaptive genetic algorithm is about 10-2,and that of the other five hybrid algorithms is about 10-3;the stability of the proposed algorithm improves by about 68%,and the average iteration improves by about 68%,and the average number of iteration convergence is reduced by about 68%,which effectively improves the reliability and feasibility of the evacuation path.(4)Aiming at the problem of increasing cumulative radiation dose in the evacuation path due to different departure moments of emergency vehicles in the third stage.Taking the evacuation time and the capacity condition of the resettlement site as constraints,a multi-constraint evacuation path planning model that minimizes the cumulative radiation dose is established,and a hybrid ant colony algorithm is proposed.The algorithm introduces the heuristic ideas of simulated annealing and A* algorithm in the iterative process of ACO algorithm,and introduces the Pareto sorting method,and adds the effect of distance on the pheromone increment into the pheromone updating method of ACO algorithm,so as to improve the global searching ability of the algorithm and the feasibility of the evacuation path.Finally,the proposed algorithm is compared with the other four algorithms,and the results show that the average cumulative radiation dose value is reduced by 42%,21%,and10% compared to the ACO,the max-min ACO system,and the A* algorithm,respectively,and the standard deviation of the proposed algorithm is 0 in 50 repetitive experiments,which indicates that the proposed algorithm has a good stability,and the hybrid algorithm is further compared with the improved ACO algorithm and shows that further compared the hybrid algorithm with the improved ACO algorithm,it shows that the proposed algorithm has stronger applicability compared to the other four algorithms.Finally,assuming that a nuclear accident occurs in the Hualong-1000(HPR1000)reactor type of Fuqing Nuclear Power Station as an example,based on the time series of the release rate of the key radionuclide source items and the local meteorological conditions,the established evacuation model and path planning algorithm are utilized to obtain a solution for the evacuation of emergency response to nuclear accidents,thus verifying the feasibility of the model and the algorithm designed in this topic,and providing theoretical basis and technical support for the decision-making of the evacuation after the occurrence of a serious nuclear accident. |
PDF Full Text Request