Research On Forecast Model Of Forest Fire Emergency Fire Fighting Helicopter Demand Considering Sample Capacity

The frequent occurrence of forest fires in my country has caused problems such as declining forest accumulation,air pollution,soil erosion,and loss of life and property to varying degrees.Therefore,it is very necessary to carry out emergency rescue of forest fires to minimize the losses caused by disasters.According to the “Thirteenth Five-Year Plan for the Construction of the National Emergency Response System”,the aviation firefighting and rescue system will be developed during the “Thirteenth Five-Year Plan” period to effectively improve the level of firefighting and disaster relief.After a forest fire occurs,the rapid and effective prediction of the number of helicopter demand can provide support for efficient helicopter dispatch,thereby improving the efficiency of fire rescue resources.Therefore,it is necessary to carry out forest fire emergency fighting helicopter demand forecasting model research.Based on the difference in sample size,this paper constructs a forecast model of forest fire emergency fire helicopter demand by classification.First,use the method of constructing the system effectiveness evaluation index system to sort out the forest fire emergency firefighting helicopter demand forecast index system.According to the index system construction principles such as hierarchical,regional,and quantifiable,from emergent events,forest fire factors and aviation In three aspects of fire protection factors,a three-level index system for forecasting the demand for forest fire emergency fire-fighting helicopters was established.Secondly,on the basis of summarizing the current research status at home and abroad,in view of the large amount of historical forest fire data,and under the premise of ensuring the accuracy of demand forecasting by the RBF neural network,the improved grey correlation method is used to reduce the influencing factors.At the same time,the improved singular value decomposition method is used to "compress" the training data,and then a forest fire emergency firefighting helicopter demand prediction model based on the improved gray correlation method-improved singular value decomposition method reduction and RBF neural network is proposed;for forest fire historical data In the case of small quantities,the helicopter demand forecasting problem is regarded as a pattern recognition problem,and the pattern recognition problem is transformed into a multi-criteria decision problem by constructing an index fuzzy segmentation model.The weighted TOPSIS method is further used to realize the emergency fire helicopter demand forecasting,and then proposed based on The forecast model of forest fire emergency fire helicopter demand based on fuzzy segmentation of indicators and weighted TOPSIS method.Finally,the feasibility and rationality of the model are tested by collecting samples of forest fire data,using example analysis and comparative analysis according to different situations.