Numerical Simulation Of Two Kinds Of Engineering Problems

This paper consists of two parts of relatively independent research content.In the first part,we study the numerical solution of the steady neutron transport equation in a slab with general boundaries.In this section,we introduce the steady neutron equation,and then get the simplified calculus equation in a slab.We turn the calculus equation into an equivalent integral equation by considering the integral of the neutron equation as a whole term.Then the integral term is discretized via the composite trapezoidal rule with equidistant nodes.Therefore,the numerical solution of the integral equation is obtained.But remember that the obtained numerical solution is the whole term of original neutron transport equation.We must replace the whole integral term in the original neutron transport equation with the obtained numerical solution.Finally,we get the numerical solution of the original neutron transport equation.In the second part,we reduce the cost of the experiment by numerical simulation.As one of the most widely used medications in the world,acetaminophen has good effect in treat pain and fever and is typically used for mild to moderate pain.To reduce the experimental cost of the determination of the solubility of acetaminophen in n-propanol and ethyl acetate,this paper uses BP(Back-Propagation)neural network to model the experiment.We use the previous traditional experimental data as training data of the BP neural network.And the training data is trained by accelerated cross validation.With the existing training data,the optimal double-layers BP neural network is selected by comparing the different BP neural network model fitting effects.The traditional complex solubility determination process is turned into a simple pattern mapping process.The average training optimization time of the double-layers BP neural network model is only 8s.Different from the traditional solubility determination process,the solubility of acetaminophen in n-propanol and ethyl acetate is determined by using optimized BP neural network model in this paper.The average determination time was 28.75?s,and the average square root error of the determination is only 1.2%.So using the BP neural network model in determination solubility can greatly reduce the experimental expense.