Traffic accidents had suddenness, complexity and severity. Efficient and reasonable dispatch of emergency rescue vehicles played a vital role in reducing casualties and property losses. The influencing factors of rescue route planning were introduced. The model was described from the single objective model and the multi-objective model. Combined with the characteristics of emergency rescue in sudden traffic accidents, the route optimization algorithm of emergency rescue vehicles was summarized from two aspects: precise algorithm and meta-heuristic algorithm. The current emergency rescue vehicle scheduling technology and development trend were summarized and prospected.

In order to formulate a scientific and reasonable scheme for highway differentiated toll collection, the background, realization manners, related theories and key technologies are systematically described, and the cases upgraded in Guangxi, Tianjin and Hebei are briefly introduced with design essentials and application effects, and outlooks on the research trend of highway differentiated toll collection are given.

To precisely distinguish progressive and stable mild cognitive impairment (MCI). The projection matrix was learned from Alzheimer′s disease samples and normal control samples. The supervised graph regularization was used to optimize the local nearest neighbor relationship of the samples. Based on the projection matrix, the spatial transformation of the MCI samples was carried out to extract the discriminative features of progressive and stable MCI. The proposed features were fused with the scores of Mini-Mental State Examination and apolipoprotein E4. The SVM classifier was trained using the fused features for the MCI classification. The experiments were conduct on the Alzheimer′s Disease Neuroimaging Initiative (ADNI) database. The classification accuracy reached to 73.33%. Compared with the existing approaches, the proposed method significantly improved the classification accuracy, sensitivity and specificity.

To address the problem that the electricity consumption behavior of market-based users was complex and variable, and the laws of electricity data were difficult to be accurately characterized, a market-based user electricity data fitting method considering behavioral characteristics was proposed. The K-means clustering algorithm was used to classify the electricity consumption behavior of customers and clarify the typical characteristics of each type of customers; the neural network model based on orthogonal polynomials was constructed, in which the neural network weight coefficients were trained by gradient descent algorithm and the orthogonal polynomials were Chebyshev polynomials, Hermite polynomials, Legendre polynomials and Laguerre polynomials. The simulation analysis was carried out using the electricity data of users in Jinan, Shandong Province, and four different orthogonal polynomials were used to fit the electricity data and calculate the evaluation indexes for different categories of users, so as to summarize the most suitable fitting methods for users with different behavioral characteristics. The simulation results showed that the power data fitting effect differed significantly among different implementation methods for similar users, and the fitting accuracy of the neural network models based on Hermite polynomials and Laguerre polynomials was relatively high, but the polynomial models with the highest power data fitting accuracy for different categories of users were different. Selecting the corresponding orthogonal polynomials to form a neural network fitting model according to the type of electricity consumption behavior was an effective way to achieve accurate fitting of user electricity data.

Combined with back propagation (BP) neural network and genetic algorithm, a high-quality and low-cost process parameter optimization method was proposed. The flow rate of two blowtorch gases (H₂-1, H₂-2, H₂-3, Ar-1, Ar-2, Ar-3, O₂-1, O₂-2, SiCl₄) during the preparation of optical fiber preform core layer by vapor axial deposition (VAD) was selected as the input variable. The quality of the prepared optical fiber preform core layer was taken as the output variable in the established neural network model. The trained neural network model was combined with the genetic algorithm with global optimization ability, and the high quality core layer of optical fiber preform was taken as the optimization objective to obtain high quality gas parameters. The obtained parameters were selected at low cost, and the high quality and low cost process parameters were obtained. The experimental results showed that compared with the manual optimization results before optimization, the prepared optical fiber preform core layer met the high quality requirements and the cost was reduced by 22.19%.