Method for forecasting running load of hybrid electric vehicle

Abstract

The invention relates to a method for forecasting the running load of a hybrid electric vehicle, which comprises the steps: a speed sequence x0'(n) of the hybrid electric vehicle that runs a period of time is collected; a running period T of the hybrid electric vehicle is obtained according to the speed sequence x0'(n) and is used as the length of each sliding time window, the speed sequence x0'(n) is segmented at time by using sliding time windows, each sliding time window is divided into historical time T1 and forecasting time T2, data of the T1 time period is carried out dimension reduction by using DCT, a load level of the T2 time period is obtained by using fuzzy logic inference according to the data of the T2 time period, the data of the T1 time period after the dimension reduction and the load level of the T2 time period are used as training data to train a support vector machine, the speed of the running hybrid electric vehicle is collected, after the sliding time window is used for collecting the T1 time length, a speed sequence x0(n) is obtained, the x0 (n) is carried out dimension reduction by using the DCT to obtain x1(n), the x1(n) is classified by using the trained support vector machine, and load levels in the T2 time period is forecasted. The invention increases the forecasting precision and the generalization capability in the prior load forecasting method, reduces the computational complexity and realizes the dynamic forecast of the running load of hybrid electric vehicle.

Description

technical field [0001] The invention relates to the technical field of power control, in particular to the field of automobile control strategies, in particular to a method for predicting the running load of a hybrid electric vehicle based on an orthogonal cosine transform and a support vector machine. Background technique [0002] In recent years, greenhouse gas emissions and fuel consumption have led to the trend of global warming. In order to reduce this impact, cars have been given more and more attention. We hope that cars can have better fuel economy and less polluting emissions without compromising its performance. [0003] Currently, the following three methods are used to improve the performance of automobiles and the efficiency of engines. The first method is to change the structure of the traditional engine or add some improved components, but this method brings about a rather limited performance improvement. The second approach is to use electric vehicles (EV) ...

AI Technical Summary

Problems solved by technology

However, the existing algorithms have defects such as high dimensionality, large amount of calculation, complex calculation process, low precision, and large error.

Images