Data-driven fast global SOC planning method

Abstract

The invention discloses a data-driven fast global SOC planning method. The method comprises the following steps: firstly, constructing a fast global SOC planning model based on a neural network; then, randomly generating a journey through simulated navigation, generating journey traffic information and detailed working conditions of the journey, solving an SOC optimal solution of the detailed working conditions by adopting a dynamic programming algorithm in combination with the current SOC value, and recording the current SOC value and the journey traffic information as input and the SOC optimal solution as output as a group of training data; after enough training data is obtained, training the fast global SOC planning model to obtain a trained fast global SOC planning model; and finally, when SOC planning needs to be carried out on the vehicle, inputting the current SOC value of the vehicle and travel traffic information provided by navigation into the trained rapid global SOC planning model. By adopting the method, the SOC can be quickly and globally planned according to the real-time travel traffic information, the advantage of a high-capacity battery of the plug-in hybrid electric vehicle is fully played, and the use oil consumption of a user is reduced.

Description

technical field [0001] The invention relates to the field of energy management of plug-in hybrid electric vehicles, in particular to a data-driven fast global SOC planning method. Background technique [0002] The energy management strategy can improve the fuel economy and emission performance of the vehicle without sacrificing the power demand by coordinating the distribution of power demand among multiple energy sources, which plays a pivotal role in the control of plug-in hybrid electric vehicles . Existing energy management strategies are generally divided into rule-based and optimization-based strategies. The rule-based strategy control has strong timeliness, but its applicability to vehicle parameters and working conditions is poor, and the formulation of rules requires rich engineering experience. Optimization-based strategies are divided into global optimization strategies and instantaneous optimization strategies. The global optimization strategy can provide the ...

AI Technical Summary

Problems solved by technology

[0003] The simple SOC planning method uses a linear decline model based on mileage. Although this model can improve fuel economy to a certain extent, there is still room for improvement

The online solution of SOC and the establishment of a fitting function between SOC and operating condition characteristic parameters are the two mainstream methods for SOC planning by means of real-time traffic information to generate travel conditions. There are real-time solutions and whether they are (approximately) optimal solutions The problem

Images