Energy management method based on working condition identification

Abstract

The invention discloses an energy management method based on working condition identification. The method comprises the steps of carrying out the working condition classification of four typical working conditions through employing a K-means clustering algorithm, and calculating the clustering center of the four typical working conditions, establishing an equivalent fuel consumption minimum control strategy according to the automobile parameters, combining a multi-island genetic algorithm and a sequential quadratic programming algorithm, and using a Task component in Isight software to construct a combinatorial optimization model, performing integrated optimization on the variables of the equivalent fuel consumption minimum control strategy and the clustering centers of the four typical working conditions by using a combinatorial optimization model to obtain a final clustering center and corresponding optimal equivalent fuel coefficients under the four typical working conditions, and extracting a section of random driving working condition characteristic parameters, performing working condition classification, and performing optimal power distribution according to the current typical working condition type and the optimal equivalent fuel coefficient. According to the method, the identification of the working condition characteristic parameters can be realized, and meanwhile, the defects of a single local or global optimization algorithm are overcome.

Description

technical field [0001] The invention relates to the technical field of hybrid electric vehicle energy management, in particular to an energy management method based on working condition identification. Background technique [0002] Compared with instantaneous optimization, global optimization, and adaptive energy management strategies based on working condition identification, the rule-based energy management strategy has relatively low fuel economy. Because the instantaneous optimization is optimized for a certain driving condition, it has poor adaptability to other working conditions. At the same time, the potential of instantaneous optimization to improve fuel economy is insufficient; the global optimization algorithm is complex and computationally intensive, and it needs to know the future road conditions. The practicability is poor; the adaptive energy management strategy based on working condition identification can better improve the fuel economy of the whole vehicle....

AI Technical Summary

Problems solved by technology

[0002] Compared with instantaneous optimization, global optimization, and adaptive energy management strategies based on working condition identification, the rule-based energy management strategy has relatively low fuel economy.

Because the instantaneous optimization is optimized for a certain driving condition, it has poor adaptability to other working conditions. At the same time, the potential of instantaneous optimization to improve fuel economy is insufficient; the global optimization algorithm is complex and computationally intensive, and it needs to know the future road conditions. The practicability is poor; the adaptive energy management strategy based on working condition identification can better improve the fuel economy of the whole vehicle. The commonly used working condition identification algorithms are neural network, support vector machine, etc. Poor computing power makes it difficult to directly apply these operating condition recognition algorithms to actual vehicle energy management strategies

In addition, optimizing the design of the hybrid system by manually adjusting parameters or writing an optimization algorithm requires a large and cumbersome workload, and the accuracy is difficult to guarantee.

Images