A novel systematic algorithm paradigm for the electric vehicle data anomaly detection based on association data mining

Abstract

Summary The electric automobile electric regenerative braking control strategy request achieves two goals, namely, enhances the entire vehicle the energy returns‐ratio as well as optimizes the pilot to the feel. When the former request brake the electric regenerative braking first, it can satisfy the brake request under the premise first to then use the electric regenerative braking recycling brake energy, and before, the latter request rational distribution the trailing wheel braking force uses in tracking pilot's deceleration intention. Under this basis, this paper proposes the novel systematic algorithm paradigm for the electric car data anomaly detection based on association data mining. With the continuous improvement of the intelligent vehicle and the continuous improvement of the function of the system, the system can be used as a hybrid vehicle detection and vehicle maintenance standard equipment and has broad application prospects. The effectiveness of the proposed system is verified through the experiment.

Machine learning‐based model for lithium‐ion batteries in BMS of electric/hybrid electric aircraft

“…The former ones are based on comprehensive measurements and their most prevalent disadvantage is the system's uncertainty. 8,9 Statistical and neural network approaches are two types of data-based models which are suitable for the systems with complex physical mechanisms. 10 However, if system uncertainties are not properly controlled in the algorithm of a databased technique during separate steps, they become compounded at each processing step and cause important issues in prediction.…”

“…Fault diagnosis methods usually are classified into two groups: data‐ and model‐based diagnosis. The former ones are based on comprehensive measurements and their most prevalent disadvantage is the system's uncertainty 8,9 . Statistical and neural network approaches are two types of data‐based models which are suitable for the systems with complex physical mechanisms 10 .…”

Special issue on multimodal information learning and analytics on cross‐media big data

“…The KNN (k‐nearest neighbors) algorithm is implemented for microblog classification in order to extract useful information about the real hazards, and an experiment is conducted to tune the parameters in KNN by comparison of outcomes of social media data analysis and the real typhoon situation. Wang and Wu propose the novel systematic algorithm paradigm for the electric car data anomaly detection based on association data mining. The correlation matrix is analyzed by Ji et al through component analysis method first, to detect the economic implication of the components of correlation matrix.…”