Cascaded Dual-Observer-Based Decoupled Estimation of Mass and Track Gradient for Permanent-Magnet-Driven Electric Monorail Cranes

Precise data regarding the overall mass of the machinery and the gradient of the track are crucial for optimizing the control of monorail cranes and enhancing energy efficiency. Within the context of electric monorail cranes (EMCs), accurately estimating the total mass of the machinery and the track gradient poses a formidable challenge. This challenge arises from the strong coupling between the overall mass of the machine and the track gradient, the robustness of parameter estimation methods under varying operational conditions, and the generalizability of the algorithm to real-world operations and rail scenarios of EMCs. To address these challenges, this paper proposes a novel parameter estimation scheme that comprehensively considers the impact of parameter coupling relationships and multiple influencing factors in the transportation scenarios of EMCs under actual working conditions. First, to overcome measurement difficulties induced by strong coupling between the EMC mass and track gradient, a decoupling estimation method based on cascaded dual observers is proposed to jointly estimate the two states. Secondly, to mitigate track slope estimation errors under complex track types and diverse operating conditions, an enhanced immune optimization algorithm, integrating a Weibull function and Levy flight mechanism, in conjunction with an unscented Kalman filter (UKF), is developed. Furthermore, to achieve high-precision and stable parameter identification results, a Weibull dynamic forgetting factor is incorporated into the RLS algorithm, leading to the design of a WDFF-RLS estimator. Finally, real vehicle experiments were conducted on complex tracks at the test site to validate the accuracy and robustness of the proposed estimation method.