Self-adaptive recursive multi-step prediction method of demand torque for mechanical-electrical compound drive system

Abstract

The invention discloses an adaptive recursive multi-step prediction method for the required torque of an electromechanical compound transmission system, which relates to an adaptive recursive multi-step prediction method. The invention is based on an auto-regression model (ARX) with external input, and uses an adaptive recursive prediction algorithm to realize the online multi-step real-time prediction of the required torque of the electromechanical compound transmission system. In the prediction algorithm, the original driver's accelerator pedal signal and the actual demand torque signal calculated by signal transformation are used as the input of the model to realize the direct prediction of demand torque, thereby reducing the cumulative error of prediction. At the same time, two weight coefficients are introduced to ensure the accuracy and adaptability of the prediction algorithm. The invention realizes the online real-time update of the prediction model through the introduced self-adaptive weight coefficient, and completes the online prediction of the required torque information of the electromechanical compound transmission system while ensuring the real-time performance and accuracy of the prediction algorithm.

Description

technical field [0001] The invention relates to an adaptive recursive multi-step prediction method, in particular to an adaptive recursive multi-step prediction method for the required torque of an electromechanical compound transmission system. Background technique [0002] The electromechanical hybrid transmission system provides drive torque according to the driver's pedal information. When the vehicle's handling state changes, due to the hysteresis of the engine's response, the motor usually provides excess torque to ensure the vehicle's dynamic performance. When the vehicle is cold started, especially when the temperature of the battery is relatively low, excessive motor torque may cause the discharge current of the battery to exceed the maximum allowable current, thereby reducing the service life of the battery. If the required torque of the vehicle can be predicted, controlling the output power of the engine based on the predicted required torque can effectively impro...

AI Technical Summary

Problems solved by technology

The main problem in information prediction is: under the premise of ensuring the prediction accuracy, use an algorithm to apply as little effective historical data as possible, reduce the calculation amount of prediction and estimation, and realize real-time prediction of information at the same time

However, the problem with this multi-step forecasting algorithm is that the predicted data is used as the input of the next forecast, which will cause the accumulation of errors. When the number of forecast steps is large, the error will increase significantly.

However, the shortcomings of poor adaptability of the algorithm have not been improved accordingly

[0004] Through the analysis and comparison of the current forecasting algorithms, there are mainly the following shortcomings: (1) the application of forecasted data in multi-step forecasting iteratively causes the accumulation of errors, which leads to poor accuracy of the forecasting algorithm; (2) the forecasting process involves matrix The computational complexity of the calculation and solution is very large, which makes the real-time performance of the prediction algorithm poor, and it is difficult to realize online application; (3) the prediction algorithm uses a fixed prediction model. When the characteristics of the system change, the prediction algorithm is not self-adaptive and cannot be timely adjustments to the forecasting model, resulting in poor forecasting accuracy

However, for the electromechanical compound transmission system where the driving environment is constantly changing, its torque demand information is a kind of rapidly changing and complex information, so when the above prediction methods are used to predict the torque demand information of the system, the prediction effect obtained is not good. ideal

Images