Development of an algorithm for generating driving cycles based on Markov chains

The driving cycle is a speed profile that reflects the characteristics of vehicle operation. They are used both at the design stage of the vehicle and for conducting field tests after the production of prototypes. The algorithm for obtaining a riding cycle based on Markov chains allows synthesizing a set of riding cycles that satisfy the required parameters. The synthesis process can be considered as lossy compression of information, which is not a disadvantage in such practical applications. In the synthesis process, one of the key tasks is to identify significant features or criteria for a set of source data that determine the success of synthesis. When using the mathematical apparatus of Markov chains, the stage of finding a set of evaluation criteria and determining their values is a separate task. A set of criteria can be synthesized based on the goals of a specific practical task and an existing data set. In the process of synthesizing the driving cycle, statistical criteria such as average speed, average acceleration and deceleration, and the variances of these values were used. The algorithm for synthesizing a riding cycle based on Markov chains has two main stages. At the first stage, a transition matrix is compiled and the values of statistical criteria are calculated, and at the second stage, the synthesis of the initial sequence takes place directly. However, such an algorithm also has disadvantages related to the use of randomness in the synthesis process. The distribution of deviations of the parameters of the elements of the initial data set has a significant effect on the convergence of the solution by the Markov chain method. The analysis of the distribution of deviations in the initial data set allows us to make an assumption about the probability of finding a successful solution by an algorithm based on Markov chains.