Development of Wheel Loader Duty Cycle Using Hybrid Markov Chain and Genetic Algorithm

Heavy diesel machinery has a significant contribution to the production of air emissions in large and industrial cities. However, little attention has been paid to this issue, and no appropriate action has been taken to address it. Determining the proper duty cycle can be an effective step in reducing emissions and fuel consumption of these vehicles. The duty cycle includes the driving cycle along with the vehicle operating cycle and makes sense for vehicles that have a specific task. In this article, 80,000 experimental data of the wheel loader (WL) in diggings site by global positioning system (GPS) is collected. After filtering, the data is converted to micro-trips and divided into four clusters using the k-means method, and finally, the Markov matrix is produced. The genetic algorithm (GA) is performed to identify the best combination of micro-trips. In this research, two types of cycles, fuel consumption, and emission were derived using the ADVISOR software. The first cycle is the WL operation at the excavation site, and the second cycle is the short loading cycle. Tehran WL duty cycle has higher fuel consumption, nitrogen oxides (NOx), and hydrocarbon (HC) emission than Environmental Protection Agency (EPA) WL and European non-road transient cycle. While the amount of carbon monoxide (CO) produced in the Tehran WL duty cycle is lower than all EPA duty cycles but higher than the European non-road transient cycle. Tehran WL duty cycle fuel consumption is 265% more than the European non-road transient cycle and 86.12% higher than EPA WL typical operation 1 duty cycle. These differences show that each country’s cycle is only specific to the same country. A derived WL cycle can be used on vehicles with a similar application such as forklift, aircraft support, and forestry equipment.