Research On Predication Model For Estimating Bus Emissions At Bus Stops

It is well known that compared with the road sections,bus stops are densely populated points.At bus stops,the bus condition changes frequently and the discharge is larger.At present,the analysis on bus emissions at the bus stops have been focused on the influencing factors.However,there are no suitable bus emission models that can be used to estimate the emissions generated near bus stops.In this paper,through studying vehicle emission model and bus emission characteristics at bus stops,a method has been built to calculate vehicle fuel emissions.The Portable Emission Measurement System and Global Position System(GPS)devices have been used to collect the emission data of CO2,CO,NOx,HC and the operating mode data of buses.Based on the operating mode data,the study zone can be calculated by using dot density method.Then,the factors can be found out which influence the emissions generated near bus stops.Secondly,by combing vehicle emission models which are based on Vehicle Specific Power(VSP)with emission data and operating mode,databases of vehicle emission rate and operating mode distribution have been established.In this paper,the diesel bus is chosen as the main research object.According to the distribution characteristics of VSP at bus stops,the VSP Bin partition method is proposed.When both of the velocity and acceleration are equal to zero,the VSP Bin is defined as Bin19.By using the method of mean value estimation and actual value fitting,a real-time forecast model and a forecast model are established.Finally,the test is conducted on the proposed model by using the SPSS software.The real-time amount of emission and total mass of emission can be well predicted with the VSP Bin division method at the bus stop and the fitting model of real emission rate.The total mass of emission has the best prediction.In the four kinds of gases,the prediction of C02 is the best,CO and NOx are the second,and the prediction effect of HC is poor.