Research On The Residential TOU Price Considering Electric Vehicles Charging

With the global energy shortage is becoming increasingly serious and bad weather especially smog weather is frequent,all fields of society has put forward higher requirements on environment protection,energy saving and emission reduction.Under this background,the electric vehicles with the advantages of energy saving,economy,low noise and zero emission are developing rapidly,and the amount of ownership increases rapidly.However,the randomness of the charging time of EVs will have a negative impact on the operation of the power grid,and threaten the reliability,economic and efficient operation of the power grid.TOU is the widely used demand response strategy,which can effectively improve the load curve,however,most studies focus only on guiding conventional load,the research of EV charging load is lack,so in this paper the research on TOU considering EVs charging is particularly necessary.First of all,based on the statistical data of residential trip,the Gauss mixture model is used to estimate the probability distribution over the end time of trip,and the conditional probability is used to indicate the correlation between end trip time and total distance,witch improve the modeling accuracy on lots of EV charging.The influence of EVs charging on the load curve,peak valley difference,variance,line active power loss and nodes voltage level of distribution network is analyzed through actual examples,which further illustrates the necessity of controlling and guiding the EVs charging orderly.Secondly,based on the characteristics of the conventional load and charge load,the selection of the electrical response characteristic model and witch the change rule and mechanism of is analyzed.The load and price elasticity matrix model is established for conventional consumption.The influence of load change rate,electricity price,load price and elastic coefficient value on unce rtainty is analyzed.For the EV charging load,under TOU price,using the consumer psychology theory to establish the demand response model to express the relationship between price difference and response degree,combined with the characteristics of each partition of the model,analysis of the impact of price guide on uncertain range is completed.And the probability distribution model of the response fluctuation range of the two kinds of load is given.Finally,in the SPSS software the classification of the annual load is completed through the K-means clustering algorithm,through the clustering center to find representative day,using representative daily consumption weights to modify the degree of membership of peak or a valley.Based on idea of ensemble classification,we can get the optimal threshold,which is used to determine the segmentation scheme suitably to the long cycle.The relationship between load variance and net loss is analyzed,which shows that reducing the load variance promotes energy saving and emission reduction.Considering the uncertainty of user load response on electricity price,we use chance constrained programming to establish the TOU price optimization model in order to minimize the load variance and difference of load peak and valley.Let the Monte Carlo sampling algorithm into the Constraint discrimination and fitness function calculation of genetic algorithm,then we get the optimal TOU price scheme.A simulation example shows that under the premise of ensuring the interests of all parties,the optimal TOU price scheme can optimize the load curve shape,improve the power system operation efficiency and promote energy saving and emission reduction.