In order to develop detail survey and integrative evaluation of wind energy resource, the consistency corrections, objective homogeneity tests (such as SNHT, Potter, and CUSUM), and homogeneity adjustment of yearly mean wind speed series were carried out during 1970-2009 from four meteorological stations which represent wind areas according to their historical metadata features. During the period of nonautomatic wind recorder, how to build the 10min interval yearly maximum wind velocity series was also explored elementarily based on observed 2min interval wind data. The results show as follows: 1) The yearly mean wind speed time series were inhomogeneous，which was caused by the change of their surroundings or other uncertain factors at Burqin, Naomaohu, and Hongliuhe stations. From the values of the SNHT homogeneity adjustment and the buildings surrounding the stations, the most largely environment influence occurred at Burqin station where the time series broke off three times and the mean accumulated adjustment value reaches 09 m·s-1 or so. There are two discontinuous points at Naomaohu station, one of which concerns location moving, and that of adjustment value is 0.1－0.2 m·s-1. There exists one discontinuous point at Hongliuhe station, which also concerns location moving like Naomaohu station, and that of adjustment value is 0.1－0.2 m·s-1. 2) On transforming the yearly maximum wind velocity of 10min interval from that of observed 2min interval under the control of some critical wind speed, it is acceptable to adopt the experiential formula fitting to the northwestern part of China at Shisanjianfang station, while it is better to employ their mean portion correlation at the other three stations.