Detailed hourly energy simulation was conducted for office buildings in the five major climate zones - severe cold, cold, hot summer and cold winter, mild and hot summer and warm winter - in China using multi-year (1971-2000) weather databases as well as typical meteorological years (TMY). The primary aim was to compare the energy simulation results from the TMY with those from individual years and their long term means. A total of 154 simulation runs were performed. Building heating and cooling loads, their components and energy use for heating, ventilation and air-conditioning were analysed. Predicted monthly load and energy consumption profiles from the TMY tended to follow the long term mean quite closely. Mean bias errors ranged from -4.3% in Guangzhou to 0% in Beijing and root-mean-square errors from 3% in Harbin to 5.4% in Guangzhou. These percentages were not always the smallest compared with the 30 individual years, however, they are at the lower end of the percentage error ranges. This paper presents the work and its findings. (c) 2007 Elsevier Ltd. All rights reserved.