Kilogram, the unit of mass, is the last one of seven base units in International System of Units (SI) which is still defined and kept by a material artifact. 1 kg is defined as the mass of the International Prototype of the Kilogram (IPK) kept at the Bureau International des Poids et Mesures (BIPM) in Paris. One of the major disadvantages of this definition is the fact that the amount of material constituting the IPK changes with time. Because a more stable mass reference does not exist, the variation of IPK is completely unknown so far. The International Committee for Weights and Measures (CIPM) recommended redefining the kilogram by fixing the numerical value of the Planck constant h and called on every national metrology institute to study the measurement of the h. To avoid possible system errors from one method, more experiments especially based on different principles are expected and encouraged for the final determination of the Planck constant. The CCM required that at least three consistent results should be obtained before the redefinition. Since 1970 s, the Kibble balance (also known as the Kibble balance) experiment has been used by a number of national metrology institutes such as NPL, NIST, METAS, LNE and BIPM. The IAC including the PTB, NMIJ and NMIA used the XRCD method to measure the Avogadro constant. To make contribution to the redefinition of kilogram, the National Institute of Metrology of China (NIM) proposed a joule balance method in 2006, which is also an electrical way but different from the watt balance method in that the dynamic phase is replaced with a static phase to avoid the trouble in the dynamic measurement. The progress of these approaches and the current situation of the redefinition of the kilogram are presented in this paper. In 2013, a model apparatus was built to verify the principle of the joule balance. Then NIM started to build its new joule balance aiming to obtain an uncertainty of 10-8 level since 2013. In Dec. 2016, the new apparatus was built and could be used to measure the Planck constant h in vacuum. In May 2017, the measurement result was submitted to the Metrologia and accepted by the CODATA TGFC as the input data. However, the measurement result has an uncertainty bigger than 10-8 and was not used for the final determination of the h value. At present, the joule balance group of NIM, together with the Harbin Institute of Technology, Tsinghua University and China Jiliang University is still making great efforts to improve the joule balance apparatus. The uncertainty of 10-8 level is expected to be achieved in the next two years.