Abstract: Nonintrusive load monitoring (NILM) is a low-cost but efficient technology to obtain detailed information of electricity consumption, which has a broad application potential in power distribution & utilization. However, the lack of objective, comprehensive and reliable testing and evaluation method impacts greatly the research and implementation of NILM technologies. Therefore, this paper focuses on the testing and evaluation of NILM algorithm and products, and comprehensively summarizes and analyzes the research status and existing problems from three aspects: test data acquisition, evaluation metric selection, and test platform construction. Firstly, existing NILM evaluation processes and methods are summarized and analyzed. Then, the guiding principles of test data requirements are clarified, and different test data acquisition methods and mainstream test data sets are summarized and compared. On this basis, in order to make the test results in different scenarios comparable, the concept of data scenario complexity is introduced and quantifiable complexity description methods are summarized. In terms of algorithm accuracy and product practicality, the performance evaluation indexes of NILM are systematically investigated, analyzed and discussed. Furthermore, it summarizes the status of NILM technology standardization, especially introduces the recently published power industry standard in China "Technical specification for nonintrusive load monitoring device". Finally on the basis of summarizing the existing testing and evaluation platforms, this paper introduces the general framework of the NILM testing and evaluation platform recommended in the industry standards, so as to provide a comprehensive reference for relevant users, researchers and technology implementers.