The Jiangmen Underground Neutrino Observatory (JUNO) is a large liquid scintillator detector designed to explore many topics in fundamental physics. One of the capabilities of the JUNO detector is to search for the baryon number violation processes, which would be a crucial step towards testing the Grand Unified Theories and explaining the matter-antimatter asymmetry of the Universe. The large liquid scintillator detector of JUNO has distinct advantages in the search for some nucleon decay modes. This talk reports the JUNO potential to search for $p\rightarrow \bar{\nu} K^+$ and neutron invisible decays (e.g., $n\rightarrow 3 \nu$ or $nn \rightarrow 2 \nu$). Both of them can produce time-, energy- and space-correlated triple coincidence signals, which may be used to effectively suppress backgrounds. It has been found that the expected sensitivities for JUNO with 10 years of data are $\tau / B(p \to \bar{\nu} K^+) > 9.6 \times 10^{33} \, {\rm years}$, $\tau/B( n \rightarrow { inv} ) > 5.0 \times 10^{31} \, {\rm years}$ and $\tau/B( nn \rightarrow { inv} ) > 1.4 \times 10^{32} \, {\rm years}$ at the 90% confidence level, which are better than the current best limits.