When the widespread access of new energy power generation and air conditioning to the park energy Internet, the uncertainty of the source-load side leads to the complex and changeable operation scenarios.The impact on the energy-saving operation of the park cannot be ignored. How to realize the efficient economic planning of the park energy Internet is an urgent problem to be studied at present. Therefore, the current stochastic programming theory is firstly summarized, and the research status at home and abroad is analyzed. Based on statistical machine learning, the following three key problems are analyzed. Firstly, the complex operation scenario simulation technology of energy Internet caused by the uncertainty of new energy power generation and demand response. The second is the typical scenario generation technology；The third is the safe and stable artificial intelligence programming solution technology. Finally, the development prospect is put forward. Statistical machine learning theory will be one of the important directions of the future development of the energy network in the park.