The continuous increase in energy consumption has made the potential of wind-power generation tremendous. However, the obvious intermittency and randomness of wind speed results in the fluctuation of the output power in a wind farm, seriously affecting the power quality. Therefore, the accurate prediction of wind power in advance can improve the ability of wind-power integration and enhance the reliability of the power system. In this paper, a model of wavelet decomposition (WD) and weighted random forest (WRF) optimized by the niche immune lion algorithm (NILA-WRF) is presented for ultra-short-term wind power prediction. Firstly, the original serials of wind speed and power are decomposed into several sub-serials by WD because the original serials have no obvious day characteristics. Then, the model parameters are set and the model trained with the sub-serials of wind speed and wind power decomposed. Finally, the WD-NILA-WRF model is used to predict the wind power of the relative sub-serials and the result is reconstructed to obtain the final prediction result. The WD-NILA-WRF model combines the advantage of each single model, which uses WD for signal de-noising, and uses the niche immune lion algorithm (NILA) to improve the model’s optimization efficiency. In this paper, two empirical analyses are carried out to prove the accuracy of the model, and the experimental results verify the proposed model’s validity and superiority compared with the back propagation neural network (BP neural network), support vector machine (SVM), RF and NILA-RF, indicating that the proposed method is superior in cases influenced by noise and unstable factors, and possesses an excellent generalization ability and robustness.

With the increasing coupling of the power system and the natural gas system, the electric–gas interconnection system has become a typical form of comprehensive energy utilization. Through the energy conversion function of the coupling unit, the system can flexibly participate in the bidding for purchasing and selling energy in a power market and a natural gas market on the premise of meeting the internal demand of multiple loads. To solve the internal coordination and optimization problem and the external flexible bidding problem in the multi-energy market, this paper proposes a robust optimization model of energy purchase and sale for the electric–gas interconnection system in a multi-energy market. Firstly, the basic structure of the electric–gas interconnection system is introduced, and the steady-state model of energy flow in the system is built based on the energy hub model. Secondly, considering the uncertainty of energy prices and the output power of renewable energy units in the system, a bidding model for energy purchase and sale of the electric–gas interconnection system in multi-energy market based on the idea of robust optimization is constructed in the framework of the Nordic energy market. Finally, empirical analysis based on the actual data is carried out, and the results prove the validity and superiority of the model. In this paper, aiming at the uncertainty of energy price, a large number of scenes are generated by Latin hypercube sampling (LHS), and then a k-means algorithm is used to reduce the scenes, so as to simulate typical scenes. Aiming at the uncertainty of the output power of the renewable energy unit in the system, a cardinal uncertainty set is used to control deviation between the actual output power and predicted output power, so that the overall robustness of the model can be controlled. The proposed model can make decision-making independent of the accurate probability distribution of uncertainty factors, and is suitable for complex multi energy systems. Meanwhile, the model possesses excellent robustness, which can effectively reduce the risk of bidding loss in the process of energy purchase and sale.