A matrixed nonlinear grey Bernoulli model for interval number prediction of power generation in East China

Purpose The interval number prediction of power generation can provide a reference for the rational planning of the power system. For the nonlinearity, uncertainty and complex trends of power generation in East China, a matrixed nonlinear grey Bernoulli model combined with the weighted conformable fractional accumulation generating operator (MWCFNGBM(1,1, <m:math xmlns:m="http://www.w3.org/1998/Math/MathML"><m:msup><m:mi>t</m:mi><m:mi>α</m:mi></m:msup></m:math> )) is proposed. Design/methodology/approach First, the original sequence fluctuations are smoothed with the weighted conformable fractional accumulation generating operator. The time power term is introduced into the nonlinear grey Bernoulli model to enhance the flexibility and adaptability of predicting nonlinear and complex sequences. The model parameters are further matrixed so that the interval number sequences can be modeled directly. The improved MPA is chosen to optimize the nonlinear parameters through the algorithm comparison. Finally, the Cramer rule is used to derive the time recursive formula. Findings The validity and superiority of the MWCFNGBM(1,1, <m:math xmlns:m="http://www.w3.org/1998/Math/MathML"><m:msup><m:mi>t</m:mi><m:mi>α</m:mi></m:msup></m:math> ) is verified by the model comparison experiment. The total power generation in East China is predicted and analyzed from 2024 to 2027. The prediction shows that it will grow steadily over the next four years. Originality/value The trend of power generation in East China is complex in the short term. It is of research significance to use the grey model for short-term interval prediction of power generation. For the data characteristics of power generation, a grey interval number prediction model for power generation prediction is proposed.