Power management strategy based on Elman neural network for grid-connected photovoltaic-wind-battery hybrid system

Rheological characteristics and behaviour prediction of lubricating grease for RV reducer across a wide temperature range

“…The output of the implicit layer goes into the takeover layer, and the output of the takeover layer returns to the implicit layer, so as to achieve the memory function of the historical data, which makes the network have the ability to adapt to the time‐varying. Compared with the feed‐forward neural network, the ENN has more computational ability and more stability [24–27]. The specific expression of the ENN is as follows: $$$$\begin{equation} \def\eqcellsep{&}\begin{array}{l} r(k) = f({W}_2u(k - 1) + {W}_1{r}_c(k) + {b}_1)\\ {r}_c(k) = r(k - 1)\\ p(k) = g({W}_3r(k) + {b}_2) \end{array} \end{equation}$$$$where $$u(k\; - 1)$$ is the input vector of the neural network, $$r(k)$$ is the vector of the hidden layer $$p(k)$$ and $${r}_c(k)$$ is the vector of the output vector and the vector of the undertaking layer.…”

“…The output of the implicit layer goes into the takeover layer, and the output of the takeover layer returns to the implicit layer, so as to achieve the memory function of the historical data, which makes the network have the ability to adapt to the time-varying. Compared with the feed-forward neural network, the ENN has more computational ability and more stability [24][25][26][27]. The specific expression of the ENN is as follows:…”

Research on the potential for China to achieve carbon neutrality: A hybrid prediction model integrated with elman neural network and sparrow search algorithm

“…ENN is a recursive network featured by an internal self-referencing layer. ENN consists of four components: input layer, context layer, hidden layer, and output layer ( Boualem et al, 2022 ). The context layer is designed to store or memorize the output values before the hidden layer.…”

“…The nonlinear expression of the ENN is as follows ( Boualem et al, 2022 ; Ruiz et al, 2018 )： where k is the training number of ENN; y denotes the n-dimensional output vector; x represents the hidden layer neuron output vector; x c denotes the feedback state vector; u means the input vector; g represents the transfer function of the output neuron; 、 and is the connection weight matrices, and f represents the transfer function of the hidden layer neuron.…”

“…Hence, the ENN has been embraced by many researchers and fields. ( Boualem et al, 2022 ; Ding et al, 2022 ; Meng et al, 2022 ; Ruiz et al, 2018 ).…”