In order to improve the accuracy of load forecasting, a short-term load forecasting model based on Elman-IOC neural network with chaotic oriented cuckoo algorithm was proposed. Firstly, the Elman neural network topology is improved by adding the input-output layer connection unit, the network parallel computing capability is enhanced and the prediction accuracy is improved. Then, in the cuckoo algorithm, the optimal location information is used to guide the random walk process. At the same time, the chaos disturbance operator is introduced to enhance the global search ability. Finally, the algorithm is applied to Elman-IOC neural network parameter optimization, and a short-term load forecasting model is established. Experimental results show that compared with other models, this model has a higher prediction accuracy.