Prediction Model of Pigsty Temperature Based on ISSA-LSSVM

The internal temperature of the pigsty has a great impact on pigs. Keeping in within certain range is pressing problem environmental control. current regulation method based mainly manual and simple automatic There rarely intelligent control, such direct methods have problems as low control accuracy, high energy consumption untimeliness, which can easily lead to occurrence heat stress conditions. Therefore, this paper proposed an improved sparrow search algorithm (ISSA) multi-strategy improvement optimize least squares support vector machine (LSSVM) form prediction model. In optimization process (SSA), initial position population was first generated by using reverse good point set; secondly, number update formula automatically adjust discoverers followers iterations improve ability algorithm; finally, adaptive t-distribution applied discoverer variation refine further algorithm. Tests were conducted 23 benchmark functions, results showed that ISSA outperformed SSA. By comparing it with LSSVM models optimized four standard algorithms, effect ISSA-LSSVM model tested. end, had MSE 0.0766, MAE 0.2105, R2 0.9818. best performance provide accurate data for pigsty.