Recently, Japan’s discharge of wastewater from the Fukushima nuclear disaster into the ocean has attracted widespread attention. To effectively address the challenge of separating uranium, the focus is on finding a healthy and environmentally friendly way to adsorb uranium using biochar. In this paper, a BP neural network is combined with each of the four meta-heuristic algorithms, namely Particle Swarm Optimization (PSO), Differential Evolution (DE), Cheetah Optimization (CO) and Fick’s Law Algorithm (FLA), to construct four prediction models for the uranium adsorption capacity in the treatment of radioactive wastewater with biochar: PSO-BP, DE-BP, CO-BP, FLA-BP. The coefficient of certainty (R²), error rate and CEC test set are used to judge the accuracy of the model based on the BP neural network. The results show that the Fick’s Law Algorithm (FLA) has a better search ability and convergence speed than the other algorithms. The importance of the input parameters is quantitatively assessed and ranked using XGBoost in order to analyze which parameters have a greater impact on the predictions of the model, which indicates that the parameters with the greatest impact are the initial concentration of uranium (C₀, mg/L) and the mass percentage of total carbon (C, %). To sum up, four prediction models can be applied to study the adsorption of uranium by biochar materials during actual experiments, and the advantage of Fick’s Law Algorithm (FLA) is more obvious. The method of model prediction can significantly reduce the radiation risk caused by uranium to human health during the actual experiment and provide some reference for the efficient treatment of uranium wastewater by biochar.

In this study, we proposed an improved Harris Hawks Optimization (IHHO) algorithm based on the Sobol sequence, Whale Optimization Algorithm (WOA), and t-distribution perturbation. The improved IHHO algorithm was then used to optimize the BP neural network, resulting in the IHHO-BP model. This model was employed to predict the bonding strength of heat-treated wood under varying conditions of temperature, time, feed rate, cutting speed, and grit size. To validate the effectiveness and accuracy of the proposed model, it was compared with the original BP neural network model, WOA-BP, and HHO-BP benchmark models. The results showed that the IHHO-BP model reduced the Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) by at least 51.16%, 40.38%, and 51.93%, respectively, while increasing the coefficient of determination (R²) by at least 10.85%. This indicates significant model optimization, enhanced generalization capability, and higher prediction accuracy, better meeting practical engineering needs. Predicting the bonding strength of heat-treated wood using this model can reduce production costs and consumption, thereby significantly improving production efficiency.

The physical and mechanical properties of heat-treated wood are essential factors in assessing its appropriateness for different applications. While back-propagation (BP) neural networks are widely used for predicting wood properties, their accuracy often falls short of expectations. This paper introduces an improved Beluga Whale Optimisation (IBWO)-BP model as a solution to this challenge. We improved the standard Beluga Whale Optimisation (BWO) algorithm in three ways: (1) use Bernoulli chaos mapping to explore the entire search space during population initialization; (2) incorporate the position update formula of the Firefly Algorithm (FA) to improve the position update strategy and convergence speed; (3) apply the opposition-based learning based on the lens imaging (lensOBL) mechanism to the optimal individual, which prevents the algorithm from getting stuck in local optima during each iteration. Subsequently, we adjusted the weights and thresholds of the BP model, deploying the IBWO approach. Ultimately, we employ the IBWO-BP model to predict the swelling and shrinkage ratio of air-dry volume, as well as the modulus of elasticity (MOE) and bending strength (MOR) of heat-treated wood. The benefit of IBWO is demonstrated through comparison with other meta-heuristic algorithms (MHAs). When compared to earlier prediction models, the results revealed that the mean square error (MSE) decreased by 39.7%, the root mean square error (RMSE) by 22.4%, the mean absolute percentage error (MAPE) by 9.8%, the mean absolute error (MAE) by 31.5%, and the standard deviation (STD) by 18.9%. Therefore, this model has excellent generalisation ability and relatively good prediction accuracy.

This study integrates the Backpropagation (BP) Neural Network with several optimization algorithms, namely Hippopotamus Optimization (HO), Parrot Optimization (PO), Osprey Optimization Algorithm (OOA), and Goose Optimization (GO), to develop four predictive models for the adhesive strength of heat-treated wood: HO-BP, PO-BP, OOA-BP, and GO-BP. These models were used to predict the adhesive strength of the wood that was heat-treated under multiple variables such as treatment temperature, time, feed rate, cutting speed, and abrasive particle size. The efficacy of the BP neural network models was assessed utilizing the coefficient of determination (R²), error rate, and CEC test dataset. The outcomes demonstrate that, relative to the other algorithms, the Hippopotamus Optimization (HO) method shows better search efficacy and convergence velocity. Furthermore, XGBoost was used to statistically evaluate and rank input variables, revealing that cutting speed (m/s) and treatment time (hours) had the most significant impact on model predictions. Taken together, these four predictive models demonstrated effective applicability in assessing adhesive strength under various processing conditions in practical experiments. The MAE, RMSE, MAPE, and R² values of the HO-BP model reached 0.0822, 0.1024, 1.1317, and 0.9358, respectively, demonstrating superior predictive accuracy compared to other models. These findings support industrial process optimization for enhanced wood utilization.

With the rapid development of industrialization, the problem of heavy metal wastewater treatment has become increasingly serious, posing a serious threat to the environment and human health. Biochar shows great potential for application in the field of wastewater treatment; however, biochars prepared from different biomass sources and experimental conditions have different physicochemical properties, resulting in differences in their adsorption capacity for uranium, which limits their wide application in wastewater treatment. Therefore, there is an urgent need to deeply explore and optimize the key parameter settings of biochar to significantly improve its adsorption capacity. This paper combines the nonlinear mapping capability of SCN and the ensemble learning advantage of the Adaboost algorithm based on existing experimental data on wastewater treatment. The accuracy of the model is evaluated by metrics such as coefficient of determination (R²) and error rate. It was found that the Adaboost–SCN model showed significant advantages in terms of prediction accuracy, precision, model stability and generalization ability compared to the SCN model alone. In order to further improve the performance of the model, this paper combined Adaboost–SCN with maximum information coefficient (MIC), random forest (RF) and energy valley optimizer (EVO) feature selection methods to construct three models, namely, MIC-Adaboost–SCN, RF-Adaboost–SCN and EVO-Adaboost–SCN. The results show that the prediction model with added feature selection is significantly better than the Adaboost–SCN model without feature selection in each evaluation index, and EVO has the most significant effect on feature selection. Finally, the correlation between biochar adsorption properties and production parameters was discussed through the inversion study of key parameters, and optimal parameter intervals were proposed to improve the adsorption properties. Providing strong support for the wide application of biochar in the field of wastewater treatment helps to solve the urgent environmental problem of heavy metal wastewater treatment.

The properties of wood change after heat treatment, affecting its applications. Glossiness, a key aesthetic property, is of great significance in fields like furniture. Precise prediction can optimize the process and improve product quality. Although the traditional back propagation neural network (BPNN) has been applied in the field of wood properties, it still has issues such as poor prediction accuracy. This study proposes an improved whale optimization algorithm (IWOA) to optimize BPNN, constructing an IWOA-BPNN model for predicting the glossiness of heat-treated wood. IWOA uses chaos theory and tent chaos mapping to accelerate convergence, combines with the sine cosine algorithm to enhance optimization, and adopts an adaptive inertia weight to balance search and exploitation. A dataset containing 216 data entries from four different wood species was collected. Through model comparison, the IWOA-BPNN model showed significant advantages. Compared with the traditional BPNN model, the mean absolute error (MAE) value decreased by 66.02%, the mean absolute percentage error (MAPE) value decreased by 64.21%, the root mean square error (RMSE) value decreased by 69.60%, and the R² value increased by 12.87%. This model provides an efficient method for optimizing wood heat treatment processes and promotes the development of the wood industry.