Since total petroleum hydrocarbons (TPH) and organic carbon (OC) are two important variables in the performance of oily sludge composting process; the prediction of their changes is of great importance to attain high removal efficiency. The main objective of this work was to model oily sludge composting process by using neuro-evolutive methodology based on artificial neural networks (ANNs) and differential evolution (DE) in order to predict TPH and OC removal in various conditions of the process. The experimental data on oily sludge composting were used to validate the model. So as to determine the optimal ANN model, a set of randomly generated models are initially generated and their parameters are evolved by the DE until a stop criterion is reached. It was found that TPH and OC were modeled well and the ANN model can provide predictions which were in accordance with the experimental data. The obtained results can be used to lessen the costs of full-scale bioremediation through eliminating the need for further experiments.

由于总石油烃（TPH）和有机碳（OC）是油性污泥堆肥过程中的两个重要变量；对它们变化的预测对于获得高去除效率非常重要。这项工作的主要目的是通过使用基于人工神经网络（ANN）和差异进化（DE）的神经进化方法，对油性污泥堆肥过程进行建模，以便预测过程中各种条件下的TPH和OC去除量。利用含油污泥堆肥的实验数据对模型进行了验证。为了确定最佳的ANN模型，首先生成一组随机生成的模型，然后由DE演化其参数，直到达到停止标准为止。发现TPH和OC建模良好，并且ANN模型可以提供与实验数据一致的预测。通过消除进一步的实验需求，可以将获得的结果用于降低全面生物修复的成本。