Tailings saturation line prediction based on genetic algorithm and BP neural network

Article type: Research Article

Authors: Tongle, Xu^* | Yingbo, Wang | Kang, Chen

Affiliations: Mechanical Engineering School, Shandong University of Technology, Shandong Zibo, China

Correspondence: [*] Corresponding author. Xu Tongle, Mechanical Engineering School, Shandong University of Technology, Shandong Zibo 255049, China. Tel.: +86 13964413228; Fax: +86 05332786982; E-mail: xutongle@163.com.

Abstract: Due to being influenced by many factors, tailings dam saturation line comes to be complex and non-linear, which is difficult to be predicted. To solve this problem, genetic neural network algorithm is proposed to build saturation line forecasting model. Some factors are identified as the root causes for saturation line change, and they are the input nodes of the neural network which is able to analyze data adaptively. Genetic algorithm, as a global searching algorithm, is used to optimize weights of BP neural network. By the proposed method, saturation line change tendency can be obtained faster and more accurately. An experiment is performed to test the advance and feasibility of the method.

Keywords: BP neural network, genetic algorithm, saturation line, forecasting model

DOI: 10.3233/IFS-151905

Journal: Journal of Intelligent & Fuzzy Systems, vol. 30, no. 4, pp. 1947-1955, 2016