Genetic Algorithm Based on New Evaluation Function and Mutation Model for Training of BPNN

A local minimum is frequently encountered in the training of back propagation neural networks (BPNN), which sharply slows the training process. In this paper, an analysis of the formation of local minima is presented, and an improved genetic algorithm (GA) is introduced to overcome local minima. The Sigmoid function is generally used as the activation function of BPNN nodes. It is the flat characteristic of the Sigmoid function that results in the formation of local minima. In the improved GA, pertinent modifications are made to the evaluation function and the mutation model. The evaluation of the solution is associated with both the training error and gradient. The sensitivity of the error function to network parameters is used to form a self adapting mutation model. An example of industrial application shows the advantage of the improved GA to overcome local minima.

Shale Fracturability Graphic Template Based onMixed Analytic Hierar-chy Process andMutation Theory

Due to the depletion of conventional energy reserves,there has been a global shift towards non-conventional energy sources.Shale oil and gas have emerged as key alternatives.These resources have dense and heterogeneous reservoirs,which require hydraulic fracturing to extract.This process depends on identifying optimal fracturing layers,also known as‘sweet spots’.However,there is currently no uniform standard for locating these sweet spots.This paper presents a new model for evaluating fracturability that aims to address the current gap in the field.The model utilizes a hierarchical analysis approach and a mutation model,and is distinct in its use of original logging data to generate a fracturability evaluation map.Using this paper’s shale fracturing sweet spot evaluation method based on a two-step mutation model,four wells in different blocks of Fuling and Nanchuan Districts in China were validated,and the results showed that the proportion of high-yielding wells on the sweet spot line could reach 97.6%,while the proportion of low-producing wells was only 78.67%.Meanwhile,the evaluation results of the model were compared with the microseismic data,and the matching results were consistent.

Efficient generation of the mouse model with a defined point mutation through haploid cell-mediated gene editing

Generation of mouse models carrying a defined point mutation,especially disease-related point mutations,is of considerable interest for research in biology and medicine.The standard method based on embryonic stem cell（ESC）-mediated homologous recombination（HR）is time-and labor-consuming.