Study On Rock Drillability Prediction And Drilling Parameter Optimization Based On Well Site Data Of Nanhai A Block

With the increase of energy consumption in the world,middle and shallow oil and gas resources are decreasing day by day,which makes exploration and development continue to advance to deep strata,and the geological environment of drilling becomes more and more complicated,resulting in a series of complex problems such as long drilling cycle and high cost.Through dynamic evaluation of rock drillability and optimization of drilling parameters,complex geological conditions can be reflected in time,so as to reduce the occurrence of accidents and achieve the purpose of reducing cost and consumption.The existing drilling parameter optimization is mainly driven by logic and experience,and there are some problems such as complex model,large amount of calculation and low efficiency.With the rapid development of artificial intelligence,intelligent drilling will become the trend of future development.How to accurately and effectively predict the mechanical drilling rate and realize the optimization of drilling parameters is of great significance for drilling operations.Taking Block A of Nanhai as the research area,the prediction model of rock resistance characteristic parameters is established in this area.On this basis,the main controlling factors affecting drilling rate are screened by correlation analysis,and the intelligent prediction of drilling rate is realized by neural network algorithm.In addition,according to the predicted drillability level value,the formation is stratified by clustering algorithm.Based on the intelligent drilling rate prediction model,the bit selection and parameter optimization of the formation in the same cluster section are carried out.The main research results are as follows:(1)The laboratory experiment of drilling resistance in the research area was carried out and the parameters of drilling resistance were obtained.The rock hardness ranges from 349 to2517MPa,and the drillability level ranges from 3.15 to 9.78.Based on acoustic time difference and mechanical specific energy,two kinds of prediction models of rock drilling resistance characteristic parameters are established.The results show that the R~2 coefficients of the established relational models are all higher than 0.9,and the model has a good fit.(2)Based on drilling and logging engineering parameters in the working area and the predicted drillability level value of the formation,the pre-processing operation was carried out by using wavelet denoising and standardization methods,and the correlation analysis of drilling rate was carried out by using Pearson correlation coefficient method and grey correlation analysis method respectively.Depth,weight on bit,rotational speed,displacement,bit size and drillability stage are selected as the input variables for the drilling rate prediction model.Three algorithms of BP,GA-BP and MEA-BP neural network were introduced respectively to construct three drilling rate prediction models,and MEA-BP prediction model had the best effect.(3)Based on the drillability level value,the clustering algorithm is used to realize stratification processing.In view of the problems of high rock hardness and poor drillability in the cluster group of 5 layers in the working area,the use of the bit is statistically analyzed,the bit is optimized by the principal component analysis method,and the recommended sequence of the bit in different layers in the working area is established.(4)Based on the intelligent drilling rate prediction model,the multi-objective optimization model of drilling parameters is established.The model takes the minimum cost per footage and the minimum mechanical specific energy as the objective function.NSGA-II algorithm was used for field application analysis,and the optimal combination of weight on bit and speed satisfying the objective function was selected.After optimization,the unit footage cost is reduced by 12.3%,and the mechanical specific energy is reduced by 13.7%.The results show that the method proposed in this paper can effectively predict the drilling mechanical penetration rate,and realize the optimization of drill bit and drilling parameters in the target area.The established optimization model has good application effect and can achieve the goal of reducing cost and consumption.