基于Transformer模型的电力隧道开挖工法预测研究

在电力资源供应不足与绿色城市建设的双重影响下,电力隧道已经成为城镇化建设的必要选择,而根据已有工程条件确定电力隧道开挖工法更是施工稳定与取得经济效益的关键。以我国多个省份提供的电力隧道开挖工法应用情况为基础构建了数据集,应用Transformer模型对数据集之间的因果序列关系进行了学习与预测研究。研究结果表明,与传统机器学习模型相比,Transformer模型做到了不定项预测,同时在电力隧道开挖工法的最优项预测方面也取得了较好的结果,其准确率、精确率、召回率和F1值分别为98.25%、98.53%、98.45%和98.47%。Transformer模型在北京市某些已有工程实例中也得到了进一步的验证,具有极大的发展潜力。相关结论可为类似研究提供参考。

冷轧扭钢筋混凝土空心板预制工法

冷轧扭钢筋是一种新型冷化强度钢材,是北京建筑工程研究所的科研成果。冷轧扭钢筋采用的母材为φ6.5～10普通热轧圆盘条,经过GQZ10A型钢筋冷轧机,冷轧并扭成螺旋状。其设计抗拉强度为母材的1.95倍。与混凝土之间的握裹力明显提高,两端不需弯钩,可用于混凝土结构和构件中。经试验和工程应用证明,其技术性能符合《钢筋混凝土结构设计规范》(TJ10-74)

Springback prediction for incremental sheet forming based on FEM-PSONN technology

In the incremental sheet forming （ISF） process, springback is a very important factor that affects the quality of parts. Predicting and controlling springback accurately is essential for the design of the toolpath for ISF. A three-dimensional elasto-plastic finite element model （FEM） was developed to simulate the process and the simulated results were compared with those from the experiment. The springback angle was found to be in accordance with the experimental result, proving the FEM to be effective. A coupled artificial neural networks （ANN） and finite element method technique was developed to simulate and predict springback responses to changes in the processing parameters. A particle swarm optimization （PSO） algorithm was used to optimize the weights and thresholds of the neural network model. The neural network was trained using available FEM simulation data. The results showed that a more accurate prediction of s!oringback can be acquired using the FEM-PSONN model.

Prediction of Gas Holdup in Bubble Columns Using Artificial Neural Network

A new correlation for the prediction of gas hold up in bubble columns was proposed based on an extensive experimental database set up from the literature published over last 30 years. The updated estimation method relying on artificial neural network, dimensional analysis and phenomenological approaches was used and the model prediction agreed with the experimental data with average relative error less than 10%.

Predictive Vegetation Mapping Approach Based on Spectral Data, DEM and Generalized Additive Models

This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.

Mixed-Weights Least-Squares Stable Predictive Control Algorithm with Soft and Hard Constraints

Mixed-weight least-squares (MWLS) predictive control algorithm, compared with quadratic programming (QP) method, has the advantages of reducing the computer burden, quick calculation speed and dealing with the case in which the optimization is infeasible. But it can only deal with soft constraints. In order to deal with hard constraints and guarantee feasibility, an improved algorithm is proposed by recalculating the setpoint according to the hard constraints before calculating the manipulated variable and MWLS algorithm is used to satisfy the requirement of soft constraints for the system with the input constraints and output constraints. The algorithm can not only guarantee stability of the system and zero steady state error, but also satisfy the hard constraints of input and output variables. The simulation results show the improved algorithm is feasible and effective.

Study and application of monitoring plane displacement of a similarity model based on time-series images

In order to compensate for the deficiency of present methods of monitoring plane displacement in similarity model tests,such as inadequate real-time monitoring and more manual intervention,an effective monitoring method was proposed in this study,and the major steps of the monitoring method include:firstly,time-series images of the similarity model in the test were obtained by a camera,and secondly,measuring points marked as artificial targets were automatically tracked and recognized from time-series images.Finally,the real-time plane displacement field was calculated by the fixed magnification between objects and images under the specific conditions.And then the application device of the method was designed and tested.At the same time,a sub-pixel location method and a distortion error model were used to improve the measuring accuracy.The results indicate that this method may record the entire test,especially the detailed non-uniform deformation and sudden deformation.Compared with traditional methods this method has a number of advantages,such as greater measurement accuracy and reliability,less manual intervention,higher automation,strong practical properties,much more measurement information and so on.

Prediction of blast boulders in open pit mines via multiple regression and artificial neural networks

The most important objective of blasting in open pit mines is rock fragmentation.Prediction of produced boulders(oversized crushed rocks) is a key parameter in designing blast patterns.In this study,the amount of boulder produced in blasting operations of Golegohar iron ore open pit mine,Iran was predicted via multiple regression method and artificial neural networks.Results of 33 blasts in the mine were collected for modeling.Input variables were:joints spacing,density and uniaxial compressive strength of the intact rock,burden,spacing,stemming,bench height to burden ratio,and specific charge.The dependent variable was ratio of boulder volume to pattern volume.Both techniques were successful in predicting the ratio.In this study,the multiple regression method was superior with coefficient of determination and root mean squared error values of 0.89 and 0.19,respectively.

Developing energy forecasting model using hybrid artificial intelligence method

An important problem in demand planning for energy consumption is developing an accurate energy forecasting model. In fact, it is not possible to allocate the energy resources in an optimal manner without having accurate demand value. A new energy forecasting model was proposed based on the back-propagation(BP) type neural network and imperialist competitive algorithm. The proposed method offers the advantage of local search ability of BP technique and global search ability of imperialist competitive algorithm. Two types of empirical data regarding the energy demand(gross domestic product(GDP), population, import, export and energy demand) in Turkey from 1979 to 2005 and electricity demand(population, GDP, total revenue from exporting industrial products and electricity consumption) in Thailand from 1986 to 2010 were investigated to demonstrate the applicability and merits of the present method. The performance of the proposed model is found to be better than that of conventional back-propagation neural network with low mean absolute error.

Erythrocyte Fragment Count Predicts Hemolysis in Roller Pumps

Objective： Hemolysis in blood pumps has been measured by various in vitro test methods, in which normalized index of hemolysis （NIH） was established. As NIH is complicated and difficult to calculate, erythrocyte fragment count is proposed in the present study to predict hemolysis in roller pumps. Methods： Five paired in vitro tests were conducted using the POLYSTAN pediatric pump（group A） and COBE pump（ group B）. Ten whole blood samples （400 ml ） were circled in the roller pump for 16 h. Erythrocyte fragments count and plasma-free hemoglobin （FHb） were measured before pumping and every two hours through circulation after four-hour-pumping. The morphological changes of erythrocyte were observed by scanning electron microscope. Results： The two groups＇ EFC and FHb levels were increased linearly during a long duration of pumping and linear regression of erythrocyte fragments count and plasma-free hemoglobin were correlated. Conclusion： Erythrocyte fragments count could be used as an index in evaluating the in vitro hemolytic properties of blood pumps.

Prediction of Leachate Generation in a Landfill Using Artificial Neural Networks

One of the problems encountered in the operation of a leachate treatment in a landfill is the quantity of the fluctuating leachate. Therefore, information on the precise prediction about the quantity of leachate produced in a landfill is required. This information can be obtained by using an ANN （artificial neural networks） model. In this study, a prediction on a leachate generation for a period of 15 days was made. The input for the ANN model consists of data such as rainfall, temperature, humidity, duration of solar radiation, and the landfill characteristics, while the output is the leachate landfills production in Minamiashigara, Japan. The ANN algorithm uses a BP （back propagation） with LM （Levenberg-Marquadrt） training type. By using the input-output data pairs, the training of ANN model was conducted in order to obtain the values of the weights that describe the relationship between the input-output data. Furthermore, with the trained ANN model, the prediction of leachate generation for a period of 15 days was made. The study result shows that the prediction accuracy ofleachate generation of ANN-C model, with a correlation coefficient （r） of 0.924, is quite good. Thus, the prediction of leachate generation using artificial neural network model can be recommended for predicting leachate generation in the future. In this study, a prediction on a leachate generation for a period of 15 days was made. The quantity of leachate generation in a landfill can be obtained by using ANN for future periods. By entering data for future periods （t ＋1） in ANN models, the leachate generation for the period （t ＋1） can be predicted.