Research on Automatic Crack Detection for Concrete Infrastructures Using Image Processing and Deep Learning

Automatic crack detection is a critical task in the generation of a crack map for existing concrete infrastructure inspection. This paper describes an automatic crack detection and classification method based on a genetic algorithm (GA) for optimizing image processing technique parameters (IPTs). Under various complex photometric conditions, the crack detection results of concrete infrastructure surface images remain noise pixels. Following that, a deep convolution neural network (CNN) method is used to automatically classify crack candidates and non-crack candidates. Furthermore, the proposed method is compared to state-of-the-art crack detection methods. The experimental results validate the reasonable accuracy in practice. The final goal was to create a crack map, which necessitated automatic pixel-level accuracy.

Author(s) Details

Cuong Nguyen Kim
Faculty of Highway & Bridge, Mien Trung of Civil Engineering, Vietnam.

Kei Kawamura
Graduate School of Science & Technology for Innovation, Yamaguchi University, Japan.

Hideaki Nakamura
Graduate School of Science & Technology for Innovation, Yamaguchi University, Japan.

Amir Tarighat
Department of Civil Engineering, Shahid Rajaee Teacher Training University, Iran.

View Book :-

An Approach of Optimization Techniques for History Matching and Production Forecasting

Petroleum as a natural resource is depleting year after year, necessitating effective management of the resource and its reservoir. Reservoir modelling and production forecasting are critical inputs in this scenario for effective management. The spatial distribution of reservoir properties describing the reservoir and associated output profiles is difficult to estimate since naturally occurring reservoirs are extremely heterogeneous and nonlinear in nature. In terms of reservoir properties and history matching, an accurate model constructed with the aid of data obtained from the reservoir, can lead to more effective reservoir management, and such models can be created using mathematical modelling and optimization techniques. This chapter covers a variety of optimization strategies that can be used for output forecasting and background matching. Simulated Annealing (SA), Scatter Search (SS), Neighborhood algorithm (NA), Particle Swarm Optimization (PSO), and Ant Colony Optimization are examples of gradient-based and non-gradient-based optimization techniques (ACO), The application of Ensemble Kalman Filters (EnKF) and Genetic Algorithms (GA) to reservoir output history matching and efficiency. The chapter also goes into recent developments and adaptations of these techniques.

Author (s) Details

Giridhar Vadicharla
Department of Chemical Engineering, University of Petroleum and Energy Studies, Dehradun, India

Pushpa Sharma
Department of Petroleum Engineering and Earth Sciences, University of Petroleum and Energy Studies, Dehradun, India

View Book :-

Estimation of Electric Power Demand Using Socioeconomic Variables: A Case of Indian Electricity Sector

The production and industrial growth of any country needs accurate forecasting of energy demand. In this research, per capita electricity demand is forecast using the Genetic Algorithm. As input data, socio-economic variables such as population, Gross Domestic Product, exports, and imports were used. Linear, quadratic and exponential models are used to forecast capacity. Performance parameters, including the Root Mean Square Error, R2 and modified R2, are determined for the models. For three separate potential scenarios, electricity demand is projected for the period 2017-2035. A strong agreement with the actual data with a high correlation coefficient (R2 = 99.45 percent) and low Root Mean Square Error ( RMSE = 0.1455) is shown by the projected demand for electricity. This method can, therefore, be used in India as an alternative energy forecasting technique.

Author(s) Details

Dr. Sanjeet Singh
Decision Sciences Area, Indian Institute of Management Lucknow, Prabandh Nagar, IIM Road, Lucknow-226013, India.

K. R. Ramakrishnan
ZS Associates India Private Limited, Pune, India.

View Book :-

Exploring Optimum Solutions for Management of House Hold Data with Population Census in Indian Context

The study discusses the population census in identified states and its influence in neighborhood states. A fitting function will be generated from the identified data which will be processed using genetic algorithm to find the most probable state which influences population among the identified set. In continuation to that we can consider the House hold data in different states as rows and different types of House Holds like Good, Livable and Dilapidated as columns as input to the GenAlgo function. The problem is initialized with a fitness function and mutation function relevant to the Household problem. The work starts with data frame that is passed back to fitfun and mutfun to enable them to take advantage of any additional data viable for them to perform their proposed functions. The idea here is to have put together a data frame containing the Good number of households and Livable households of the population to identify the best performers of states in development activities. This work tries to map population growth with development activities like House hold data in finding a balance in growth among different states in India.

Author(s) Details

Addepalli V. N. Krishna
School of Engineering and Technology, CHRIST (Deemed to be University), Bengaluru-560074, India.

View Book :-


Application of Genetic Algorithm Solution Approach to Voltage Drop Issues on 33 kV/11 kV Injection Feeders: A Case Study of Ogbomoso, South West, Nigeria

The place of good quality and quantity of electricity supply by electric power provider in national growth cannot be underestimated. But, sadly the quantity and quality of electricity in most third world countries such as Nigeria is plagued by quite a number of power quality disturbances and technical losses inherent within the system. Voltage drop affects the quantity of available electricity and it is a major concern of electric power providers as it challenged their sole responsibility of supplying customers with the required voltage level at all times. Surprisingly, the causes and effects of voltages drops on 33kV/11kV transmission systems have not been extensively looked at in Nigeria. This paper presents application of genetic algorithm solution approach to voltage drop issues on 33kV/ 11kV Injection feeders: A case study of Ogbomoso, South West, Nigeria. The result of the analysis showed that the receiving end voltage is of low proportion compared to the sending end voltage. The parametric modeling of voltage drop revealed several causes of voltage drop in the study area. Different cable sizes were used to mitigate the effect voltage drop, it was discovered that, to attain minimum voltage drop in this station, the 65 mm2 cable used has to be augmented to 85 mm2 or reduce to 50 mm2 while the number of the injection stations should be increase.

Read full article: