Sudip Mandal

Bio: Sudip Mandal is an academic researcher from Jalpaiguri Government Engineering College. The author has contributed to research in topics: Gene regulatory network & Search algorithm. The author has an hindex of 7, co-authored 25 publications receiving 172 citations. Previous affiliations of Sudip Mandal include University of Calcutta.

Application of FPA and ANOVA in the optimization of liquid flow control process

Abstract: Received: 6 January 2018 Accepted: 23 March 2018 In process industry liquid flowrate is one of the important variable which need to be controlled in a process to obtain the better quality and reduce the cost of production. As the liquid Flow rate in a process industry depends upon a number of parameter so the process will give the unexpected output as it is caused by the improper setting of parameters. The improper parameter settings could threaten the processes. In this paper, we utilize the Flower Pollination Algorithm (FPA) methods and ANOVA to obtain the optimum conditions of a flowrate in a process industry and to gain the percentage of contributions of each parameter by. A verification test was carried out to inspect the optimum output among the ANOVA & FPA. For generating the objective function 120 sets of data is used in ANOVA while 18 sets of data are used for the verification purpose.

Comparative study: FPA based response surface methodology & ANOVA for the parameter optimization in process control

Abstract: Optimization plays a key role in a process control industry to optimize and prediction of the system’s performance. Most of the process control are multi-variable and to control the parameters to optimized the system performance through the classical method is inflexible, unreliable and time-consuming. Thus, an alternative method will be more effective for parameter optimization & prediction. In this research investigates parameters affecting the liquid flow for the various studied. Design of Experiments based on metaheuristic algorithm is conducted for the analysis of influencing factors. Response surface methodology (RSM) & ANOVA are widely used as a mathematical and statistical tool for system performance optimization. RSM can be employed to optimize and analyze the effects of several independent factors on a treatment process to obtain the maximum output. This paper is to present a comprehensive review on the usability & effectiveness of RSM & ANOVA based on flower pollination algorithm for process parameters modelling and optimization of liquid flow processes. From the appraisal it indicates that the FPA based RSM is gives the more predicted output than the FPA based ANOVA is approximately 9.0389e-6.

A Modified Particle Swarm Optimization Algorithm based on Self-Adaptive Acceleration Constants

Abstract: Particle Swarm Optimization (PSO) is one of most widely used metaheuristics which is based on collective movement of swarm like birds or fishes. The inertia weight (w) of PSO is normally used for maintaining balance between exploration and exploitation capability. Many strategies for updating the inertia weight during iteration were already proposed by several researchers. In this paper, a Modified Particle Swarm Optimization (MPSO) algorithm based on self-adaptive acceleration constants along with Linear Decreasing Inertia Weight (LDIW) technique is proposed. Here, in spite of using fixed values of acceleration constants, the values are updated themselves during iteration depending on local and global best fitness value respectively. Six different benchmark functions and three others inertia weight strategies were used for validation and comparison with this proposed model. It was observed that proposed MPSO algorithm performed better than others three strategies for most of functions in term of accuracy and convergence although its execution time was larger than others techniques.

Metaheuristic Based Parametric Optimization of TIG Welded Joint

Abstract: Martensitic stainless steels are hard, brittle and notch sensitive; crack formation during welding is frequent. Selection of the levels of welding parameters i.e. the input variables seems to be important and useful in the context of achieving optimum/maximum strength of the welded joint. In the present work, focus is given on identification of the proper combination of input parameters in TIG welding of martensitic stainless steel AISI 420. Welding current, gas flow rate and welding speed have been taken as input parameters. Ultimate tensile strength (UTS) and Ductility or Elongation of the welded joint obtained from tensile test is taken as response parameter. Initially, response surface methodology based face-centered central composite design has been used for mathematical model building and regression analysis. Next, several recently proposed metaheuristics are applied for parametric optimization of TIG welding process to maximize the response parameters. From, the simulated results, a critical operating region for efficient TIG welding is identified in term of maximum UTS and Ductility. Confirmatory tests are also performed to validate our proposed methodology.

Machine learning

Abstract: Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Particle Swarm Optimization Feature Selection for Breast Cancer Recurrence Prediction

Abstract: Women who have recovered from breast cancer (BC) always fear its recurrence. The fact that they have endured the painstaking treatment makes recurrence their greatest fear. However, with current advancements in technology, early recurrence prediction can help patients receive treatment earlier. The availability of extensive data and advanced methods make accurate and fast prediction possible. This research aims to compare the accuracy of a few existing data mining algorithms in predicting BC recurrence. It embeds a particle swarm optimization as feature selection into three renowned classifiers, namely, naive Bayes, K-nearest neighbor, and fast decision tree learner, with the objective of increasing the accuracy level of the prediction model.