Temporal outlier detection using fuzzy logic and evolutionary computation

Blast furnace (BF) is the main method of modern iron-making. Ensuring the stability of the BF conditions can effectively improve the quality and output of iron and steel. However, operations of BF depend on mainly human experience, which causes two problems: (1) human experience is not objective and is difficult to inherit and learn and (2) it is difficult to acquire knowledge that contains time information among multiple variables in BF. To address these problems, a data-driven method is proposed. In this article, we propose a novel and efficient algorithm for discovering underlying knowledge in the form of temporal association rules (TARs) in BF iron-making data. First, a new TAR mining framework is proposed for mining temporal frequent patterns. Then, a novel TAR mining algorithm is proposed for mining underlying, up-to-date, and effective knowledge in the form of TARs. Finally, considering the updating of the BF database, a rule updating method is proposed that is based on the algorithm that is proposed in this article. Our extensive experiments demonstrate the satisfactory performance of the proposed algorithm in discovering TARs in comparison with the state-of-the-art algorithms. Experiments on BF iron-making data have demonstrated the superior performance and practicability of the proposed method.

Incremental sequential patterns for multivariate temporal association rules mining

In this paper, we developed a sound conceptual framework for temporal fuzzy association rules mining based on fuzzy information granulation. First, the definition of the support rate of traditional fuzzy association rules is extended to the temporal data, including the fuzzy support rate of both continuous and discontinues temporal fuzzy item set and their association rules. It is found that the computation of support rate under our definition is a simply dynamic programming problem with very low complexity. Then an algorithm based on Apriori is demonstrated. At last, experiments were carried out to show the good performance of our new algorithm in mining continuous rules and discontinuous rules by our definition.

Temporal association rule mining is a variation of association rule mining for finding relationship between items with respect to particular time periods. It is very useful in making business-related decisions, such as catalogue design, cross-marketing, cross-selling, and inventory control. Further, for effective inventory management, economic order quantity (EOQ) of items is determined by considering cross-selling effect among items. In this paper, we propose a new method for EOQ estimation. The method is based on finding the strongest relation between items by using temporal association rule mining. First, opportunity cost of frequent item-sets is determined by the temporal association rule mining, and then this opportunity cost is used to determine the EOQ for imperfect quality items. Furthermore, effect on ordering policy is determined for imperfect quality items and perfect quality items by considering cross-selling effect. A numerical example is illustrated to validate the results.

Nowadays due to the rapid advances in the field of information systems, transactional databases are being updated regularly and/or periodically. The knowledge discovered from these databases has to be maintained, and an incremental updating technique needs to be developed for maintaining the discovered association rules from these databases. The concept of Temporal Association Rules has been introduced to solve the problem of handling time series by including time expressions into association rules. In this paper we introduce a novel algorithm for Incremental Mining of General Temporal Association Rules (IMTAR) using an extended TFP-tree. The main benefits introduced by our algorithm are that it offers significant advantages in terms of storage and running time and it can handle the problem of mining general temporal association rules in incremental databases by building TFP-trees incrementally. It can be utilized and applied to real life application domains. We demonstrate our algorithm and its advantages in this paper.

Temporal pattern mining is the recent research among researchers contributing in the areas of data mining, medical mining, spatial data mining, health informatics and gaining significant interest in Internet of things, one of the top ten fields of interest expected to rule the computing world in the next few years. Temporal pattern mining is a knowledge discovery process which concentrates on mining temporal databases for discovering hidden temporal information which includes finding temporal patterns, temporal and spatio- temporal association rules, performing temporal clustering and classification to name a few of them. In this paper, the major objective is to find temporally outlier patterns from the temporal database of disjoint time-stamped transactions in a single database scan without a need to scan the database multiple times. We aim to eliminate n-1 database scans, performed when we use conventional approach of finding temporal patterns. We demonstrate the approach using a case study. The results show that the proposed approach is computationally efficient which is essentially because of single scan performed.

Design of efficient control algorithms for autonomous mobile robot movement in unknown and changing environment with obstacles and walls is a difficult task. There exist different strategies to design control systems to perform the robot movement. In this article possibility to use fuzzy logic, if-then rules, and genetic algorithm for autonomous mobile robot control is presented. Control using fuzzy logic is softer and better then control using if-then rules because of absence of motors speed jumps. It is shown that the rough control system can be designed using an expert knowledge. Then genetic algorithm can be used to improve the quality of the control system.

Today, Photovoltaic (PV) systems are very popular for power generation because they enjoy the advantage of being pollution free abundant natural resource, low maintenance cost and noiseless operation. Photovoltaic (PV) systems needs maximum power point tracking (MPPT) to maximize the PV array power output which is independent of the temperature, irradiation coPIC18F452nditions and electrical characteristics of the load. The Research Work introduces photovoltaic MPPT system based on PIC18F452 to achieve maximum power tracking. Moreover, the combination of boost circuit and PIC18F452 microcontroller is used for tracking of the maximum power generation from PV system. Moreover, Research presents a practical implementation of Perturbation and Observation (P&O), Incremental Conductance (IncCon) and Fuzzy Logic algorithms to achieve MPPT control. These algorithms are widely used because of its low-cost and ease of realization. Proposed P&O, Incremental Conductance and Fuzzy Logic algorithms are implemented and tested under different loads, and the test results are analyzed and compared. The results show the performance of the Fuzzy Logic and Incremental Conductance algorithm in tracking MPP is better than the P&O algorithm and the experimental results of Fuzzy Logic and Incremental Conductance algorithm indicate the feasibility and improved system functionality with high-efficiency. Also, results concludes that the design can effectively accomplish the purpose of the maximum power tracking.

This paper reports on an approach that contributes towards the problem of discovering fuzzy association rules that exhibit a temporal pattern. The novel application of the 2-tuple linguistic representation identifies fuzzy association rules in a temporal context, whilst maintaining the interpretability of linguistic terms. Iterative Rule Learning (IRL) with a Genetic Algorithm (GA) simultaneously induces rules and tunes the membership functions. The discovered rules were compared with those from a traditional method of discovering fuzzy association rules and results demonstrate how the traditional method can loose information because rules occur at the intersection of membership function boundaries. New information can be mined from the proposed approach by improving upon rules discovered with the traditional method and by discovering new rules.

This article shows a new result in term of maximum power point tracking control which is the efficiency of the ripple correlation control compared to the fuzzy logic command under spatial climatic conditions (a temperature between 40C°and - 60C°and an irradiation of 0 W/m2 or 1370 W/m2) in order to maximize the electrical energy produced by multi-junction photovoltaic panels of a satellite (the example of Nimbus 2) located in the earths low orbit. The simulation results indicate the ripple correlation control efficiency in terms of stability and accuracy compared to the fuzzy logic algorithm which causes energy losses due to large oscillations around the maximum power point.

Conventionally modelling and simulation of complex nonlinear systems has been to construct a mathematical model and examine the system’s evolution or its control. This kind of approach can fail for many of the very large non-linear and complex systems being currently studied. With the invention of new advanced high-speed computers and the application of artificial intelligence paradigms new techniques have become available. Particularly neural networks and fuzzy logic for nonlinear modelling and genetic algorithms [Goldberg, D. (1989)] and evolutionary algorithms for optimisation methods have created new opportunities to solve complex systems [Bai, Y., Zhuang H. and Wang, D. (2006)]. This paper considers issues in design of multi-layer and hierarchical fuzzy logic systems. It proposes a decomposition technique for complex systems into hierarchical and multi-layered fuzzy logic sub-systems. The learning of fuzzy rules and internal parameters in a supervised manner is performed using genetic algorithms. The decomposition of complex nonlinear systems into hierarchical and multi-layered fuzzy logic sub-systems reduces greatly the number of fuzzy rules to be defined and improves the learning speed for such systems. In this paper a method for combining subsystems to create a hierarchical and multilayer fuzzy logic system is also described. Application areas considered are - the prediction of interest rate, unemployment rate predication and electricity usage prediction. Genetic Algorithms can be used as a tool for design and generation of fuzzy rules for a fuzzy logic system. This automatic design and generation of fuzzy rules, via genetic algorithms, can be categorised into two learning techniques namely, supervised and unsupervised. In supervised learning there are two distinct phases to the operation. In the first phase each individual is assessed based on the input signal that is propagated through the system producing output respond. The actual respond produced is then compared with a desired response, generating error signals that are then used as the fitness for the individual in the population of genetic algorithms. Supervised learning has successfully applied to solve some difficult problems. In this paper design and development of a genetic algorithm based supervised learning for fuzzy models with application to several problems is considered. A hybrid integrated architecture incorporating fuzzy logic and genetic algorithm can generate fuzzy rules that can be used in a fuzzy logic system for modelling, control and prediction

The aim of this chapter is to show the effectiveness of soft computing in the design of adaptive controllers for a single-link, flexible manipulator. The first part utilizes fuzzy logic (FL) and genetic algorithms (GAs) in the design of an offline modular neural network controller while the second part combines fuzzy logic and GAs in the construction of online proportional derivative (PD), proportional, integral (PI) and proportional, integral, derivative (PID)-type FL controllers. Experimental results, demonstrating the effectiveness of the methods for an experimental single-link, flexible manipulator are presented and discussed.

Design and implementation of a fuzzy logic (FL) coprocessor is presented in this paper. The main goal of a FL coprocessor is to speed up the operations typical for FL algorithms. Current implementations of FL algorithms are dedicated and general purpose. The main drawback of solutions based on dedicated hardware is lack of configurability. Software implementation on general purpose processors is slow for many applications. The basic components of the FL coprocessor presented in this paper are designed in Handel-C using Celoxica's DK Design Suite. The inference engine inside the FL coprocessor is a Mamdani-type engine and the FL coprocessor only supports trapezoid and triangular membership functions. With the DK Design Suite, area estimations are analyzed and throughput is increased by applying pipelining and parallelism transformations.

Described in this paper are the design and implementation of a microcontroller-based fuzzy logic peak current-mode controller (PCMC) for dc-dc converters. A fuzzy logic PCMC boost converter operating in the continuous conduction mode has been designed using an 8-bit microcontroller. A one-chip solution has been achieved. No complex algorithm is required to estimate the controlled current by using the on-board peripherals of the microcontroller. The fuzzy logic algorithm uses a rule-based description of converter operation. By using rule-based look-up tables and other techniques, the fuzzy logic algorithm is implemented on the low-cost microcontroller successfully. Experimental results indicate successful operation of the microcontroller-based fuzzy logic PCMC power converter system.

Traditional financial data mining and information disclosure research faces problems such as huge data volume, uneven data quality, and untimely information disclosure. This paper aims to use fuzzy logic algorithms to process financial data, improve data mining efficiency, explore new information disclosure methods, and design an information disclosure evaluation model based on fuzzy logic algorithms. The experiment constructed a financial data mining model based on an adaptive neural fuzzy inference system (ANFIS), optimized the model parameters using a fuzzy genetic algorithm (FGA), and used a fuzzy comprehensive evaluation model to form information disclosure content with the help of logical reasoning. The research results show that the model accuracy of financial data mining reaches 96.38%, which has a good effect on information disclosure, and the algorithm efficiency is better than that of traditional algorithms.