Global Journal of Computer Science and Technology, C: Software & Data Engineering, Volume 22 Issue 2

brought in his computing and instrumentation engineering knowledge geared towards his working experience of more than ten (10) years alongside with the domain practitioners to understand more of the activities of the plant. To attain these task or objectives, is by integrating action research method and object- oriented development (OODA) for the modelling. a) Action Research Method According to Bryman and Bell (2011), action research is “an approach in which the action researcher and a client collaborate in the diagnosis of the problem and in the development of a solution based on the diagnosis. One of the main characteristic traits of action research relates to collaboration between researcher and member of organization to solve organizational problems. In line with the quantitative and qualitative research approach proposed by Kusumarasdyati, (2016), the method has advantages added Bryman & Bell (2011). Both the numerical values for the quantitative and non-numerical values for the qualitative were all under technical, practical, and emancipatory action research. The numerical values are from the boiler drum recommended quality and feed water recommended quality, whereas the non-numerical entails the boiler problems and causes extracted from the O&M of fertilizer plant manual. Also, the researcher conducted one-on-one interview with fourteen (14) of the fertilizer plant boiler domain practitioners, carrying out dynamic simulations during the design of proposed system, changes were made based on their recommendations on the application on consistency, correctness, and its precision. The dataset numeric values are corresponding to the physical or chemical state of the boiler system, control measures such as shutdown, bidirectional or one directional flow, open on excess fluid, or close on excess fluid can be used to control the performance of the steam package boiler to keep it in a workable condition. The performance and health of the steam package boiler can be control, measured and monitored using process measuring, control and sensing devices such as pressure gauges, temperature sensor (transducers), flow and level gauges, actuating valve etc. which are incorporated in the proposed system designed. The idea is to take all the useful dataset from the boiler manual according to its performance and capability in different operation points to model better system for plant control and maintenance as stated by Michael (2016), same as applied in user centered design approach to situation awareness, Nwiabu & Adeyanju (2012). b) Objective-Oriented Development Approach Nabil & Govardhan (2012) describe object- oriented (OO) approach as a system development methodology that follows an iterative and incremental approach to systems development. In each increment or phase, the developers move through the activities of gathering requirements, analyzing the requirements, designing the system, implementing the design, and testing the system. Adetokunbo et al., (2013) added that object-oriented approach models its processes using objects. That is, the solution of problems can be seen as a set of objects or computations performed in the context of objects. Data and the processes that act on the data are encapsulated within every object. Each object’s data (attributes or states) are the properties that relate to the object. Nabil & Govardhan, (2012), added that object- oriented development approach (OODA) uses a set of diagramming techniques known as the Unified Modeling Language or UM and focuses on the three architectural views of a system: functional, static, and dynamic, makes use of iterative and incremental steps, it gives opportunity to manage changes as they occur to user requirements. So, it is more prone to user satisfaction. In this research work, the steam package boiler is object to be controlled and maintained. The Java programming language is suitable for the proposed system development because each module in the system is seen as classes, this gives the environment to develop a user interface for interaction with the system. IV. E xpert S ystem A rchitecture The expert system architecture of the steam package boiler rule-base system and fuzzy logic is shown in Fig1. It shows the encoded knowledge of the domain expert put into the system’s database where the inference engine would make use of it to process and determine the working state and errors of the system. When the system runs on the manual operation mode (MOM), the user (field operators/engineer) enters the correct parameter set points which enable a corresponding output from the input signal and expected value of the affected plant system as authorized by their plant shift-in-charge (SIC) or plant operations manager. In the expert system shell, data flow bi-directional to check values from plant (field instruments) and from field operator or process system engineer navigating the system. The description of the system architecture as shown in Figure1 below, has the processes of both fuzzification and defuzzification, which plays major roles in the system architecture. The fuzzy logic is found in the part of the inference engine, where the software system converts rules to code and codes back to rules. Steam Package Boiler Expert System for Control and Maintenance of Fertilizer Plants Using Rule-Base Fuzzy Logic Global Journal of Computer Science and Technology Volume XXII Issue II Version I 27 Year 2022 ( ) C © 2022 Global Journals