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

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 30 Year 2022 ( ) C © 2022 Global Journals Figure 3: System Activity Diagram On MOM, when the system malfunctioning is detected by a drift in setpoint parameters the user (field operator/engineer) inputs values for corrections which enables efficient control and smooth running of the steam boiler device, otherwise, the plant operation manager required for the maintenance team (Mech, Electrical and Instrument) to carry out maintenance by bye-passing required field equipment’s. If the system is runs on AOM, and there is erratic or faults detections, and values are drifting from required default parameter setpoints, the system would automatically make corrections which aid continues running of boiler plant. The process of designing this system, is aimed at providing detailed dataset from Table1 and Table2and other system elements that enable the development and implementation reliable. The system is designed is such a way that, the output variables or condition of the steam package boiler controls its input variables or condition of the boiler device. The pH value at 25oCin the table above has a range of value suitable for the application of fuzzy logic system. The FLS is designed as follows. Algorithm 1. Define linguistic variables and terms (start) 2. Construct membership function for the terms (start) 3. Construct knowledge base of the rules (start) 4. Construct crisp data into fuzzy datasets using membership functions (Fuzzification). 5. Evaluate rules in the Rule Base (Inference Engine) 6. Combine results from each rule (Inference Engine) 7. Convert output data into non-fuzzy values (Defuzzification) Development 1. State(s) = {very-low, low, normal, high, very-high} 2. Constructing membership function Table 3: Ph Value at 25oc Showing Membership Values M/Class Class/R Grade LP x = 9.5 L/value MP x = 9.25 H/Average S x = 9 Average MN x = 8.75 Lower/Average LN x = 8.5 Smallest/Value The above table3 has been used to design a membership function graph for the membership values in the table, as calculated with the datasets of pH value at 25oC ranging between8.5ppm (actual minimum value) to 9.5ppm (actual maximum value). Graph 1: A graph of membership function of pH value at 25oC ranging between 8.5ppm (Mini value) to 9.5ppm (maxi value). 1and 4. Constructing a Knowledge Based Rule from the membership function. Table 4: Membership function table -membership function replaced with actual numeric values. The actual rule from the inference engine will be as follows: State Target (s) - Very/L Low Normal High Very/H Very/L No/C 8.75 9 9.25 9.5 Low 8.5 No/C 9 9.25 9.5 Normal 8.5 5.75 No/C 9.25 9.5 High 8.5 8.75 9 No/C 9.5 Very/H 8.5 8.75 9 9.25 No/C