Background to remove constraints produced by current
BackgroundNumerical models arefrequently used to simulate the flow and water quality problems. Usually,selecting a suitable numerical model to solve a practical water quality problemis a highly specialised task requiring detailed knowledge on the applicationand limitation of models. Due to the complexity, there is an increasing demandto integrate artificial intelligence (AI) with these mathematical models inorder to assist selection and manipulation.ProjectPurposeThe advancement inartificial intelligence (AI) over the past few decades has made it possible tointegrate technologies into numerical modelling systems to remove constraintsproduced by current numerical models which are insufficiently user friendly.There are several algorithms and methods which can be used, in this reportthese techniques are explore, the techniques are as follows, knowledge basedsystems, genetic algorithm, artificial neural network and fuzzy inferencesystem. Reasonfor using AIMany model users do notpossess the requisite knowledge to glean their input data, build algorithmicmodels and evaluate their results. This may produce inferior designs causingunderutilization or total failure of the model.
Due a computer uses memory andspeed, a balance between speed and accuracy need to be struck.KnowledgeBased SystemsThis technique usessymbolic and logical reasoning algorithm Knowledge based systemsmimic and automate the decision making and reasoning processes of human expectsin solving problems.GeneticAlgorithmThis technique uses anevolutionary algorithm that uses selection, reproductive, crossover andmutation.
Genetic Algorithm usescomputational models of natural evolutionary process in developing computerbased problems solving systems.ArtificialNeural NetworkThis technique uses datadriven models approached with highly interconnected processing elementsArtificial Neural Networkuses an information processing paradigm that is inspired by biological nervoussystems in simulating underlying relationships that are not fully understood.Fuzzyinference systemThis technique uses mapelements of a fizzy set to a universe of membership valuesFuzzy inference systemsuse modelling complex and imprecise systems when objective or the constraintsare vague using a function theoretic membership form belonging to the closeinterval from 0 to 1.
ConclusionThis study hasreviewed the progress on the integration of AI into water quality modelling. Theintegration of various AI techniques, including knowledge based systems, geneticalgorithms, artificial neural networks and fuzzy inference system, intonumerical modelling systems have been reviewed where it was found that thesetechniques can contribute to the integrated model in different aspects and maynot be mutually exclusive to one another. Some futuredirections for further development and their potentials are explored andpresented. It is believed, with the ever-heightening capability of AItechnologies, that further development of numerical modelling in this directionwill be promisi