Research On Arsenic Removal In Underground Water With Combined Pollution

With rapid developing social level and continuous improving life quality, drinkingwater safety issues arouse more and more attention. In the National major Water Projectof "Eleventh Five-Year Plan", urban drinking water safety technology becomes apriority among priorities. In recent years, arsenic poisoning events occurred frequentlydue to the long-term consumption of micro-polluted arsenic groundwater. The seriousdamage for human health of soluble inorganic arsenic in underground water has alsobeen widely concerned by the international community. Therefore, studying on typicalcase of arsenic groundwater with combined pollution and achieving reasonableautomatic control efficiently has strategic significance to protect safety supply ofdrinking water, improve the management efficiency of water supply enterprises andensure sustained, healthy and stable development of society.This thesis focused on arsenic raw groundwater with combined iron andmanganese pollution in Dongzhou waterworks in Zhengzhou. With pilot trial andlaboratory experiment, the process of aeration-contact-oxidation-filtration withoutcoagulation was used for arsenic removal research. Based on research, the BP neuralnetwork model can be built for arsenic removal automatic control system inwaterworks.The three-stage pilot trial got the optimal parameters in arsenic removal process,including a combination of clinker and quartz sand as filter material, FeCl3and KMnO4as arsenic removal chemicals and48h as best filter cycle. At the same time, arsenic,iron and manganese were all up to the standard after the process. With stable operation,lasting filter cycle and low turbidity, and the other water quality index all reached therelevant regulations of Drinking Water Standard (GB5479-2006). The influence of rawwater quality indexes on the effect arsenic removal was different between pilot trial andlaboratory experiment. So we should focus on the specific water quality conditionswhen choosing reasonable chemical dosage to achieve the ideal arsenic removal effect.Based on lots of experimental data, several most significant factors during arsenicremoval process were selected to set up two kinds of BP neural network model withprogressive relation in research. Among them, the arsenic model can accurately simulateeffluent arsenic concentration variation under the different water quality and objectiveconditions. It’s suitable for simulating and characterizing variation between waterquality parameters, objective conditions and the effluent arsenic concentration.Meanwhile, with accurately simulating drug dosage under different condition, the drugdosage model, as core simulation algorithm module, applied to build arsenic removal automatic control dosing system to improve the current control system in waterworks.