Optimal Control Of Drilling Fluid Disposal System Based On Online Water-Quality Monitoring System

The drilling fluid wastewater treatment device which based on Mechanical Vapor Recompression(MVR)takes the low temperature negative pressure evaporation as the core and combined with the physical treatment technology such as pressure,filtration and adsorption.The device solves the traditional treatment process(recovery and reuse method,curing method,biological treatment,evaporation and concentration treatment)which is difficult to apply to different components of the drilling fluid,the terminal treatment and high cost engineering problems,so it is very suitable for China's limited space drilling a large number of drilling fluid treatment.However,there have some problems in the device at present: 1,auxiliary processing module(such as aeration,demulsification,pressure filtration and activated carbon adsorption)effects cannot be judged;2,because of the diversity of drilling fluid,the use of a unified process will cause the energy consumption Waste and increase the processing time.Therefore,the paper carries on the research to optimization control on the equipment.First of all,targeted experiments were carried out for aeration,demulsification,pressure filtration and activated carbon adsorption treatment module in the paper.It was found that the dissolved oxygen content in the drilling fluid was less than 2.55mg/L,and the aeration treatment was carried out for 20 min minutes.The demulsification stage did not need to be used in the treatment of water-based drilling fluid.When the concentration of suspended solids is higher than 20mg/L,it needs to be filtered into the filter before entering the MVR evaporator.In the post-processing stage,when the COD content in the drilling fluid was greater than 188.2mg/L,de-oiled and activated carbon adsorption both worked on the drilling fluid;when the COD content in the drilling fluid was greater than 120mg/L and less than188.2mg/L only the activated carbon adsorption worked on the drilling fluid;when the COD content in the drilling fluid was less than 120mg/L,the drilling fluid was carried to the tank.In the activated carbon adsorption stage,the amount of activated carbon is 112.2g/ml and the COD 188.2mg/L drilling fluid wastewater,adsorption treatment time 20 min to meet the "Integrated Wastewater Discharge Standard"(GB8978-1996)secondary standard(120mg/L),COD content decay rate of 40.12%.Secondly,a water quality on-line monitoring system is introduced for device.Through the analysis of the mechanism of the treatment process,the water quality parameters to be monitored are dissolved oxygen,suspended matter and COD during the drilling fluidtreatment.As the COD content is measured for a long time,moreover,the equipment cost and the analysis cost are both expensive,the paper uses the soft sensor model to predict the COD content to influence the COD content of the drilling fluid which puts the main water quality parameters of the COD content of the drilling fluid wastewater(dissolved oxygen,suspended matter,turbidity,pH content)as the input,and COD content as the output.This paper compares the BP(Back Propagation)neural network with the improved BP neural network-unscented Kalman neural network prediction model,thus it is found that the precision of the unsatisfactory Kalman neural network prediction model is higher than the prediction model of BP neural network,and the prediction error is only about 10%.So the prediction model of the unsatisfactory Kalman neural network prediction model is more in line with the requirement of COD measurement.The simulation results show that the non-trace Kalman neural network prediction model realizes the real-time performance of water quality on-line detection,because it has high accuracy and can predict the COD content in the drilling fluid wastewater in real time and combine with the dissolved oxygen which can be measured in real time.Finally,the device is optimized control.In order to achieve real-time online water quality testing,I installed in the installation of a set of integrated dissolved oxygen,suspended solids,turbidity,pH water quality on-line detector.And in order to achieve the diversification of the flow of drilling fluid sewage,I added a two-way valve and pipeline,so as to complete the parallel optimization of the water pipe network.As the device has the problem of high energy consumption and low degree of automation,the fuzzy control rules which are based on the physical characteristics of the control object and the experience of the engineering practice.The fuzzy control rules were established to control the start and stop of the aeration,pressure filtration,oil removal and activated carbon adsorption module and adjust the evaporator temperature by adjusting the opening of the MVR evaporator balance valve.Moreover,in the sub-control system of activated carbon adsorption module design applied degenerative feedback closed loop control system.So the device can cut down the energy consumption and reduce the operation cost on the basis of guaranteeing the water quality to reach the standard.The experimental study on the optimized control of the device: After the treatment in the device,the water quality parameters of drilling fluid wastewater from Hechuan's 1# is achieved two standard of "integrated wastewater discharge standards"(GB8978-1996).The device treats the drilling fluid sewage for 10 hours,the energy consumption per hour 27.8kw.h.Compared with the previous experimental energy consumption per hour 28.4kw.h,the device wite the optimized control can save energy consumption per hour 0.6kw.h.