Development Of A Gel Physical Property Analyzer Based On Parallel Computing

As an important functional material,gel has been widely used in the energy industry,power industry,aerospace field,food cosmetics,pharmaceutical biology and other fields.The physical parameters of the gel are of great significance for studying the rheological properties and reaction kinetics of the gel system.In various applications of gel,the amount of plugging agent is very large and the physical properties of gel are also very high.The physical property analyzer developed for gel plugging agent can be easily extended to various application fields,which has high demonstration effect and wide application value.At present,the common gel physical properties measurement equipment has complex operation,slow calculation speed and high price,which is difficult to meet the work requirements.Therefore,a gel physical property analyzer based on digital image is developed,which can quickly and effectively measure the gel related parameters in real time.At the same time,the use of parallel computing can significantly improve the data processing efficiency of the equipment,shorten the data processing time,and improve the automation of the instrument.The paper analyzes the calculation method of gel physical properties.On this basis,the hardware and software parts of the gel physical properties analyzer were designed according to the needs of actual analysis and calculation.According to the hardware design scheme,a reasonable hardware device including pressure pump and camera is selected,which can better complete the functions of simulated gel flow and capillary image acquisition.According to the hardware device,the paper proposes an improved segmentation algorithm from the perspective of improving calculation speed and accuracy.Based on the Otsu method,the algorithm distinguishes the target from the background by adding certain judgment conditions,which can effectively filter out the influence of light and shadow on the recognition results.When the concave liquid surface curve of the gel is incomplete,a polynomial curve fitting algorithm is proposed to fit the concave liquid surface curve of the gel in the capillary.On this basis,the core parameters(35)L are determined by optical flow method and monocular ranging principle,so as to calculate the physical properties of the gel.Based on the research of hardware design and algorithm,a high-precision gel physical property analyzer based on parallel computing is developed.The physical characteristic analyzer has the characteristics of high precision,strong real-time performance,convenient operation,non-contact and portable.The actual measurement of the physical properties of three batches of gel in the oil field shows that the physical properties analyzer developed in this paper avoids the problems of contact test samples and complicated operation,and can accurately obtain the actual physical properties.The average relative error of the measurement results is increased from 3% to about 2%,which meets the accuracy requirements of practical engineering applications.The acceleration effect of the whole system is tested and counted.The results show that the system speed is significantly improved after the introduction of parallel computing,and the acceleration ratio can reach tens of times.