Optimal Design of Parallel-Spool High-Pressure Solenoid Valve
Dong Dai,Li Xiaoning,Guo Zhonghua
DOI: https://doi.org/10.3969/j.issn.1002-6819.2014.20.006
2014-01-01
Abstract:An underwater driving system requires a high-pressure solenoid valve with a high flow rate and high speed. A novel solenoid valve with parallel spools was proposed to meet the demand. The opening response time of the solenoid valve had a great influence on the performance of the underwater driving system. In order to reduce the opening response time of the novel parallel-spool pneumatic pilot operated high-pressure solenoid valve, an optimal design method was proposed. First, a mathematical model of the opening process of the valve was built up. A prediction model of the opening response time of the valve was deduced from the mathematical model established. In the prediction model, the opening response time was subdivided into four parts including the armature preparation time, the armature movement time, the exhaust time of the control chamber, and the movement time of the main valve spool. The corresponding formulas to solve the response time were derived. Secondly, based on the prediction model, an optimization program of the parameters of the pilot valve was designed by using a Multi-population Genetic Algorithm which was developed from a Standard Genetic Algorithm. The optimization program was written in MATLAB programming language using the Sheffield Genetic Algorithm Toolbox. To reduce the opening response time, four key parameters including the nominal diameter of the pilot valve, the radius of the armature, the width of the coil, and the length of the coil needed to be optimized. After optimization, the opening response time of the valve was 47.6 ms which was reduced by 14%. The impact of the key parameters on the opening response time of the valve was discussed respectively by using the optimization program. With the increase of the nominal diameter of the pilot valve and the length of the coil, the opening response time of the valve first decreases and then increases. Namely, there are optimal values for those two parameters. The opening response time fluctuated with the increase of the diameter of the armature, but the amplitude was small. Along with the rising of the width of the coil, response time was slightly decreased at first, then increased rapidly. On the basis of the analysis, multiple local extremums were found. That will result in many difficulties to the algorithm in converging to the global extremum. The optimization of the solenoid valve was an optimization with multiple local extremums. The optimization algorithm should possess a good capability to search for a global optimum. The optimization results obtained by using the Multi-population Genetic Algorithm showed that the optimization program designed was effective. Compared with the Standard Genetic Algorithm, the Multi-population Genetic Algorithm performed much better than the Standard Genetic Algorithm in avoiding the premature phenomenon, and its convergence speed was faster. At last, a test prototype of the novel valve was processed based on the optimized parameters and an opening response test of the valve was carried out. The experimental results showed that the opening response time tested was 48.3 ms at the pressure of 10 MPa, that was 1.5% larger than optimization result. The error between the experiment and the optimization was small. After optimization, the solenoid valve can meet the demands of the underwater driving system well.