Research On The Influences Of Design Parameters On The Performances Of 3kW Diesel Generator Sets In Plateau Environment

Diesel generator sets are power equipments that convert the chemical energy of diesel into electrical energy.Which composed of a diesel engine and a generator.It has the characteristics of convenient movement,rapid starting,stable power supply,convenient operation and maintenance,and strong environmental adaptability.It is widely used in hospitals,banks,airports,hotels,communications,ship electricity,stone mining,engineering repair,military and other industries.After years of development,diesel generator sets have good working performances in plain conditions.Most diesel generator sets are limited to an altitude below 1000m on the market.However,the environmental conditions of the plateau are complex.With the increase of altitude,the density of the air decreases,the atmospheric pressure drops sharply,and the oxygen content in the air also drops sharply,which leads to the deterioration of the combustion condition of the diesel engine.Diesel generators used in the plains cannot work properly in the plateau environment.The 3kW diesel generator sets are important facilities in the border defense force and have extremely important strategic role.Therefore,this paper takes a trial-produced3kW diesel generator sets as the research object,and analyzes the influences of its design parameters on the performance indicators of the diesel generator sets under different altitude conditions.Based on the orthogonal tests,with the output power,specific fuel consumption,starting time and smoke as the evaluation indicators,the optimal design parameters of the diesel generator sets working at an altitude of 3000m and an altitude of4250m were obtained.At the same time,the basic theories of performance optimization of diesel generator sets were deeply studied,and the mathematical models between the performance indicators and design parameters of diesel generator sets were established,and the mathematical models of the regression equation obtained were optimized by Optimum in Design-Expert,so that it could be used at an altitude of 4250m.The output power was greater than 2.7kW in the environment.And based on the artificial neural network improved by genetic algorithm,the performance index models of diesel generator sets were established,and the prediction of performance index of diesel generator sets under different altitudes and design parameters were realized.The main research contents of this paper are as follows.Firsty,the principle and influencing factors of combustion and emission of diesel generator were theoretically analyzed.Selected six design parameters,such as compression ratio,preheating scheme,fuel injection pressure,circulating fuel supply,air filter element diameter and fuel supply advance angle of diesel generator sets as the influencing factors.Field tests were carried out on the plain,at an altitude of 1500 meter,3000 meter and 4250 meter.Taking two kinds of diesel generators with compression ratio of 19.5 and 21.5 as the research objects,tested the output power,specific fuel consumption,starting time and smoke intensity and the influence of the compression ratio on the performance indicators of diesel generators at different altitudes were analyzed.Using three schemes of intake port preheating,in-cylinder preheating,and simultaneous intake port and in-cylinder preheating,explored the influence of the preheating scheme on the starting performance of diesel generators at different altitude.Performance tests were carried out with five different fuel injection pressures,circulating fuel supply,air filter element diameters,fuel supply advance angles.and the relationship curves were drawn.Obtained the influences of four parameters on the performance index of diesel generator sets by analyzing the relationship curves.Secondly,using L₉（3⁴）standard orthogonal table,and the output power,specific fuel consumption,starting time,smoke intensity and comprehensive performance indicators were used as evaluation indicators.Taking the fuel injection pressure,the circulating fuel supply,the diameter of the air filter element diameter and fuel supply advance angle as the influencing factors,three levels were set respectively.The orthogonal tests designs were carried out at an altitude of 3000m and 4250m.The range and variance analysis were carried out on the orthogonal test data,and the primary and secondary factors,the influence significance of the four influencing factors on each evaluation index and the optimal design parameters corresponding to each evaluation index of diesel generator sets in two test environments were obtained.Then,fuel injection pressure,circulating fuel supply,air filter element diameter and fuel supply advance angle were selected as independent variables,output power Y₁,specific fuel consumption Y₂,starting time Y₃,and smoke intensity Y₄ were selected as response values.The experimental scheme obtained by the central composite design in the Design-Expert software was tested at an altitude of 4250m.Quadratic fitting was performed on the experimental data,and the regression equations between each variable and the response value were obtained.Analysises of variance and response surface analysis were performed on the regression equation obtained to test the significance of the overall model and the significance of the effects of each variable and the interaction between variables on the response value.The optimal solution was found for each response value by adding constraints,objective functions and weights of response values in the Optimum.Finally,selected the appropriate parameters through the independent variables corresponding to the optimal solution,improved the original test prototype and tested its performance indicators to verify the reliability of the constructed quadratic regression models.Finally,altitude,fuel injection pressure,circulating fuel supply,air filter element diameter and fuel supply advance angle were used as input in this paper.Output power,specific fuel consumption,starting time and smoke were output.And used genetic algorithm to optimize the weights and thresholds in the BP neural network operation,and built diesel generator sets performance index prediction model.The simulation tests of diesel engine plateau performance were carried out by using MATLAB.Took 111 sets of data as training samples,selected the tansig function and the logsig function as the transfer function of the middle layer and the output layer,the traingdx function as the training function,the number of nodes in the middle layer was set to 7,the number of iterations was 1000,the expected mean square error was 0.00001,The learning rate lr was 0.01,the maximum evolutionary generation was 30,the population size was 50,the crossover probability was 0.95,and the mutation probability was 0.3 for training.Taking 18 sets of data as test samples,the model output predicted value and actual test value were analyzed.The accuracy and reliability of the established diesel generator sets performance indexes prediction model were verified.