| Slope stability problem is the key geo-technical problems need to be solved in water conservancy and hydropower, railway, highway and disaster prevention project construction. Limit equilibrium method is the earliest and most widely used method to evaluate the slope stability problems.The limit equilibrium method for slope stability analysis usually contains two steps: first step, for any given slip surface in slope, analyzes the state of limit equilibrium and calculate the safety factor of the certain slip face; The second step is, search for the most dangerous slip surface in the slope from all possible slip surfaces and the most dangerous slip surface corresponds to the stability of the overall stability safety factor.Morgenstern-Price Method(hereinafter referred to as the MP method), which the author studies on in this paper, is the most general and limit equilibrium method which has the best assumptions, and the solving process is strict and precise. However, the method for solving the assumption of constant λ and safety facor Fs, often set a set of initial values, and then the numerical iteration method is applied to approximate the exact solution. Numerical solving process of the conventional MP method is relatively complex, the iterative process is involved in the problems of local convergence or no convergence, and therefore difficult to popularization and application in engineering practice. Starting from the basic concept and theory of MP method, a MP method of genetic optimization method which can be used in stability analysis is established. The method is a global optimization method. In the system of research work, the following results has achieved:Starting from the basic theory and concept of MP method, this paper derived the relationship of bulk push thrust and moment transmission, formed the detailed solving of thrust Ei and moment Mi on both sides of the slip mass. Assume that a set of λ and Fs, according to the initial conditions on one side of the landslide E0 = 0, M0 = 0, thrust En and press moment Mn on the other side of the landslide of this formula can be obtained.When the λ and Fs approach the exact value, thrust En and press moment Mn should converge to zero. According to this idea, the fitness function of genetic algorithm is established, through the fitness function, the superiority of λ and the Fs solutions can be evaluated. Proceed the genetic evolutionary computation. With the given improved MP method, λ and Fs can be calculated.The basic idea is to transfer λ and Fs into binary encoding and series as a chromosome, loop through the "choice", "cross", "variation" operation, until the evolution results meet the optimal conditions, λ and global optimal solution of the Fs.In order to further search the most dangerous sliding surface of slope, this paper has also established a principle based on two-dimensional MP method of linear sliding surface simulation strategy. It is a strategy that based on the equal width b assumption which slip crack surface around the starting position of L and R and the end of each article strips with vertical direction angle to determine the point position of slip surface. Such simulation offers a strategy to avoid the unreasonable critical slip surface named "zigzag" and "hollow". Under the simulation strategy, bent slip crack surface control variable number is only(n + 1), and part of the process can be directly used in the MP method article points as a result, at the same time in the guarantee the correctness of the solution set of greatly reduced the number of variables and simulation computation. According to the simulation strategy, search the most dangerous slip crack surface problem can be expressed as the objective function, with Ll, Lr, α1,Δ α2 ~ Δ αn- 1 as the control variables in nonlinear optimization problem.This paper establishes the fitness function of the no-circular slip surface, and provides the method of using genetic algorithm to search the most dangerous slip surface.The detailed strategy of this method is: in series Ll, Lr, α1 and Δαn-1~ Δα2 for a total of(n+1) control variables,which can be binary encoded into a chromosome of 24+8(n) digit.Loop "choice", "cross", "variation" operation, until the evolution results meet the optimal conditions and found the most dangerous slip surface and the minimum safety factor of slope.Around the most dangerous sliding surface search process prone to "zigzag" and "hollow" problem causes and judgment method and approach of the study. The reasons of “zigzag”: randomly generated slip line intersect with the line more than two times(common in multistage slope), which caused by the inside part of the critical slip surface is not in the slope. About "hollow" phenomenon causes: the random generation of Lr and Ll determines the location of the L and R, cause L am in sometimes, in the case of right and the "R" in the left, to generate the critical slip surface of disorder. In this paper, the "zigzag" and "hollow" in the process of these two kinds of search provides a judgment formula errors. Further by adopting the idea of "penalty function", avoided this results from two kinds of errors.Based on the study of the improved MP method and method of searching critical slip surface, a slope stability analysis and calculation program in the Visual Basic development platform is developed. The calculation program can realize the following functions: first, the two-dimensional slope of any given slip surface, improved MP method can be used to calculate the critical slip surface stability coefficient value; Second,above all the possible critical slip surface in slope, solution focused on the most dangerous slip surface and the minimum stability coefficient of slope can be given. Take ACADS classic examination an example to verify this program, using Slide software and "the referee" as comparison. Validation results show the method is reliable and accurate.Applying the above research results to Niaoqueping Landslide in order to analysis the stability.The analysis of Niaoqueping landslide include the potential slip surface and the overall stability of the slope stability. Further analysis of the critical slip surface is represented by chromosomes genetic evolution process. Analysis shows that using the improved MP method for slope stability analysis is an effective and accurate method of global optimization which has the advantage of large scope of the research and high search precision.. |
PDF Full Text Request