Solving piecewise linear equations in abs-normal form

نویسندگان

  • Andreas Griewank
  • Tom Streubel
چکیده

With the ultimate goal of iteratively solving piecewise smooth (PS) systems, we consider the solution of piecewise linear (PL) equations. As shown in [Gri13] PL models can be derived in the fashion of automatic or algorithmic differentiation as local approximations of PS functions with a second order error in the distance to a given reference point. The resulting PL functions are obtained quite naturally in what we call the abs-normal form, a variant of the state representation proposed by Bokhoven in his dissertation [vB81]. Apart from the tradition of PL modeling by electrical engineers, which dates back to the Master thesis of Thomas Stern [Ste56] in 1956, we take into account more recent results on linear complementarity problems and semi-smooth equations originating in the optimization community [CPS92, Sch12, FP03]. We analyze simultaneously the original PL problem (OPL) in abs-normal form and a corresponding unfolded system (UPL), which is closely related to a linear complementarity problem (LCP). We show that the UPL, like KKT conditions and other singly switched systems, cannot be open without being injective. Hence we find that some of the intriguing PL structure described by Scholtes [Sch12] is lost in the unfolding from OPL to UPL. To both problems one may apply Newton variants with appropriate generalized Jacobians directly computable from the abs-normal representation. Alternatively, the ULP can be solved by Bokhoven’s modulus method and another fixed point solver that is asymptotically faster but requires a slightly stronger convergence condition. We compile the properties of the various schemes and highlight the connection to the properties of the Schur complement matrix, in particular its signed real spectral radius as analyzed by Rump in [Rum97]. Numerical experiments and suitable combinations of the fixed point solvers and stabilized generalized Newton variants remain to be done. Preprint submitted to Linear Algebra and its Applications November 21, 2013

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

ABS-Type Methods for Solving $m$ Linear Equations in $frac{m}{k}$ Steps for $k=1,2,cdots,m$

‎The ABS methods‎, ‎introduced by Abaffy‎, ‎Broyden and Spedicato‎, ‎are‎‎direct iteration methods for solving a linear system where the‎‎$i$-th iteration satisfies the first $i$ equations‎, ‎therefore a‎ ‎system of $m$ equations is solved in at most $m$ steps‎. ‎In this‎‎paper‎, ‎we introduce a class of ABS-type methods for solving a full row‎‎rank linear equations‎, ‎w...

متن کامل

ABS Solution of equations of second kind and application to the primal-dual interior point method for linear programming

 Abstract  We consider an application of the ABS procedure to the linear systems arising from the primal-dual interior point methods where Newton method is used to compute path to the solution. When approaching the solution the linear system, which has the form of normal equations of the second kind, becomes more and more ill conditioned. We show how the use of the Huang algorithm in the ABS cl...

متن کامل

A new multi-step ABS model to solve full row rank linear systems

ABS methods are direct iterative methods for solving linear systems of equations, where the i-th iteration satisfies the first i equations. Thus, a system of m equations is solved in at most m ABS iterates. In 2004 and 2007, two-step ABS methods were introduced in at most [((m+1))/2] steps to solve full row rank linear systems of equations. These methods consuming less space, are more compress ...

متن کامل

Multigrid method for nonsmooth problems

Multigrid methods have been shown to be an efficient tool for solving partial differential equations. In this paper, the idea of a multigrid method for nonsmooth problems is presented based on techniques from piecewise linear differentiation. In detail, the original nonsmooth problem is approximated by a sequence of piecewise linear models, which can be written in abs-normal form by using addit...

متن کامل

A Piecewise Approximate Method for Solving Second Order Fuzzy Differential Equations Under Generalized ‎Differentiability‎

In this paper a numerical method for solving second order fuzzy differential equations under generalized differentiability is proposed. This method is based on the interpolating a solution by piecewise polynomial of degree 4 in the range of solution. Moreover we investigate the existence, uniqueness and convergence of approximate solutions. Finally the accuracy of piecewise approximate method b...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2013