Identifies the virtual sensor starting from the measured system inputs and outputs
Method of virtualsensor object.
This function identifies a pwas virtual sensor starting from measures of system inputs (u), system measurable outputs (y) and system outputs (z) measurable only for a finite time window. It can be used for pwas discrete time system identification too. The observer is identified starting from T samples of input u, of the measurable output y and of the unmeasurable output z of the observed system. The weigths of the pwas map are estimated with a Regularized Least Square regression (see function ridgeRegression).
object = identify(object,u,y,z,P,[ut],[yt],[zt])
u is a [npoints x nu] matrix whose i-th row corresponds to the i-th measurement of the nu inputs of the system. The same is for y (for the measurable outputs) and z (for the unmeasurable outputs). u, y and z can also be cell arrays whose i-th element correspond to a dataset obtained iterating the system starting from different initial conditions. P is an array containing the number of subdivisions per dimension of the simplicial partition. The domain of the pwas function is estimated from data. ut, yt and zt (optional) are matrices or cell arrays just as u, y and z, but they contain a test dataset which can be used to estimate the best value of the regularization parameter lambda. If they are not provided the optimal value for lambda is computed with a Generalized Cross Validation technique.
object = identify(object,u,y,z,P,[ut],[yt],[zt],D)
The domain is provided from outside the function. D is a matrix in the form:
object = identify(object,u,y,z,P,[ut],[yt],[zt],D,options)
options is a structure with the following fields:
- order: Tikhonov regularization order. It can be 0 or 1, default 0.
- lambda: regularisation weight. It must be a scalar > 0. If it is not provided it is estimated inside the function.
- nsplits: number of splits of the dataset used to solve the regularised least square problem with an iterative approach (in order to save memory). In practice, instead of solving , you can solve , in which and . nsplits corresponds to n. A low value of nsplits makes ridgeRegression faster but it can result in memroy occupation problems. Default value, 1.
- constraints: only if reducedComplexity = 0. array of structures defining optional constraints on the pwas function you want to obtain. Each element of the array defines different constraints. Tre structures have the following fields:
- type: it can be either 'bounds' or 'equality'. If it is 'bounds' constraints in the form lbound <= f(x) <= ubound are imposed. lbound and ubound can be provided through fields lbound and ubound. If they are not provided they are imposed as the minimum and maximum values of the data Y and Yt contained in the dataset. If type is equality, equality constraints in the form f(x) = k can be imposed for x lying on a hyper-plane parallel to the domain components.
- lbound: lower bound for constraints of type 'bounds'
- ubound: upper bound for constraints of type 'bounds'
- variables: needed for equality constraints. It is an array of strings indicating for which set of points you want to impose the constraints. For example if you want to impose a constraint on f, for x1 = 3 and x3 = 2, variables must be ['x1 = 3', 'x3 = 2'].
- value: needed for equality constraints. It is the value you want the function to assume in correspondence of the points defined in field variables. To impose the constraint f(x) = 0, for x2 = 1 and x4 = 5 you gave to set variables = ['x2 = 1', 'x4 = 5'] and value = 0. * gamma: domain expansion parameter. If the domain D is not provided from outside the function, it is automatically extrapolated from data X. The tight domain (Dt) contains exactly data X. Such domain is grown of a factor gamma such as D = gamma Dt. Default value 1.1. * solver: string specifying the solver you want to use to solve the QP problem (this is necessary only for constrained ridge regression, otherwise the solution is analytical). Possible choices are 'quadprog' (default), 'cvx', 'cplex', 'yalmip' 'clp'. * verbose: if verbose is set to 1, messages are displayed indicating the status of the ridge regression process.
- Tomaso Poggi (firstname.lastname@example.org)
Copyright is with:
- Copyright (C) 2010 University of Genoa, Italy.