estimate constrained values of a lookup table -凯发k8网页登录

open a parameter estimation session.

in the simulink model, select parameter estimator from the apps tab, in the gallery, under control systems to open a session with the name lookup_increasing in the parameter estimator.

create an experiment and import the i/o data.

on the parameter estimation tab, click new experiment. type [time1,ydata1] in outputs and [time1,xdata1] in inputs of the edit experiment dialog box. click ok. a new experiment with name exp is created in the experiments area of the app. rename the experiment estimationdata by right-clicking the default experiment name, exp, and selecting rename. for more information, see .

run an initial simulation to view the measured data, simulated model values, and the initial table values by typing the following commands at the matlab^® prompt.

sim('lookup_increasing')
figure(1); plot(xdata1,ydata1,'m*',xout,yout,'b^')
hold on; plot(-5:5,cumsum(ydelta),'k','linewidth',2)
xlabel('input data'); ylabel('output data');
legend('measured data','initial simulated data','initial table output')

the initial table output values and simulated data do not match the measured data.

select parameter for estimation.

on the parameter estimation tab, click select parameters. the edit: estimated parameters dialog box opens. in the parameters tuned for all experiments panel, click select parameters to open the select model variables dialog box. check the box next to ydelta, and click ok.

the ydelta values are selected for estimation by default in the edit: estimated parameters dialog box.

apply a monotonically increasing constraint on the table output values. for more details about the table, see lookup table output.

in the edit: estimated parameters dialog box, click the arrow next to the ydelta values. in the expanded menu, set minimum ydelta values to [-inf,zeros(1,10)]. thus, while the first value in ydelta can by anything, subsequent values which are the difference between adjacent table output values, must be positive.

select estimationdata experiment for estimation.

on the parameter estimation tab, click select experiment. by default, estimationdata is selected for estimation. if not, check the box under the estimation column, and click ok.

estimate the table values using default settings.

on the parameter estimation tab, click estimate.

the parameter trajectory plot shows the change in the parameter values at each iteration.

the estimation progress report shows the iteration number, number of times the objective function is evaluated, and value of the cost function at the end of each iteration.

the estimated parameters are saved in a new variable, estimatedparams, in the results area of the app. to view the estimated parameters, right-click estimatedparams and select open.

the estimated ydelta(2:end) values are positive. thus, the output of the table, which is the cumulative sum of the values stored in ydelta, is monotonically increasing.

create an experiment to use for validation and import the validation i/o data.

on the parameter estimation tab, click new experiment. type [time2,ydata2] in outputs and [time2,xdata2] in inputs of the edit experiment dialog box. name the experiment validationdata by right-clicking the default experiment name, exp, in the experiments area of the app, and selecting rename. for more information, see .

select the experiment for validation.

click select experiments on the parameter estimation tab. the validationdata experiment is selected for estimation by default. clear estimation and select the box for validation.

select the results to validate.

on the validation tab, click select results to validate. clear use current parameter values, select estimatedparams, and click ok.

select the plots to display during validation.

the parameter estimator displays the experiment plot after validation by default. add the residuals plot by selecting the corresponding box on the validation tab.

click validate.

examine the plots.