drydemo.html

交流 模糊控制 交流 模糊控制

winH1 = gcf;</pre><pre xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd" style="color:gray; font-style:italic;">
Selecting input 1 ...
ANFIS model 1: y(k-1) --&gt; trn=0.2043, chk=0.1888
ANFIS model 2: y(k-2) --&gt; trn=0.3819, chk=0.3541
ANFIS model 3: y(k-3) --&gt; trn=0.5245, chk=0.4903
ANFIS model 4: y(k-4) --&gt; trn=0.6308, chk=0.5977
ANFIS model 5: u(k-1) --&gt; trn=0.8271, chk=0.8434
ANFIS model 6: u(k-2) --&gt; trn=0.7976, chk=0.8087
ANFIS model 7: u(k-3) --&gt; trn=0.7266, chk=0.7349
ANFIS model 8: u(k-4) --&gt; trn=0.6215, chk=0.6346
ANFIS model 9: u(k-5) --&gt; trn=0.5419, chk=0.5650
ANFIS model 10: u(k-6) --&gt; trn=0.5304, chk=0.5601
Currently selected inputs: y(k-1)

Selecting input 2 ...
ANFIS model 11: y(k-1) y(k-2) --&gt; trn=0.1085, chk=0.1024
ANFIS model 12: y(k-1) y(k-3) --&gt; trn=0.1339, chk=0.1283
ANFIS model 13: y(k-1) y(k-4) --&gt; trn=0.1542, chk=0.1461
ANFIS model 14: y(k-1) u(k-1) --&gt; trn=0.1892, chk=0.1734
ANFIS model 15: y(k-1) u(k-2) --&gt; trn=0.1663, chk=0.1574
ANFIS model 16: y(k-1) u(k-3) --&gt; trn=0.1082, chk=0.1077
ANFIS model 17: y(k-1) u(k-4) --&gt; trn=0.0925, chk=0.0948
ANFIS model 18: y(k-1) u(k-5) --&gt; trn=0.1533, chk=0.1531
ANFIS model 19: y(k-1) u(k-6) --&gt; trn=0.1952, chk=0.1853
Currently selected inputs: y(k-1) u(k-4)

Selecting input 3 ...
ANFIS model 20: y(k-1) u(k-4) y(k-2) --&gt; trn=0.0808, chk=0.0822
ANFIS model 21: y(k-1) u(k-4) y(k-3) --&gt; trn=0.0806, chk=0.0836
ANFIS model 22: y(k-1) u(k-4) y(k-4) --&gt; trn=0.0817, chk=0.0855
ANFIS model 23: y(k-1) u(k-4) u(k-1) --&gt; trn=0.0886, chk=0.0912
ANFIS model 24: y(k-1) u(k-4) u(k-2) --&gt; trn=0.0835, chk=0.0843
ANFIS model 25: y(k-1) u(k-4) u(k-3) --&gt; trn=0.0609, chk=0.0604
ANFIS model 26: y(k-1) u(k-4) u(k-5) --&gt; trn=0.0848, chk=0.0867
ANFIS model 27: y(k-1) u(k-4) u(k-6) --&gt; trn=0.0890, chk=0.0894
Currently selected inputs: y(k-1) u(k-3) u(k-4)

Elapsed time = 1.091000
</pre><img xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd" src="drydemo_img07.gif"><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd" style="color:#990000; font-weight:bold; font-size:medium; page-break-before: auto;"><a name=""></a></p><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd">For input selection, another more computation intensive approach is to do
an exhaustive search on all possible combinations of the input
candidates. The function that performs exhaustive search is exhsrch,
which selects 3 inputs from 10 candidates.  However, exhsrch usually
involves a significant amount of computation if all combinations are
tried.  For instance, if 3 is selected out of 10, the total number of
ANFIS models is C(10, 3) = 120.
</p><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd">Fortunately, for dynamical system identification, we do know that the inputs should not come from either of the following two sets of input candidates exclusively:
</p><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd">Y = {y(k-1), y(k-2), y(k-3), y(k-4)}
</p><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd">U = {u(k-1), u(k-2), u(k-3), u(k-4), u(k-5), u(k-6)}
</p><p xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd">A reasonable guess would be to take two inputs from Y and one from U to
form the inputs to ANFIS; the total number of ANFIS models is then
C(4,2)*6=36, which is much less.  The above plot shows that the selected
inputs are y(k-1), y(k-2) and u(k-3), with a training RMSE of 0.0474 and
checking RMSE of 0.0485, which are better than ARX models and ANFIS via
sequential forward search.
</p><pre xmlns:mwsh="http://www.mathworks.com/namespace/mcode/v1/syntaxhighlight.dtd" style="position: relative; left:30px">group1 = [1 2 3 4];	<span style="color:green">% y(k-1), y(k-2), y(k-3), y(k-4)</span>
group2 = [1 2 3 4];	<span style="color:green">% y(k-1), y(k-2), y(k-3), y(k-4)</span>
group3 = [5 6 7 8 9 10];	<span style="color:green">% u(k-1) through y(k-6)</span>

anfis_n = 6*length(group3);
index = zeros(anfis_n, 3);
trn_error = zeros(anfis_n, 1);
chk_error = zeros(anfis_n, 1);
<span style="color:green">% ======= Training options </span>
mf_n = 2;
mf_type = <span style="color:#B20000">'gbellmf'</span>;
epoch_n = 1;
ss = 0.1;
ss_dec_rate = 0.5;
ss_inc_rate = 1.5;
<span style="color:green">% ====== Train ANFIS with different input variables</span>
fprintf(<span style="color:#B20000">'\nTrain %d ANFIS models, each with 3 inputs selected from 10 candidates...\n\n'</span>,<span style="color:blue">...</span>
    anfis_n);
model = 1;
<span style="color:blue">for</span> i=1:length(group1),
    <span style="color:blue">for</span> j=i+1:length(group2),
        <span style="color:blue">for</span> k=1:length(group3),
            in1 = deblank(input_name(group1(i), :));
            in2 = deblank(input_name(group2(j), :));
            in3 = deblank(input_name(group3(k), :));
            index(model, :) = [group1(i) group2(j) group3(k)];
            trn_data = data(1:trn_data_n, [group1(i) group2(j) group3(k) size(data,2)]);
            chk_data = data(trn_data_n+1:trn_data_n+300, [group1(i) group2(j) group3(k) size(data,2)]);
            in_fismat = genfis1(trn_data, mf_n, mf_type);
            [trn_out_fismat t_err step_size chk_out_fismat c_err] = <span style="color:blue">...</span>
                anfis(trn_data, in_fismat, <span style="color:blue">...</span>
                [epoch_n nan ss ss_dec_rate ss_inc_rate], <span style="color:blue">...</span>
                [0 0 0 0], chk_data, 1);
            trn_error(model) = min(t_err);
            chk_error(model) = min(c_err);
            fprintf(<span style="color:#B20000">'ANFIS model = %d: %s %s %s'</span>, model, in1, in2, in3);
            fprintf(<span style="color:#B20000">' --&gt; trn=%.4f,'</span>, trn_error(model));
            fprintf(<span style="color:#B20000">' chk=%.4f'</span>, chk_error(model));
            fprintf(<span style="color:#B20000">'\n'</span>);
            model = model+1;
        <span style="color:blue">end</span>
    <span style="color:blue">end</span>
<span style="color:blue">end</span>

<span style="color:green">% ====== Reordering according to training error</span>
[a b] = sort(trn_error);
b = flipud(b);		<span style="color:green">% List according to decreasing trn error</span>
trn_error = trn_error(b);
chk_error = chk_error(b);
index = index(b, :);

<span style="color:green">% ====== Display training and checking errors</span>
x = (1:anfis_n)';
subplot(2,1,1);
plot(x, trn_error, <span style="color:#B20000">'-'</span>, x, chk_error, <span style="color:#B20000">'-'</span>, <span style="color:blue">...</span>
    x, trn_error, <span style="color:#B20000">'o'</span>, x, chk_error, <span style="color:#B20000">'*'</span>);
tmp = x(:, ones(1, 3))';
X = tmp(:);
tmp = [zeros(anfis_n, 1) max(trn_error, chk_error) nan*ones(anfis_n, 1)]';
Y = tmp(:);
hold on; 
plot(X, Y, <span style="color:#B20000">'g'</span>); 
hold off;
axis([1 anfis_n -inf inf]);
set(gca, <span style="color:#B20000">'xticklabel'</span>, []);

<span style="color:green">% ====== Add text of input variables</span>
<span style="color:blue">for</span> k = 1:anfis_n,
    text(x(k), 0, <span style="color:blue">...</span>
        [input_name(index(k,1), :) <span style="color:#B20000">' '</span> <span style="color:blue">...</span>
            input_name(index(k,2), :) <span style="color:#B20000">' '</span> <span style="color:blue">...</span>
            input_name(index(k,3), :)]);