NARX with multiple Inputs

Dear All;

I am trying use NARX for time series prediction, the problem that i have multiple inputs and i receiving the following error:

Error using preparets (line 185)

The number of input signals does not match network's non-feedback inputs.

Error in NN_Prediction (line 48)

[Xs,Xi,Ai,Ts] = preparets(net,inputSeries,{},targetSeries);

How can i resolve it ?

MATLAB Code:

clc ;

clear ;

%% Read input and output values

data=readtable ('FoulingData.xlsx','sheet','AI_Data');

I1=data.DP_Tube;

I2=data.Tout_Tube;

I3=data.Tout_Shell;

I=[I1 I2 I3];

I=I'

T=data.Fouling;

T=T'

% Normalization between -1 and 1

[In,ps] = mapminmax(I);

[Tn,ts] = mapminmax(T);

X=num2cell(In);

T=num2cell(Tn);

%Data Preparation

N = 3; % Multi-step ahead prediction

% Input and target series are divided in two groups of data:

% 1st group: used to train the network

inputSeries = X(:,1:end-N);

targetSeries = T(1:end-N);

% 2nd group: this is the new data used for simulation. inputSeriesVal will

% be used for predicting new targets. targetSeriesVal will be used for

% network validation after prediction

inputSeriesVal = X(:,end-N+1:end);

targetSeriesVal = T(end-N+1:end); % This is generally not available

% Network Architecture

delay = 2;

neuronsHiddenLayer = 10;

% Network Creation

net = narxnet(1:delay,1:delay,neuronsHiddenLayer);

net.numInputs = 3; % adding an input

net.inputConnect =[1 1 1; 0 0 0]; %connecting 3 inputs to the first layer

[Xs,Xi,Ai,Ts] = preparets(net,inputSeries,{},targetSeries);

net = train(net,Xs,Ts,Xi,Ai);

view(net)

Y = net(Xs,Xi,Ai);

% Performance for the series-parallel implementation, only

% one-step-ahead prediction

perf = perform(net,Ts,Y);

% Multistep prediction

[Xs1,Xio,Aio] = preparets(net,inputSeries(1:end-delay),{},targetSeries(1:end-delay));

[Y1,Xfo,Afo] = net(Xs1,Xio,Aio);

[netc,Xic,Aic] = closeloop(net,Xfo,Afo);

[yPred,Xfc,Afc] = netc(inputSeriesVal,Xic,Aic);

multiStepPerformance = perform(net,yPred,targetSeriesVal);

view(netc)

figure;

plot([cell2mat(targetSeries),nan(1,N);

nan(1,length(targetSeries)),cell2mat(yPred);

nan(1,length(targetSeries)),cell2mat(targetSeriesVal)]')

legend('Original Targets','Network Predictions','Expected Outputs')

[Sunil Patidar]

Can you try removing the bellow mentioned two lines of codes as this might be changing your network's structure.

net.numInputs = 3; % adding an input
net.inputConnect  =[1 1 1; 0 0 0]; %connecting 3 inputs to the first layer

Also, Here is a link to an example workflow of how to perform prediction with a closed loop network. While it is not your exact workflow it may be a helpful example:

https://www.mathworks.com/help/deeplearning/ref/preparets.html