Artificial Neural Network Prediction Code in Matlab, please help !!!
25 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
dau
el 26 de Nov. de 2013
Respondida: Shah
el 29 de Mayo de 2018
Dear ! I have a data set with a column is rainfall and the next column is water level (1992-2006) as: https://docs.google.com/file/d/0By5tEg03EXCpd09rcG5hTWRzMHM/edit I would like to ask your about what are the Inputs values and Target Values? My friend told me that to Predict water level in the future is not need Rainfall data but only need Water level for recent years ago. I am confusing about that since Neural Network is needing an Input and Target values. I have a code for predict water level but rely on the input data has known, within 1992-2006 only. What I am need is using data from 1992-2006 to predict in the year 2014 example. I know you are very excellent about that, please fix again help me the code, thanks.
% Data_Inputs=xlsread('1.xls'); % Import file
%The training data sample are randmonized by using the function'randperm'
%Shuffling_Inputs=Data_Inputs(randperm(end),1:2); % integers (training sample)
Training_Set=Data_Inputs(1:end,1);%specific training set
Target_Set=Data_Inputs(1:end,2); %specific target set
Input=Training_Set'; %Convert to row
Target=Target_Set'; %Convert to row
X = con2seq(Input); %Convert to cell
T = con2seq(Target); %Convert to cell
%%2. Data preparation
N = 365; % 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);
inputSeriesVal = X(end-N+1:end);
targetSeriesVal = T(end-N+1:end);
% Create a Nonlinear Autoregressive Network with External Input
delay = 2;
inputDelays = 1:2;
feedbackDelays = 1:2;
hiddenLayerSize = 100;
net = narxnet(inputDelays,feedbackDelays,hiddenLayerSize);
% Prepare the Data for Training and Simulation
% The function PREPARETS prepares timeseries data for a particular network,
% shifting time by the minimum amount to fill input states and layer states.
% Using PREPARETS allows you to keep your original time series data unchanged, while
% easily customizing it for networks with differing numbers of delays, with
% open loop or closed loop feedback modes.
[inputs,inputStates,layerStates,targets] = preparets(net,inputSeries,{},targetSeries);
% Setup Division of Data for Training, Validation, Testing
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;
% Train the Network
[net,tr] = train(net,inputs,targets,inputStates,layerStates);
% Test the Network
outputs = net(inputs,inputStates,layerStates);
errors = gsubtract(targets,outputs);
performance = perform(net,targets,outputs)
% View the Network
view(net)
% Plots
% Uncomment these lines to enable various plots.
%figure, plotperform(tr)
%figure, plottrainstate(tr)
%figure, plotregression(targets,outputs)
%figure, plotresponse(targets,outputs)
%figure, ploterrcorr(errors)
%figure, plotinerrcorr(inputs,errors)
% Closed Loop Network
% Use this network to do multi-step prediction.
% The function CLOSELOOP replaces the feedback input with a direct
% connection from the outout layer.
netc = closeloop(net);
netc.name = [net.name ' - Closed Loop'];
view(netc)
[xc,xic,aic,tc] = preparets(netc,inputSeries,{},targetSeries);
yc = netc(xc,xic,aic);
closedLoopPerformance = perform(netc,tc,yc)
% Early Prediction Network
% For some applications it helps to get the prediction a timestep early.
% The original network returns predicted y(t+1) at the same time it is given y(t+1).
% For some applications such as decision making, it would help to have predicted
% y(t+1) once y(t) is available, but before the actual y(t+1) occurs.
% The network can be made to return its output a timestep early by removing one delay
% so that its minimal tap delay is now 0 instead of 1. The new network returns the
% same outputs as the original network, but outputs are shifted left one timestep.
nets = removedelay(net);
nets.name = [net.name ' - Predict One Step Ahead'];
view(nets)
[xs,xis,ais,ts] = preparets(nets,inputSeries,{},targetSeries);
ys = nets(xs,xis,ais);
earlyPredictPerformance = perform(nets,ts,ys)
%%5. Multi-step ahead prediction
inputSeriesPred = [inputSeries(end-delay+1:end),inputSeriesVal];
targetSeriesPred = [targetSeries(end-delay+1:end), con2seq(nan(1,N))];
[Xs,Xi,Ai,Ts] = preparets(netc,inputSeriesPred,{},targetSeriesPred);
yPred = netc(Xs,Xi,Ai);
perf = perform(net,yPred,targetSeriesVal);
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');
0 comentarios
Respuesta aceptada
Greg Heath
el 30 de Nov. de 2013
1. Plot
a. Rainfall vs time
b. Water Level vs time
c. Water Level vs Rainfall
d. Autocorrelation function of Water Level
e. Crosscorrelation function of Water Level and Rainfall
2. Calculate
a. Significant values and lags of the Water Level autocorrelation function
b. Significant values and lags of the Water-Level/Rainfall crosscorrelation function
3. Determine which time-series functions yield the best predictions. DO NOT use the default 'dividerand' which destroys correlations. Use 'divideblock'.
a. Timedelaynet
b. Narnet
c. Narxnet
4. If b and/or c have a satisfactory openloop performance, close the
loop and continue training with the original data.
Hope this helps.
Thank you for formally accepting my answer
Greg
3 comentarios
Greg Heath
el 4 de Dic. de 2013
You are very lucky because I have posted many time-series examples in NEWSGROUP and ANSWERS.
Greg Heath
el 1 de Sept. de 2016
THIS IS A SIMPLE REGRESSION SCENARIO. THE APPROPRIATE FUNCTION TO USE IS FITNET. FOR THE HELP AND DOCUMENTATION ASSISTANCE USE THE COMMANDS
HELP FITNET
DOC FITNET
THEN SEARCH THE NEWGROUP AND ANSWERS FOR POSTED EXAMPLES USING THE SEARCHWORDS
FITNET
GREG FITNET
HOPE THIS HELPS.
GREG
Más respuestas (1)
Shah
el 29 de Mayo de 2018
i have the same issue, my dataset consist of colums/input parameters (Total water,Extr water mm,Cum Runoff mm,Drainage Mm Precipitation,Irrigation #,Irrig effect mm,Water table cm,Surface runoff,Pot ET mm/d,Evapotrans mm/d, Transpir mm/d,Transpiration) and i want to predict future values of peak discharge=Q = PIA P is runoff coefficient which depends on the characteristics of the catchment area. It is a ratio of runoff: rainfall. I is the intensity of rainfall in m/sec for the duration at least equal to “time of concentration”. And A is area of the catchment in m2.
Time of concentration:
It is the time taken by the rain water falling at the remotest point of the drainage basin to reach the discharge measurement point. It is given by the formula
tc = 0.000324 L0.77/S0.358
where tc is time of concentration in hours,
L is length of the drainage basin in m measured along river channel upto the farthest point on the periphery of the basin.
S is average slope of the basin from the farthest point to the discharge measuring point under consideration.
i am applying ANN/NN to predict peak discharge. how i can predict using above given inputs?
0 comentarios
Ver también
Categorías
Más información sobre Modeling and Prediction with NARX and Time-Delay Networks en Help Center y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!