if any one have code related to this topic "Modeling of Electricity Consumption Forecasting Using Artificial Neural Network and Support vector Machine" kindly let me know please.

3 次查看（过去 30 天）

Saba Yousaf 2018-9-29

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/421399-if-any-one-have-code-related-to-this-topic-modeling-of-electricity-consumption-forecasting-using-a

评论： Stephan 2018-9-29

My thesis topic is "Modeling of Electricity Consumption Forecasting Using Artificial Neural Network and Support vector Machine". i used following code to predict the consumption results like R value, MSE , MAPE, MAE etc.

Opts = detectImportOptions('datasetn.txt'); Data = readtable('datasetn.txt', Opts);

input = {'T', 'Wdi', 'S', 'R','W'}; i = Data(:, input); p=table2array(i) %input variables

kWh = Data.kWh; kWh = strrep(kWh, ',', ''); kWh = strjoin(kWh, '\n'); kWh = str2num(kWh); Data.kWh = kWh

respone=Data.kWh % target response

net=feedforwardnet(10); net = configure(net, p', respone') [net, info] = train(net, p',respone'); % p input and respone the target / the info will contain the masks

y1=net(p') %predicted response Rt=corrcoef(y1, respone') %correlation coefficent MSE=mean((respone'-y1).^2); % overall MSE mAE=mean(respone'-y1))/365 %overall MAE

but i dont know what i have to write in experiment chapter because all others papers that i read almost 20 to 30, are much more advanced and they used some ANN methods like "Artificial Neural Network Based Models Generated by Multi Objective Genetic Algorithm" and many more . but i just train using "Levenberg-Marquardt backpropagation". now kindly let me know please what i have to discuss here.