Hi Anderson This is very helpful and I am just planning to get pH at the concrete surface by using imaging any idea how to extract the data from the concrete surface by using imaging
oi anderson eu tenho uma otimização de um processo que possuo 1 variavel indenpendente e 8 variavel dependente, consigo tomizar o processo usando a metodologia de superficie de resposta?
Olá Bruno! Neste caso fica meio complicado utilizar a MSR pois você só tem 1 variável independente, dai não tem como gerar superfícies. Você só teria uma linha ao invés de um mesh. E você precisaria de mais de 2 níveis (-1 e +1), acho que no mínimo 9 níveis. Você vai conseguir resolver este tipo de problema com regressão múltipla ou com MANOVA. Procure por "Multivariate Multiple Regression" ou apenas "Multivariate Regression" (estes termos se confundem bastante na literatura, e muitas vezes são tratados como a mesma coisa). Minitab, Origin, R, STATISTICA fazem este tipo de regressão para você. Se você tivesse 2 variáveis independentes, dai daria pra fazer com a MSR, gerando 1 modelo (e uma superfície) para cada variável dependente. Depois você poderia utilizar a função desejabilidade para encontrar o ponto de ótimo entre todos os outputs.
Hello! I have not worked with ternary microemulsion systems. The most I did was draw the ternary graphs, but that's all. I remember works by some colleagues who compared the tie lines in systems with different solvents, but it was really a long time ago, and that's all I will remember
Excelente vídeo, Anderson. Obrigada. Estou tendo dificuldade na edição da legenda agora; há como modifica-la sem alterar as cores no gráfico de superfície de resposta? A legenda ficou muito grande e gostaria de encurta-la. Obrigada.
Olá @Cleuciane Tillvitz do Nascimento Você consegue alterar o tamanho da caixa da janela clicando sobre a legenda, e reduzindo o tamanho da caixa. Mas isto pode deixar tudo muito apertado, com o texto muito grande. Para controlar o tamanho da fonte dentro da legenda, clique com o botão direito sobre a caixa da legenda, vá em properties e altere o campo Size. Dai é só encontrar uma boa relação entre o tamanho da caixa e o tamanho da fonte
You get the model formula from the software you used to make the adjustment. As you have 2 independent variables, you will have two different models. Regarding the number of levels, after the model has been obtained and validated, it no longer makes a difference whether the model was built with 2 or 3 levels
Nice tutorial! But hey, what if I have to plot a surface whose model has more than two variables? I'm working with a design which contains x, y, z, xy, xz, yz and xyz. How can I use more than two variables if origin only lets me have x and y while z is used for the response?
Hello @Lichewitz ! Thank you!! Well, you just have to fix one of the variables at a desired value (generally is the central point)! let's say you have this equation: Z = mean + A*x1 + B*x2 + C*x3 If you want the plot of x1 and x3, you choose a value for x2 (which must be within the limits you used to perform the regression). Then you have: Z = (mean + B*x2) + A*x1 + C*x3 This term (mean + B*x2) is a constant in this example Another example: If you want the plot of x2 and x3, you choose a value for x1 (which must be within the limits you used to perform the regression). Then you have: Z = (mean + A*x1) + B*x2 + C*x3 This term (mean + A*x1) is a constant in this example
@@andersoncanteli Got it! But assuming I have equation with, say, four variables, and I want to select only two, when the time comes for me to input the equation in the matrix values, do I need to multiply variables I'm not showing by their coded value, changing the intercept value? For example, if my model is (R = mean + ax + by + cz + dw), and I want to plot x and y while keeping z and w at their lower level... do I have to substitute z and w for -1, resulting in the equation (R = mean -c -d + ax +by) ? Or am I missing something here? And thanks again, your tutorial was very very helpful!
@@dipali4525 Weird, it should work exactly like the tutorial 😕. After I published the video, I never touched the files again, so the file is correct and it worked 5 years ago (as you can see in the video). Unless you provide some kind of detail, the only thing I can suggest is that you try to repeat exactly the same steps used in the tutorial, using the same equation I used. If repeating the tutorial works, it means that the method is working.
@@dipali4525 I'm not 100% sure as I haven't used Origin in a few years, but you might need to right click on the surface and go to styles (or something like that). In the window that will open, you should have the option to remove/add grid on the surface in one of the tabs. Maybe you just need to add some color in the grid. It must be something simple
@@mustaphabello8016 maybe you just have to choose OriginPro... You can also open OriginPro first, then go to File -> Open -> go to directory that the file is -> open it
![HIWWHEE | OHM [Official MV]](http://i.ytimg.com/vi/OJJLtTLqLag/mqdefault.jpg)
Thanks for your help! The best tutorial for this kind of plots!! :)
I'm glad you liked it :)
Hi Anderson
This is very helpful and I am just planning to get pH at the concrete surface by using imaging any idea how to extract the data from the concrete surface by using imaging
HELLO How to plot bottom surface in version 2024b i cant find the surface projection botton. kindly help in this regard thanks
Hello! Unfortunately I don't have access to recent versions of Origin, so I won't be able to help you this time
oi anderson eu tenho uma otimização de um processo que possuo 1 variavel indenpendente e 8 variavel dependente, consigo tomizar o processo usando a metodologia de superficie de resposta?
Olá Bruno! Neste caso fica meio complicado utilizar a MSR pois você só tem 1 variável independente, dai não tem como gerar superfícies. Você só teria uma linha ao invés de um mesh. E você precisaria de mais de 2 níveis (-1 e +1), acho que no mínimo 9 níveis. Você vai conseguir resolver este tipo de problema com regressão múltipla ou com MANOVA. Procure por "Multivariate Multiple Regression" ou apenas "Multivariate Regression" (estes termos se confundem bastante na literatura, e muitas vezes são tratados como a mesma coisa). Minitab, Origin, R, STATISTICA fazem este tipo de regressão para você.
Se você tivesse 2 variáveis independentes, dai daria pra fazer com a MSR, gerando 1 modelo (e uma superfície) para cada variável dependente. Depois você poderia utilizar a função desejabilidade para encontrar o ponto de ótimo entre todos os outputs.
hello mam , how to compare or quantify 2 different ternary systems of microemulsion region
Hello! I have not worked with ternary microemulsion systems. The most I did was draw the ternary graphs, but that's all. I remember works by some colleagues who compared the tie lines in systems with different solvents, but it was really a long time ago, and that's all I will remember
Excelente vídeo, Anderson. Obrigada. Estou tendo dificuldade na edição da legenda agora; há como modifica-la sem alterar as cores no gráfico de superfície de resposta? A legenda ficou muito grande e gostaria de encurta-la. Obrigada.
Olá @Cleuciane Tillvitz do Nascimento
Você consegue alterar o tamanho da caixa da janela clicando sobre a legenda, e reduzindo o tamanho da caixa. Mas isto pode deixar tudo muito apertado, com o texto muito grande. Para controlar o tamanho da fonte dentro da legenda, clique com o botão direito sobre a caixa da legenda, vá em properties e altere o campo Size.
Dai é só encontrar uma boa relação entre o tamanho da caixa e o tamanho da fonte
@@andersoncanteli Perfeito. Muito obrigada.
Can you teach me how to plot multi 3D section(plot 3D section at different Z value)
How to get the matrix ?
how to get formula for two independ variables and three levels
You get the model formula from the software you used to make the adjustment. As you have 2 independent variables, you will have two different models. Regarding the number of levels, after the model has been obtained and validated, it no longer makes a difference whether the model was built with 2 or 3 levels
Mine is Originlab 2016, I couldn't find the experimental data from layer content in 3d surface graph.
I don't have Originlab 2016, so I don't know where it could be :/ Sorry
Anderson Canteli It is solved, thank you.
Nice tutorial! But hey, what if I have to plot a surface whose model has more than two variables? I'm working with a design which contains x, y, z, xy, xz, yz and xyz. How can I use more than two variables if origin only lets me have x and y while z is used for the response?
Hello @Lichewitz
! Thank you!!
Well, you just have to fix one of the variables at a desired value (generally is the central point)!
let's say you have this equation:
Z = mean + A*x1 + B*x2 + C*x3
If you want the plot of x1 and x3, you choose a value for x2 (which must be within the limits you used to perform the regression).
Then you have:
Z = (mean + B*x2) + A*x1 + C*x3
This term (mean + B*x2) is a constant in this example
Another example:
If you want the plot of x2 and x3, you choose a value for x1 (which must be within the limits you used to perform the regression).
Then you have:
Z = (mean + A*x1) + B*x2 + C*x3
This term (mean + A*x1) is a constant in this example
@@andersoncanteli Got it! But assuming I have equation with, say, four variables, and I want to select only two, when the time comes for me to input the equation in the matrix values, do I need to multiply variables I'm not showing by their coded value, changing the intercept value?
For example, if my model is (R = mean + ax + by + cz + dw), and I want to plot x and y while keeping z and w at their lower level... do I have to substitute z and w for -1, resulting in the equation (R = mean -c -d + ax +by) ? Or am I missing something here? And thanks again, your tutorial was very very helpful!
@@lichewitz8905 If you are using the coded equation, as seems to be the case, then it is as you described
Thanks very much :)
thank you so much especially for the color theme !
Hello, i downloaded the link for colour palette. But it is not working
Hello! Are you using the same version that I used in this tutorial?
Yes, I am using the same version
@@dipali4525 Weird, it should work exactly like the tutorial 😕. After I published the video, I never touched the files again, so the file is correct and it worked 5 years ago (as you can see in the video).
Unless you provide some kind of detail, the only thing I can suggest is that you try to repeat exactly the same steps used in the tutorial, using the same equation I used. If repeating the tutorial works, it means that the method is working.
Anderson, Thanks. I am able to copy the color format. However, the gridlines on my surface plot are missing. Can you please suggest, how to add?
@@dipali4525 I'm not 100% sure as I haven't used Origin in a few years, but you might need to right click on the surface and go to styles (or something like that). In the window that will open, you should have the option to remove/add grid on the surface in one of the tabs. Maybe you just need to add some color in the grid. It must be something simple
Please Can I have the link of figure . Thank you for your help.
goo.gl/WKFoQz
@@andersoncanteli my computer is asking me to choose the program or application to open the link
@@mustaphabello8016 maybe you just have to choose OriginPro... You can also open OriginPro first, then go to File -> Open -> go to directory that the file is -> open it
thanks,
just get minitab
In minitab have not the plot same as in video. Only surface plot or main plot
in minitab you don't have a many choice of color when you plot 3D