Locate in a cell array rows which share same strings and elements

Suppose I have a cell array in MATLAB of the following type:

    Station_Name  Node_Number  Generation_type  Generated_Power (MW)

    Barry              5        Gas             120  
    Brigg              3        Nuclear         135
    Brixton            1        Gas             110 
    Shoreham           2        Solar           90 
    Kelsey             5        Gas             85 
    Jinnie             4        Nuclear         65
    Pollock            2        Gas             150
    Surret             2        Gas             160

(The actual problem contains much more stations, nodes, and generation types though I am writing this here for simplification).

I want to sort the data to an array of the following type:

    Node_Number    Generation_type      Total_Generated_Power

    1              Solar                 
    1              Gas  
    2              Hydro 
    2              Gas
    .
    .
    .
    3        
    4
    5

So I want to create a script that will automatically produce an array with a different row for each generation type, for each one of the nodes of the cell array (Actually more than 5 nodes and more than 3 generation types).

My initial thought is that I need to make a for loop that will initially check for the elements that are same at the second column, and for those elements the strings at the third column will be compared to see which are the same. Then for those stations that share the same generation type at the same node their generated power will be added together. Then a new variable will need to be created as the new array which will hold the total generated power for each type in each node.

The for loop would be initialised from 1 till the max number of nodes, and would check each one of the rows of the array. The use of the `strcmp` for finding the stations that share the same string i.e same generation type should also be used, from what I understand.

Any tips on how to approach this would be very helpful. Thank you in advance for your time.

Answer 1

[Walter Roberson]

Consider using findgroups() and splitapply()

... though for something like that I would probably tend to use several unique() calls and accumarray()