Nerd For Tech
Published in

Nerd For Tech

AGSI Data Storage


AGSI Data Storage Report


maXbox Starter 99 — Data representation of gas in storage as a timeline AGSI dataset.

There are many kinds of data scientists:

a) Those who dive in data are the best.

This data science tutorial explains the so called AGSI data storage and his visualisation of the time line. AGSI is the Aggregated Gas Storage Inventory and offers you the possibility to be kept up to date whenever a new service announcement or update from one of our data providers is posted on the website.

The Gas Infrastructure Europe (GIE) is also providing related data such as the Storage Map and Storage Investment Database at

The result of the data will be the chart below:

Pic: 1154_agsi_plot13.png

We use WinHttp.WinHttpRequest, JSONObjects and TEECharts library with loading and testing the plot. Also an API-key is needed, get the key first at:

The data represents gas in storage at the end of the previous gas day. Data is updated every day at 19:30 CET and a second time at 23:00. Before we dive into code this is the main part of the script:

plotform:= getForm2(1400, 600, clsilver, ‘Sciplot4maXbox’);


JSON2Plot(plotform, letGenerateJSON2(HttpResponse));

The main part generates a form, invokes the API and plots the data.

GIE is providing an API (Application Programming Interface) service on its AGSI and ALSI transparency publication platforms.

Using API access, consumer can bypass the AGSI and ALSI website and get hold of the data directly and continuously. It enables to extract, filter, aggregate the data and create any subset as required, without having to download each dataset separately from the website. The API export format is JSON. For example a subset of 150 days:

Pic: 1154_agsi_plot14.png

The published datasets are based on the EIC code mapping table provided to ACER. Storage data is aggregated by company, and country.

With the call I pass country, start-date and amount of days:


All available datasets can be downloaded also in Excel, CSV and JSON format. The data in this report shows an aggregated view — individual datasets by company and storage facility are also accessible.

So lets start with the API call:


This command and script runs WinHttp.WinHttpRequest. When you fail you get a bunch of exceptions like the following:

Exception: WinHttp.WinHttpRequest: The data necessary to complete this operation is not yet available; or you missed a valid key:

AGSIPost: Failed at getting response:403{

“error”: {

“code”: 403,

“message”: “The request is missing a valid API key.”,



The funny thing is the JSON formatted exception. Be also carefully to expose your key as I get from Git: GitGuardian has detected the following Google API Key exposed within your GitHub account.

Next is the formatting of the get-call with a valid API-key request in the function energyStream()

function getEnergyStreamJSON2(AURL, feedstream, aApikey:string): string;

encodURL:= Format(AURL,[HTTPEncode(asp[0]),(asp[1]),asp[2]]);

writeln(encodurl) //debug

hr:= httpRq.Open(‘GET’, encodURL, false);



And where is the fabulous content-type? As far as I understood there are only two places in a web-request where to set a content-type:

  1. The client sets a content type for the body he is sending to the server (e.g. for get and post).
  2. The server sets a content type for the response.

A sender that generates a message containing a payload body has to generate a Content-Type header field in that message unless the intended media type of the enclosed representation is unknown to the sender; otherwise we failed at getting response: 503503 — Service Unavailable.



It means that the content-type HTTP header should be set only for PUT and POST requests. GET requests can have “Accept” headers, which say which types of content the client understands. The server can then use that to decide which content type to send back.

As an option you can also use TALWinInetHttpClient. It is a easy to use WinInet-based protocol and supports HTTPs. HTTP client component which allows to post and get any data from the Web via HTTP protocol.

function TALHTTPClient_Get(aUrl: AnsiString;feedstream, aApikey: string): string;

Var LHttpClient: TALWininetHttpClient; asp: TStringArray;

LHttpClient:= TALWininetHttpClient.create;
asp:= splitStr(feedstream,’,’);
LHttpClient.url:= Format(AURL,[HTTPEncode(asp[0]),(asp[1]),asp[2]]);
LHttpClient.RequestMethod:= HTTPmt_Get; //HTTPrm_Post;

result:= LHttpClient.Get1(LHttpClient.url); //overload;

Any missing or incomplete data is also be visible on AGSI. Next we step to the conversion of our JSON response for the plot with TJSONObject:

function letGenerateJSON2(HttpRqresponseText: string): TJSONArray;
var jo: TJSONObject;

jo:= TJSONObject.Create4(HttpRqresponseText);
result:= jo.getjsonarray(‘data’);

//write out to check
for it:= 0 to result.length-1 do

writeln(‘EJson: ‘+ExceptiontoString(exceptiontype, exceptionparam));

And this JSON Array as above function returns, we pass to the next plot:

procedure JSON2Plot(form1: TForm; jar: TJSONArray);
var chart1: TChart; cnt: integer; sumup,tmp2,tmp: double; gday: string;

form1.onclose:= @Form_CloseClick;
chart1:= ChartInjector(form1);
sumup:=0; tmp2:=0; tmp:=0;

for cnt:= 0 to jar.length-1 do begin
tmp:= jar.getjsonobject(jar.length-1-cnt).getdouble(‘injection’);
tmp2:= jar.getjsonobject(jar.length-1-cnt).getdouble(‘full’);
sumup:= sumup+tmp;
gday:= jar.getjsonobject(jar.length-1-cnt).getstring(‘gasDayStart’);
chart1.Series[2].Addxy(cnt,jar.getjsonobject(jar.length- 1-cnt).getdouble(‘withdrawal’),’’,clyellow);

writeln(‘EPlot: ‘+ExceptiontoString(exceptiontype, exceptionparam));

PrintF(‘Landrange %d: Injection sum: %.2f’,[jar.length-1,sumup]);

As we can see we have 4 series to plot (including timeline):

  1. Injection (Injection during gas day)
  2. Full (Storage / WGV (in%))
  3. Withdrawal (Withdrawal during gas day (2 digits accuracy)).
  4. GasDayStart (The start of the gas day reported)

The time series is a result of the gas day and a trend is available.

“Gas day” means the period from 5:00 to 5:00 UTC the following day for winter time and from 4:00 to 4:00 UTC the following day when daylight saving is applied. Gas day is to be interpreted as UTC+1 for CET or UTC+2 in summer time for CEST. (Definition: see CAM Network Code specific-cations).

API access is provided in a REST-like interface (Representational State Transfer) exposing database resources in a JSON format with content-type in the mentioned Response Header.


The code of the data science vision contains example usage, and runs under Python3, Delphi, Jupyter-Notebook, Lazarus and maXbox4. Note that the API service is made available to the public free of charge. ONLY the data as currently available on the platforms is made available.

Tip: To extract data direct from the system, you can click on one of these links in a browser (web traffic versus API traffic ;-)):



GIE is providing an API (Application Programming Interface) service on its AGSI and AGSI+ storage data. The API documentation is on progress and provides examples and guidance on how to use the service and is available after registration to get an API-key. Below zoom of the past half year:

Pic: 1154_agsi_plot15.png

The scripts and images can be found:


Doc and Tool:

Script Ref: 1154_energy_api_agsi_plot14.txt

Appendix: shows an WinAPIDownload class from maXbox4 integration

{* — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — *)

TWinApiDownload = class(TObject)


fEventWorkStart: TEventWorkStart;

fEventWork: TEventWork;

fEventWorkEnd: TEventWorkEnd;

fEventError: TEventError;

fURL: string;

fUserAgent: string;

fStop: Boolean;

fActive: Boolean;

fCachingEnabled: Boolean;

fProgressUpdateInterval: Cardinal;

function GetIsActive: Boolean;


constructor Create;

destructor Destroy; override;

function CheckURL(aURL: string): Integer;

function Download(Stream: TStream): Integer; overload;

function Download(var res: string): Integer; overload;

function ErrorCodeToMessageString(aErrorCode: Integer): string;

procedure Stop;

procedure Clear;

property UserAgent: string read fUserAgent write fUserAgent;

property URL: string read fURL write fURL;

property DownloadActive:Boolean read GetIsActive;

property CachingEnabled:Boolean read fCachingEnabled write fCachingEnabled;

property UpdateInterval:Cardinal read fProgressUpdateInterval

write fProgressUpdateInterval;

property OnWorkStart: TEventWorkStart read fEventWorkStart

write fEventWorkStart;

property OnWork: TEventWork read fEventWork write fEventWork;

property OnWorkEnd: TEventWorkEnd read fEventWorkEnd write fEventWorkEnd;

property OnError: TEventError read fEventError write fEventError;


by Max Kleiner 10/10/2022



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Max Kleiner

Max Kleiner's professional environment is in the areas of OOP, UML and coding - among other things as a trainer, developer and consultant.