RAZVOJ APLIKACIJE ZA PODPORO ANALITIKI PRI PREDVIDEVANJU PORABE ELEKTRIČNE ENERGIJE

Kikelj, Andrej

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	RAZVOJ APLIKACIJE ZA PODPORO ANALITIKI PRI PREDVIDEVANJU PORABE ELEKTRIČNE ENERGIJE
Authors:	ID Kikelj, Andrej (Author) ID Kofjač, Davorin (Mentor) More about this mentor... ID Brezavšček, Alenka (Comentor)
Files:	UNI_Kikelj_Andrej_2014.pdf (1008,59 KB) MD5: 4AA307C48AE805C334DD11C142EE37CC
Language:	Slovenian
Work type:	Undergraduate thesis
Organization:	FOV - Faculty of Organizational Sciences in Kranj
Abstract:	Zaradi zagotavljanja konkurenčnosti na trgu morajo podjetja vse več investirati v razvoj programske opreme, ki omogoča hitrejše, natančnejše in učinkovitejše poslovanje. V podjetju Elektro gorenjska prodaja d. o. o. se je pojavila potreba po optimizaciji procesa analitike predvidevanja porabe električne energije pri odjemalcih z več odjemnimi mesti ter skupinah odjemalcev. Diplomsko delo zajema analizo problematike pri delu napovedovanja porabe, oblikovanje koncepta programske rešitve, oblikovanje podatkovne baze za programsko rešitev ter sam razvoj programske rešitve. Problem predstavljajo procesi uvoza in obdelave podatkov, kjer imajo analitiki preveč rutinskega ročnega dela. Ključno delo analitika je apliciranje izkušenj s področja predvidevanja porabe glede na gospodarske in druge dejavnike, zato mu mora programska rešitev kar se da avtomatizirati rutinski proces uvoza in prečiščevanja podatkov. Problematični so različni formati podatkov, ki se razlikujejo tako po zapisu podatkov kot časovnih pasovih. Prepoznavanje in odstranjevanje nekarakterističnih podatkov je brez vizualnega prikaza dnevne porabe oteženo. Kot rešitev smo oblikovali koncept za aplikacijo, ki analitiku omogoča hitrejše delo s podatki in zmanjša možnost človeških napak. Aplikacija avtomatizira rutinske procese uvoza in pretvorbe podatkov, pretvorbo časovnih pasov ter združevanje podatkov glede na odjemalce in skupine odjemalcev. V nastavitvah lahko analitik nastavi dela proste dni ter praznike, ki se ob izračunu povprečja porabe ne upoštevajo. Po uvozu podatkov aplikacija omogoča prikaz le-teh v grafični obliki, kjer ima analitik ob vizualnem prikazu po urah v dnevu možnost izključevanja podatkov, ki bistveno odstopajo od povprečja, s klikom na krivuljo podatka. Aplikacija omogoča analitiku filtriranje podatkov po specifičnih zahtevah, tako da ima možnost prikaza npr. podatkov za vse sobote in nedelje v januarju in februarju za leti 2012 ter 2013. Glede na podatke, ki jih analitik ni odstranil, se izpiše povprečje in predvidena poraba, interpolirana na število dni v podatkih, s pomočjo katere lahko analitik hitreje izračuna predvideno letno ali mesečno porabo.
Keywords:	razvoj programske rešitve, razvoj baze, obdelava in prikaz podatkov
Place of publishing:	Maribor
Year of publishing:	2014
PID:	20.500.12556/DKUM-43369
COBISS.SI-ID:	7281427
NUK URN:	URN:SI:UM:DK:LX5O57M6
Publication date in DKUM:	10.04.2014
Views:	1470
Downloads:	154
Metadata:
Categories:	FOV
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Secondary language

Language:	English
Title:	DEVELOPMENT OF APPLICATION TO SUPPORT ANALYTICS FOR ELECTRICITY CONSUMPTION PREDICTION
Abstract:	To maintain market competitiveness, companies must invest more and more resources into development of software that allows faster, more precise and more efficient workflow. The company Elektro Gorenjska Prodaja d.o.o. had found that there is a need to optimize the analytical process of consumption of electrical power for groups of consumers and consumers with multiple supply points. The thesis consists of analysis of issues connected to power consumption prediction, designing a concept for a software solution, database design for the solution and the development of the solution itself. The main issue is the process of import and data processing, where analysts are faced with more manual processing than desirable. Analysts create most value added when applying their experience from the field of power consumption analysis in regards to current economical and other factors that affect power consumption, thus the goal of the application is to automate the routine work of import and data processing as much as possible. The main issues here are data formats, which vary in data structure and data time zones. Without a visual representation of the data, identification of the uncharacteristic data is difficult. As a solution to the aforementioned issues, we've designed a concept for an application, which will increase the analyst data processing productivity and reduce the chance of human error. The application will automate the routine import and data conversion processes, time zone detection and conversion and data aggregation according to consumers and consumer groups. The application settings allow the analyst to set holidays and work days, so that the application will exclude data for work free days. After the data is imported, it can be shown on an hourly graph, where the analyst can exclude data with greater than desired deviation of the average values, with a simple mouse click on the data curve that needs to be excluded. The application allows the analyst to filter data by specific criteria, so that the analyst can, for example, only show data for all the Saturdays and Sundays in January for years 2012 and 2013. According to data that the analyst kept in the calculation, the application shows the average and predicted consumption, interpolated for the total number of days in the data, which enables the analyst to calculate predicted consumption on a yearly or monthly basis.
Keywords:	application development, database development, data processing and display

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back