Factors Impacting Energy Consumption
For the data analysis project, we are considering the data set collected under the Solar cities project lead by the university of Ballarat. This dataset consists of different factors that may impact on the pattern of energy consumption by the different consumers depending on the various factors. This factors includes estimated age of the users, wall_construction type of the houses, roof colour, number of bedrooms, bathrooms, living rooms, size_sqm, window_type, window_coverings.
In addition to that, factors cfl_count, halogen_count, led_count, incandescent_count, fluor_count, insulation, pv_capacity, interval_date, power_usage is also important in determining the different kind of energy consumption changes in the Loddon Mallee and Grampians region.
Through analysis of the dataset the drivers of the energy consumption in the above mentioned areas depending on various factors.
After importing the data set in the IBM Watson analytics tool at first we tried to find out the results for the following questions which includes the following graphs from the analysis.
contribution of power usage over a year by roof colour
After the analysis of the data in the Watson we found that minimum of power consumption in the given area is in the specified area is in the year 2012 which gradually increased in the year 2013 and 2014. In the year 2015, the power consumption reduced compared with the year 2014. Here it can be observed that the, the roofs with the dark and intermediate colours have major share in the energy consumption.
Contribution of power usage over a year by PV capacity
In our analysis of the data set we found that, the power consumption by the house holds and the business are high where the PV capacity is equal to zero.
Contribution of Power usage over a year by PV_Capacity and Insulation
From the data set provided in when it is investigated to find the contribution of power usage by the PV capacity and the Insulation it is found that the, the following combinations for the households and the business are responsible for too much energy consumption. This is depicted in the following graph,
From the above graph it is easily visible that, with (0,0), (0,1), (0,2) has the highest power consumption throughout the years 2012, 2013, 2014,2015. In addition to that, it is also visible that, there is increment in the power consumption by the households and the business organizations which have PV_CAPACITY 1500 and insulin is 2.
Results from Data Analysis
Power usage by estimated age
Following is the breakdown of the power usage by different estimated age groups given in the data set. From the following heat map, it can be easily observed that houses over the age sixty and above are using most of the power in the given geographic area
Months with most energy consumption
Analysing the given data set in Watson we found the following graph that depicts the power usage in the different months in the year 2012, 2013, 2014, 2015. From the analysis of the graph we found that in the year 2012, most energy is consumed in the month August which summed up to 28802.64 units. Similarly for, the years 2013, 2014, 2015 we get the most energy consuming months are July-2013(35669.25 units) , July-2014 (47195.75 units) and lastly July-2015(48181.94 units)
Months in which least power used
By sorting the energy consumption values in ascending order we get the following graph which depicts the minimum power consumption in the given years. Here in the analysis we get that in the year 2012, the least energy consuming moth is February which consumed only 5595.71 units. In the year 2013, February again is the least power consuming month. For the years 2014, 2015 the least power consuming months are given by
Drivers of Power usage
For this part we got that following are the drivers for the power usage (here the factors that have strength more than 25% are considered),
WALL_CONSTRUCTION and SUBURB
SIZE_SQM and SUBURB
BEDROOMS and SUBURB
INCANDESCENT_COUNT and SUBURB
FLUOR_COUNT and SUBURB
ESTIMATED_AGE and SUBURB
FLUOR_COUNT and ESTIMATED_AGE
CFL_COUNT and WALL_CONSTRUCTION
CFL_COUNT and ESTIMATED_AGE.
suburbs have the most houses with pv_capacity
In our analysis we found that, among the 8 suburbs in the given data set we found that, the Heywood suburb has the most number of houses with PV_CAPACITY. Second is Portland Suburb and the last is Myamyn suburb.
Houses in suburbs more likely to have pv_capacity
In our analysis we have the following result which depicts the number of houses of different ages having the PV_CAPACITY,
Here we can say that the houses between the ages twenty to twenty-nine are more likely to have PV_CAPACITY.
Power usage by owned and rented Houses in the Given dataset
From the analysis we get the following graph that depicts the power usage by the different houses which are either rented, owned, mortgaged, rent free and other. From the following chart it is evident that the owned houses use more power than the rented houses. Being specific the owned houses uses total 960929.98 units and the rented houses uses 208406.2 units.
Drivers of Power Usage
suburb dwellings that uses most power
From the analysis we found that, the dwellings in the Portland uses most power in the given data set. The result is depicted in the following graph.
Power usage by the size of the house
For the given data set the usage by the different houses and the businesses are also investigated. In this investigation it is found that, the smaller houses (under 199 SQM) are consuming more power compared to the houses with the larger square meterage. Numerically the total consumption by the, small houses are summed up to 650889.28 units and for the larger houses this total is 13400.4 units (for houses with size more than 300 SQM).
With the changes in stories of the building the trend does not change as shown in the following bubble chart in this scenario too the total consumption of energy is higher for small houses.
Wall construction type with the age of the houses
In order to explore the type of construction of the wall according to the age of the houses are depicted below with chart,
What age houses and from which areas and with how many bedrooms use the most power
Here it is found that the houses with the estimated age Fifty to Fifty-Nine in the port land area with 3 bedrooms consumes too much power. Numerically the total power usage is summed up to 119096.38 units
Relation between power usage and roof colour and roof material
In the given data set there is no attribute roof material thus for this scenario we are considering the wall construction material as the roof material. After our analysis with the Watson analytics we get that, the roofs with dark colour and made with BRICK are consuming most amount of power in that region (340809.88 units). Second is the combination of roof colour intermediate and built with brick (210973.59 units).
Utilization of power by the houses that have double glazed windows and window coverings
From the analysis of the given data set it is found that the houses with curtains and single glaze windows uses more power compared to the double glaze windows. The green bubble in the chart is showing the total usage by the houses with single glaze and curtains whereas the double glazed windows with coverings are represented by blue bubble in the chart.
For this data analytics project we have used this tool as IBM Watson is helpful in processing of the unstructured data. with its cognitive computing it can also help in and act as a decision support system for the decision makers in any business. In addition to that it helps in improving data analytics performance while providing best decision available from the data set (Chen,Argentinis & Weber, 2016). Moreover, this tool can handle enormous quantities of data. it provides different starting points that helps in building new queries depending on deferent variables in the selected data set.
Efficiency and Power Consumption
From the above analysis it is found that the main drivers of power usage in the given areas are mainly
WALL_CONSTRUCTION and SUBURB
INCANDESCENT_COUNT and SUBURB
FLUOR_COUNT and SUBURB
ESTIMATED_AGE and SUBURB
Therefore, in order increase the efficiency in the power consumption it will be important to improve the wall construction material and the colours of the roofs of the houses. Moreover, as the dwellings with the double glazed windows uses lesser amount of power thus the use of the Double glazed windows with coverings should be encouraged to improve the efficiency in the given areas.
While working on the IBM Watson I found it a little complex to Interpret the charts when there are too many distinct values in the data set. As the tool marks the values with numeric one. On the other hand, the use of this tool in order to compare different values through chart makes it easy to analyse any data set.
References
Aggarwal, M., & Madhukar, M. (2017). IBM’s Watson Analytics for Health Care: A Miracle Made True. In Cloud Computing Systems and Applications in Healthcare (pp. 117-134). IGI Global.
Chen, Y., Argentinis, J. E., & Weber, G. (2016). IBM Watson: how cognitive computing can be applied to big data challenges in life sciences research. Clinical therapeutics, 38(4), 688-701.
Derico, A., Leader, A. C., Kather, R., Engineer, O. S. A., & West, D. (2017). Make Data Simple: IBM Watson Data Platform & Data Science Experience.
Devarakonda, M., & Tsou, C. H. (2015, January). Automated Problem List Generation from Electronic Medical Records in IBM Watson. In AAAI (pp. 3942-3947).
Diamond, M., & Mattia, A. (2017). Data Visualization: An Exploratory Study into the Software Tools Used by Businesses. Journal of Instructional Pedagogies, 18.
Gao, T., Dontcheva, M., Adar, E., Liu, Z., & Karahalios, K. G. (2015, November). Datatone: Managing ambiguity in natural language interfaces for data visualization. In Proceedings of the 28th Annual ACM Symposium on User Interface Software & Technology (pp. 489-500). ACM.
High, R. (2012). The era of cognitive systems: An inside look at IBM Watson and how it works. IBM Corporation, Redbooks.
Lak, P., Kavaklioglu, C., Sadat, M., Petitclerc, M., Wills, G., Miranskyy, A., & Bener, A. B. (2017, November). A probabilistic approach for modelling user preferences in recommender systems: a case study on IBM watson analytics. In Proceedings of the 27th Annual International Conference on Computer Science and Software Engineering (pp. 38-47). IBM Corp..
Mylopoulos, J. (2017, November). Goal-Oriented Regulatory Intelligence: How Can Watson Analytics Help?. In Conceptual Modeling: 36th International Conference, ER 2017, Valencia, Spain, November 6–9, 2017, Proceedings (Vol. 10650, p. 77). Springer.
Tsoi, K. K., Chan, F. C., Hirai, H. W., Keung, G. K., Kuo, Y. H., Tai, S., & Meng, H. M. (2018). Data Visualization with IBM Watson Analytics for Global Cancer Trends Comparison from World Health Organization. International Journal of Healthcare Information Systems and Informatics (IJHISI), 13(1), 45-54.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
