Utilidex is a software company founded in 2012, primarily catering to the energy industry. The company delivers a range of services to support end users, consultants, energy traders, and suppliers with energy data management, billing, trading, and budgeting offerings, all of which are cloud-based under their integrated suite supported by the Microsoft Azure platform.
Alongside expansions into new markets in Europe and Australia, Utilidex is also expanding its portfolio with other offerings.
What is CHP?
CHP (Combined Heat and Power) is an energy-efficient technology used to generate electricity while also utilizing the heat generated in the process that would otherwise be wasted. They can be located at large end users for localized generation and can provide heating, hot water, and electricity to a district network.
Conventionally, nearly two-thirds of heat used to generate electricity is vented into the atmosphere. By capturing this heat and avoiding transmission losses, CHPs can achieve efficiencies of over 80% compared to the combined 50% for procuring from the grid and generating heat from an on-site boiler.
Types of CHPs
The common CHPs however are reciprocating engines, combustion turbines, steam turbines, and microturbines
CHPs can be of various types based on the technology they use. The common ones however are reciprocating engines, combustion turbines, steam turbines, and microturbines.
They can also use different fuels, but the most prominent fuel is natural gas. The principles of heat and electricity generation remain the same overall. A CHP system can be heart-led or electrically led based on the priorities of the user.
The Project
For Utilidex, a key area of focus has been the CHP (Combined Heat and Power) space. Hence, they partnered up with University College London’s (UCL) School of Management for exploring this area further by offering an internship to Kunal Kalani, a student in the MSc Business Analytics program at UCL.
The project was aimed at creating an optimization model for planning CHP operations to maximize profitability by intelligently exporting surplus energy based on the gas and electricity wholesale market prices in the UK.
Approach/Analysis
The project goal was to come up with a solution for agile decision-making for the operations of CHP units to run the plant as well as import and export electricity at the appropriate times, based on the gas and electricity markets, which would minimize the cost of energy for the CHP owner/operator.
Zero-Carbon Footprint Goal
In-line with the zero-carbon footprint goal set out by the NHS for 2045, a lot of hospitals and trusts have invested in CHPs
To model an optimizer and test it, data was used for 3 hospitals in the UK. In-line with the zero-carbon footprint goal set out by the NHS for 2045, a lot of hospitals and trusts have invested in CHPs for greener energy.
This also allows selling excess power produced when needed to offset the costs further. Due to the availability of CHPs in this sector and data was used from hospitals. However, the project findings are relevant to other industries and sectors as well.
Heat Prediction Aspect
One of the three hospitals had no CHP while the other two were CHP-equipped sites. Hourly data gathered was for gas consumption and electricity consumption for 8 months for each of these hospitals. Based on the weather data for these locations, estimations were made for the heat demand of these hospitals.
This demand is critical to creating the heat prediction aspect of the model using a degree-day analysis. Similarly, average electricity usage per hour was calculated and used to estimate the electricity consumption for the next day.
Size Of The Chips
These estimations were required so that the optimizer had an idea of how much hourly heat and electricity is needed for the next day to make the decisions based on these numbers and the prices on the market for the following day.
Another assumption was made about the size of the CHPs for these sites. As the benefit of the optimization is in the export of excess electricity generated on-site, it was assumed that all 3 sites had 15% spare capacity left from the CHP at peak electricity load for the site.
Mixed Integer Linear Programming
A MILP was used to model various constraints which then allowed the model to arrive at the lowest costs
A Mixed Integer Linear Programming (MILP) was used to model various constraints which then allowed the model to arrive at the lowest costs for each day.
The simulations were run for periods of an entire month for each of the hospitals to cover variations in the market prices, seasons, and the hospital’s patterns of usage as well.
The Findings / Results
Based on the results from the optimizer, the following comparisons could be made for the three hospitals:
Cost savings
Over 1 month each, all three simulations showed savings over the actual costs of energy by the site. The majority of savings were generated through running the CHP intelligently for exporting electricity for sale.
This initial proof of concept showed monthly savings between 7.8% and 21.2% of the original energy cost to the site, based on the month of the year and the site itself. To further analyze the performance of the optimization modeling the following charts can be examined:
Optimized CHP operation
A basic example of the decision-making can be seen below where the CHP is run during price spikes and not run when the prices are low:
Hospital 1 – Optimized CHP operation based on Spot Electricity Price
In the chart above, electricity import and export patterns can be seen as decided by the optimizer. The optimizer chooses to export electricity by running the CHP when the prices are high to reduce costs and rather turn in a profit from selling the generated electricity.
On the other hand, it chooses to not run the CHP and import electricity when the prices are very low. Also, the optimizer can choose to import and export simultaneously to drive costs down further if the site can do so. This can be seen between January 11th and 12th on the chart.
Further Work / Looking Beyond
The proof of concept created is just the beginning of what can be achieved in this space by Utilidex. The optimization program will be continued to be refined and developed to suit several different clients and applications in various areas of the energy sector.
The model can handle specific constraints of the client's CHP setup and contractual obligations. There will be the future capability of optimizing multiple CHP units in conjunction as well as optimizing CHP units in tandem with renewable sources like Solar and Wind power. Looking beyond CHPs, the model is also capable of handling battery charging and discharging, and much more.