Written by Alexandra Wall, Project Manager at nLine
In 2022, the five-year, $316 million Ghana Power Compact came to a close. nLine sensor data supported the evaluation of a line bifurcation project under the Compact. This blog is adapted from Millennium Challenge Corporation’s (MCC) final learnings from this impact evaluation and highlights lessons for the design of future power infrastructure programs. MCC’s key lessons learned include the importance of using newer grid measurement technologies in the design as well as the evaluation of power infrastructure projects, and the need to focus on customer-level voltage quality as a key power reliability outcome.
Ten years ago, when the Ghana Power Compact was being designed, real-time power quality data that was affordable and collected independently of the utility was not commonly available in African markets. In 2014, available data included conventional sources such as utility data on losses and outages (provided only at high or medium-voltage transmission levels) or targeted load flow analyses, as well as utility customer surveys and targeted interviews with key energy stakeholders.
For example, in Accra, Ghana, the highest spatial- and temporal-resolution measurements were from the utility’s SCADA system, which covers only high-voltage transmission lines and part of the medium-voltage distribution network. Measurements of low-voltage outages were coming from customers calling the utility to report outages, a source known to be fraught with inconsistencies (i.e., poorly captured customer experiences of outages). The designers of the Ghana Power Compact had to use these available data sources to inform which grid investment activities should be undertaken and in which parts of the grid network in Accra.
nLine recognized this lack of geographically and temporally precise data, and developed its GridWatch technology to measure customer-level power quality and reliability with a higher degree of granularity and precision.
Using GridWatch data at the design phase could have enabled the project to select the worst performing feeders. This is the type of thing that can be used at the early stage of project design to pinpoint areas. — Jeffrey Garnett, Associate Director, MCC Monitoring & Evaluation Division
The Ghana Power Compact aimed to transform the country’s energy sector by providing more reliable and affordable electricity to Ghanaians and ensuring a sustainable, reliable grid for power distribution. One way the Compact strived to achieve this was through line bifurcation investments on the low-voltage distribution network in Accra.
For the impact evaluation of the line bifurcation investments, MCC hired UC Berkeley, which used nLine’s GridWatch technology to collect high-resolution power quality data for its evaluation. Approximately two years after the formal close of the Ghana Power Compact, MCC shared final learnings from this impact evaluation. Below are two of MCC’s lessons learned for the design of future power infrastructure programs (please read MCC’s final learnings from the impact evaluation for a full, detailed summary of learnings).
Lesson: Harness newer grid measurement technologies in the design, screening, and prioritization of power infrastructure projects
The high-resolution data from the GridWatch platform yielded extensive insights into the duration, frequency, and location of outages on the distribution network, while highlighting recurring patterns over time and across low, medium, and high-voltage segments of the local grid. Such granular data on grid performance was not available to MCC during the feasibility and design stages of the Ghana Power Compact.
With the advent of remote sensing platforms that can capture longitudinal grid performance data with relative ease, the design of future power infrastructure investments can be far more robust, leveraging enhanced precision in the analysis of root causes of outages, and even helping to pinpoint the weakest performing portions of the grid down to the feeder level. For instance, analysis from GridWatch data showed that had the worst-performing lines been identified and selected for transformer injections in Accra (instead of the comparatively average-performing lines actually targeted), observed impacts on voltage could have been up to 20 percent higher, leading to 14 fewer hours per month of undervoltage experienced by customers.
Going forward, MCC country teams should identify and incorporate newer technology platforms into their due diligence planning. Such approaches should be enlisted particularly for planned investments in the distribution network of partner countries, with plans for deployment and data collection included in early stages of program development. If installed very early in the process, such analyses can help prioritize investments with greater accuracy by targeting the areas with highest needs. This can also assist in establishing baseline performance, while improving the rigor of target-setting and overall economic appraisal of program investments.
Lesson: Further explore the role of voltage quality in mediating consumer outcomes
Within the Sustainable Development Goal 7 community, relatively little focus has historically been placed on customer-level voltage quality, even though voltage fluctuations can significantly impact firm productivity and be just as disruptive as power outages—if not more—for utility customers (What’s Reliability Without Voltage?, 2022). Since day one, it has been nLine’s ethos to leverage our technology and data to answer as many critical energy access and energy justice-related questions as we can. Due to the suspected impact and pervasiveness of voltage issues in Accra, nLine was driven to include customer-level voltage quality as a measurement outcome of the Ghana Power Compact’s line bifurcation project. With MCC’s support and investment in innovative research, we were able to enable voltage measurements with GridWatch sensors.
Based on the evaluation’s findings, MCC country teams should continue exploring the impact of poor voltage quality on consumers. In addition, future analyses should investigate using alternative metrics for voltage quality, considering the ways in which unstable voltage may impact electricity users, as well as working to better characterize the coping strategies adopted by consumers.
Measuring Impact with Innovative Data Sources
To improve reliability, it is important to measure it [IEEE and SEforALL]. To understand how well infrastructure investments improve reliability, it is important to have baseline measurements [MCC]. We are now in an era where utility data or customer surveys are no longer the only options for understanding customer-level power quality and reliability in the Global South. While nLine’s data was used for monitoring and evaluating outputs of the MCC Ghana Power Compact, future Power Compacts can benefit from deploying data collection systems during the design stage that can gather independent, high-resolution, real-time data to identify the key constraints to the provision of reliable power for customers and to target areas for improvement (e.g., where to add transformers or upgrade lines). This independent audit of distribution network quality and reliability can help support design, as well as monitoring and evaluation efforts, and ultimately measure if investments in energy infrastructure are resulting in improved power quality and reliability.
In more innovative efforts, high-resolution voltage data, when paired with consumption data, could be used to increase the accuracy of grid topology and technical loss estimates. And more speculatively, we are doing research on using voltage data to map the structure of the grid in regions where high-resolution network maps are incomplete and to automatically identify the root-cause of power outages when they occur.
Where It All Began: The story of nLine’s origins with the Ghana Power Compact
Eight years ago, nLine was not yet nLine. The core mission and ideas behind nLine lived through our two co-founders’ (Noah Klugman and Joshua Adkins) PhD research at UC Berkeley under their academic advisor (Prabal Dutta). Their research explored novel techniques for measuring power outages. In 2016, the Development Impact Lab — a UC Berkeley-led consortium that scales development technologies and incorporates measurement technologies into social science impact evaluations and institutional monitoring and evaluation (M&E) — provided support for the piloting and testing of this real-time, power outage data collection system in Accra, Ghana.
.jpg%3Ftable%3Dblock%26id%3Dd021a2cf-5ea2-4ac7-ba05-2fe69fdcea55%26cache%3Dv2&w=3840&q=75)
That same year, a team of economists at UC Berkeley were awarded independent evaluation of a project under the five-year, $316 million MCC Ghana Power Compact. Noah and Josh leveraged GridWatch — the data collection system they first designed — to provide scale measurements of grid reliability to enable a rigorous estimation of the effectiveness of the compact investments in reducing power outages and improving the reliability of the power distribution network.
Fast forward to today when the five-year Ghana Power Compact has come to a close, and nLine has transitioned from a university-based research project to a social enterprise delivering power quality and reliability measurements for 10 clients across 10 countries.
MCC was nLine’s first client and the Ghana Power Compact was nLine’s first project; this experience not only honed our technology systems and sensor deployment strategies but, more importantly, created a space for nLine and MCC to push the boundaries on using novel datasets to inform Compact design, monitoring, and evaluation processes. To that end, we are currently supporting the evaluation of a medium-voltage network upgrade project under the Senegal Power Compact.
.jpg%3Ftable%3Dblock%26id%3D00430644-1149-453f-92b7-8a31d7482732%26cache%3Dv2&w=3840&q=75)
.jpg%3Ftable%3Dblock%26id%3De22c1c7b-cb22-4605-82d6-47c127a4007d%26cache%3Dv2&w=1080&q=75)
Interested in learning how nLine can support your data needs? Please reach out, we would love to hear from you at https://nline.io/contact or info@nline.io.