Data-driven solutions to renewable energy project failure

Why do some renewable energy projects fail?

If renewable energy is going to overtake fossil fuels we need to get a handle on the main reasons for project failure and start finding solutions fast. Lots of studies point to barriers that cause renewable energy projects to fail. These tend to fall into four main categories: technical, economic, political and social.

While these can differ according to country and location, this report focuses on the specific technical barriers that are common to most countries. They are: grid connection constraints (e.g. capacity, transparency), lack of grid data availability and an absence of data-driven risk management.

Grid connection constraints

Grid connection constraints and grid capacity are the main barriers to renewable energy deployment in countries with abundant solar and wind. The lack of predictability for planning permits, long delays in permit processing and a lack of clarity in the pricing structure and cost-sharing for grid owners and developers all hinder efficient project development.

Difference between centralised centralised and decentralised power supply Source: Gunter Fraunhofer IEE

To make things even more complicated, transmission lines in different countries were designed for a centralised system. This creates some challenges for distributed renewable energy generation at the point of common coupling (PCC), which can cause the reactive power balance of the grid to be disturbed. Reactive power control solves the issue, but the variability and partial predictability of renewable energy resources creates another layer of uncertainty that should be factored into the grid operation. Without the right type of data, connection becomes complicated.

This creates limits to the amount of energy fed into the grid from renewable energy sources that development and investment teams may not be aware of until it’s too late! In the EU new policy measures are aimed at avoiding frequency unbalance, guaranteeing grid flexibility and responsiveness, to ensure the thermal limit of the grid components is not surpassed and to stay below the network rated factor, but this is a long term work-in-progress. Project and investment teams will need to get smarter in order to continue to build profitable income streams from renewable energy in Europe.

Another constraint is that each connection point to an existing grid is unique and requires detailed power system modelling to assess the impact of the renewable energy connection on the grid. Additionally, location-dependence means that additional transmission lines may be necessary, due to most renewable energy projects being remote from existing transmission networks. Getting access to reliable grid data is one solution that can certainly help with that.

Grid data opacity

With the lack of available grid data in most countries, project and investment teams at early-stage are left with more questions than answers. For new entrants, this means that some projects do not even get beyond the planning stage.

Both transmission system operators (TSOs) and distribution system operators (DSOs) are often reluctant to share grid data. The two main reasons for this are 1 ) unwillingness to provide potentially sensitive information that could compromise the safety of a power plant and, 2) reluctance to indirectly reveal revenues. Some services providers share partial grid data, but withhold data about geo-coordinates of grid elements, transmission cable location and characteristics, underground cables, electrical substations at different voltage levels, load data, renewable energy sources and sites.

The opacity of grid data in most countries means that project and investment teams have to do a lot of guesswork to estimate loads, electrical properties of power lines, locations of substations, etc. But this type of guesswork comes at a cost: the dataset may misrepresent the power grid. This reality is not new to most decision-makers. To work around the chaos, they embark on a time-consuming quest to collect data from various sources with the hope of finding some answers, only to be confronted with the disappointing reality that their models and deductions are painfully insufficient. The implication? They are unable to understand the risks at the PCC.

Implications of the lack of grid data leads to inaccurate capacity expansion modelling when considering adding other energy sources. It also becomes difficult to accurately model grid instabilities, assess frequency fluctuations, study grid strengthening measures, perform production cost modelling, assess integration studies, conduct congestion management, identity reliability constraints, and ultimately reduce the level of uncertainty related to renewable energy injection.

The best grid data needed to accurately study the effects of renewable energy on an existing grid will depend on the model used and the purpose of the data. Generally, data related to global frequencies, nominal voltage levels, reactance of power lines, demand, active power generation, net transfer capacity, electric parameters of power lines, grid topology, tap ratio and transformer information are essential parameters to represent the power grid accurately. Not easy!

Data-driven risk management

Despite national and international efforts to compile renewable energy data, there is still a lack of useful quality data to mitigate renewable energy risks. This is further compounded by both technical and resource constraints (e.g. financial, human capital) that make it challenging to collect and correlate renewable energy data in real-time while considering different scenarios in parallel to strategically optimise the decision-making process. As a result, the final project definition is almost certainly destined for failure, lacking the appropriate risk evaluation and control mechanisms to guarantee otherwise.

Even the best renewable energy database has to be frequently updated on static and dynamic maps, to accurately reflect changes to infrastructure, terrain, wind speeds, solar irradiance, landfills, forests, humidity, and much more. This data-driven approach gives stakeholders the tools they need to carefully study factors related to meteorology, access points, land use, transmission line distances, location advantages, and competition. The study is then used to strategically guide decision-making alternatives that aim to mitigate renewable energy risks associated with unpredictability, non-controllable variability and location-dependence.

Without data-driven risk management, stakeholders overestimate power supply, invest heavily in development, and could eventually watch operation and maintenance costs skyrocket. In worst-case scenarios, astronomical losses and meagre power generation are the final anathema of poor strategic risk execution and planning.

Data-driven solutions

In terms of solutions big data and renewable energy can be a match made in heaven. The union guarantees better and consistent grid predictability that can ultimately reduce dependence on non-renewable energy and stimulate more investments earlier on in the project lifecycle.  Energy Tech Review—a self-professed guardian angel connecting energy providers and businesses—noted the vote among researchers in favour of data as the most crucial of the three drivers of a modern, clean energy system: intelligently-managed hardware and software being the next two. They proclaim, “The wave of digitisation will also impact the renewable energy sector as it will propel the deployment of data analytics.”

The journey towards renewable energy upscaling and barrier removal definitely “starts with data”. Renewable energy resource data and geographic information system (GIS) underpin crucial decisions in target setting, policymaking, investment and power sector planning.

Decision-data-analysis-nexus (Cox, et al. 2018)

However, a cursory look at renewable energy data without any qualitative and quantitative analyses will result in GIGO (garbage in, garbage out). Improved decision-making comes from combining high-quality data with practical predictive analysis to drive the deployment of renewable energy in ways that foster resilience, innovation, upscaling, and streamlining. In other words, not just any data will do.

Predictive analysis transforms data from scattered patterns, repeated signals, temperature variations, intermittent wind speeds, weather forecasts, historical observations, and much more, into actionable insights that key stakeholders can use to guide their decisions. When it is coupled with high-quality data, stakeholders are no longer feeling their way in the dark. Instead, they are equipped with visualisation tools that enhance their risk management strategies, strengthen their investment options, lay the groundwork for ambitious policies guided by robust metrics, promote strategic renewable energy planning, and effectively weigh the long-term economic potential of a sustainable energy mix. This reduces risk aversion and uncertainty and increases predictability by allowing communities to see renewable energy as an extension of their existing environment and not an uncomfortable nuisance.

Stages of analytical analysis – Source: Butt & Meisinger

The obvious questions now are which barriers can be overcome and who are the players already addressing them? It turns out that more and more software companies are already easing these pains and creating value adding options ranging from better pipeline management to blockchain-based and IoT solutions created specifically for renewable energy customers.

Here is a brief look at four data-driven solutions that help overcome development barriers: ENIAN, Aucerna, Energy Web and Odyssey Energy Solutions.

We’ve already seen two barriers that hinder renewable energy deployment: lack of grid-level data and data-driven risk management. ENIAN’s PowerGrid API provides granular grid and power plant data for 222 countries. Their API use case strengthens capacity expansion and product cost modelling, making the process of risk identification and project execution more data-driven and efficient. Instant information about connectivity and capacity provides valuable insights for grid modelling to determine the effects of renewable integration on an existing grid or grid expansion with more high voltage lines and substations. With access to geospatial data, it’s possible to identify profitable siting options by assessing land use in specific regions and study constraints to lower risk.

The profitability of a renewable energy project has a rippling effect on getting financing, remaining competitive, and having a long-term view from the earliest stages in the project lifecycle. A better understanding of profitability can be gained through efficient cash flow modelling, based on data-driven power output, IRR and LCOE modelling rather than back of the envelope calculations. ENIAN offers this as an all-in-one intelligent pipeline management solution that helps developers, investors and advisors move faster and more efficiently in the early project stages.

ERP is an integral part of renewable asset management, operation and maintenance, and profitability for operational phase projects. Aucerna connects project teams across multiple platforms to manage assets efficiently. The result is that processes are improved so that activities and decisions that do not generate profits, productivity and security are removed from the pipeline. By having a bird’s eye view of a renewable energy project, risks are minimised, and essential decisions are made to mitigate any future risks.

Secrecy about grid data complicates modelling and hampers decision-making. The Energy Web Decentralized Operating System from Energy Web uses data from distributed energy resources to improve how this data is provided and shared. With blockchain, data across various platforms inform decisions by connecting customers, physical assets and existing grid infrastructure with digital applications. Data about the operational capabilities of grid resources makes grid assessment more transparent by removing layers of uncertainty about how the integration of renewable energy affects the grid.

The output of wind farms and solar farms cannot always be on “operation mode”. In other words, being weather-dependent resources means they are intermittent and to some extent unpredictable.  What’s more, is that things get complicated when the effects of terrain, topography, infrastructure and temporal data sets are factored into the equation. But on the other hand, wind farms and solar farms are constantly generating considerable amounts of data, that when communicated back to the grid in the form of analog or digital information, improve its operation, efficiency, reliability and output. This is possible with the Internet of Things (IoT) where data is integrated from sensors, control devices and energy technologies to control how the grid behaves. To put it simply, the network takes care of itself. This smart grid concept is one way Odyssey Energy Solutions uses data to develop detailed analyses of distributed energy portfolio performance across numerous data sources.

In a Nutshell

Software is solving the pains that many renewable enegy teams experience by providing high-quality data coupled with digitised analysis and modelling methods to transform terabytes of information into focussed decision-making tools. This helps decision-makers to make informed, better decisions, accelerating project planning and execution, improving risk management and reducing costs throughout the value chain.

Ultimately data creates myriad new and exciting possibilities that can lead to significant breakthroughs in the fight against climate change. Data availability (or lack thereof) can cause ripple effects in all aspects of a the energy-economy, helping to overcome development barriers and accelerate the global transition to a more sustainable system for all.