“StableFlame combines economic success with environmental protection. The technology is easy to integrate, saves energy, and thus reduces costs. This allows us”
StableFlame revolutionizes waste incineration with advanced AI technology.
We offer innovative solutions for efficient and sustainable waste treatment.
In a typical waste incineration plant, many different types of waste are delivered: sewage sludge, household waste, bulky waste, industrial waste, etc.
To stabilize the combustion, additional fuels such as natural gas or heating oil must be used.
Due to frequent temperature and pressure changes resulting from unstable combustion, critical plant systems experience premature fatigue.
The energetic potential of the waste in the waste bunker is unknown, therefore it cannot be used efficiently.
The content of the waste bunker is visualized in terms of its calorific value by a digital twin.
From the moment of unloading at the gate to the point of incineration, the waste’s steps through the waste treatment process is digitally documented for subsequent analysis and traceability.
By using the information of the calorific values of the waste provided by the digital twin, an intelligent algorithm homogenizes the waste and optimizes the filling of the hopper for stabilization of incineration.
On the left, you will see a typical view that a crane operator has of the waste bunker. From this viewpoint, the crane operator is expected to identify the calorific value of the waste at various locations within the bunker and then homogenize it. Several side conditions must be taken into account, such as the need to let the waste dry for several days before incineration. Without proper records, fulfilling this task is nearly impossible. Moreover, the detection of calorific value is fraught with significant inaccuracies and varies greatly from one crane operator to another, depending on their experience. Typically, the energetic potential lying dormant within the waste bunker is not utilized, which increases operating costs, reduces profit, and leads to greater CO2-emissions.
On the right, you see the self-learning digital twin with operative data from our pilot plant at MVV Umwelt GmbH in Mannheim. The rectangles, called “bunker cells”, represent quantities of waste at specific coordinates in the waste bunker. The color of the bunker cell represents the calorific value of the waste. Every action in the waste bunker, such as trucks unloading new waste at the gates or the crane transporting waste from one bunker cell to another or to the hopper, is recorded. A history of past actions is also available for reasons of traceability and documentation. The information from the digital twin is provided to an optimization algorithm, which calculates optimal strategies for using the entire energetic potential of the waste bunker, taking into account the technical side conditions of each plant, such as the necessary feeding rate of the hopper depending on the current operating condition. By connecting to the plant's operating system, the digital twin becomes more accurate over time, and even unknown wastes can be learned independently. StableFlame offers you a self-learning system that can autonomously adapt to different plants and types of waste.
MVV Umwelt GmbH has set the ambitious goal of becoming climate positive by 2035 – five years ahead of the original schedule. By continuing its climate protection initiatives and employing technological innovations, MVV significantly contributes to climate protection. In the future, MVV will be capable of permanently removing CO2 from the atmosphere. This will enable the company not only to offset its unavoidable residual emissions but also to make an active contribution to climate protection.
The StableFlame process is distinguished by several positive effects: It allows for a reduction in auxiliary firing by using intelligently homogenized waste close to the optimal calorific value for combustion in the hopper feeder. This eliminates the need to burn additional fuel, which would be required if waste of too low a calorific value were used. This leads directly to a reduction in CO2 emissions and the operating costs of the thermal waste combustion plant. At the same time, the risk of waste with too high a calorific value entering the combustion chamber is minimized. A consistent and constant feeding of the hopper ensures a higher waste throughput, thereby improving steam throughput, increasing the plants efficiency.
Furthermore, feeding the hopper with waste close to the optimal calorific value significantly reduces wear and tear on the plant. This results in longer maintenance intervals, thereby providing additional operational days per year when the plant does not have to shut down but can continue to produce electricity. This additional electricity production allows for generating extra revenue for the operators.
*The estimates presented are based on the experiences and findings from the pilot plant. Depending on the specific characteristics of the respective plants, these values may vary.
StableFlame, a spin-off from EDI – Engineering Data Intelligence, marks a turning point in thermal waste management. Our co-founder and CEO, Katrin Schütz, is distinguished by her extensive entrepreneurial expertise and leadership experience. Axel Sikora, our co-founder and Head of the Advisory Board, contributes his vast skills in entrepreneurship and innovation. Together, they complement the technological expertise of EDI, particularly in the field of Artificial Intelligence (AI). Together, the team unites its strengths to revolutionize the practice of thermal waste management with StableFlame. By combining professional know-how with practical experience, StableFlame aims to establish efficient and sustainable methods for energy recovery from waste.
“StableFlame combines economic success with environmental protection. The technology is easy to integrate, saves energy, and thus reduces costs. This allows us”
"The technology of StableFlame makes a significant contribution to environmental protection and simultaneously achieves massive scale effects through its self-learning capabilities. As a passionate entrepreneur, I am convinced that with this innovation, we can not only significantly reduce the CO2 footprint of waste management but also greatly increase cost efficiency."
"In 2015, we founded EDI - Engineering Data Intelligence, with the aim of sustainably transforming traditional sectors through the application of artificial intelligence. During a three-year research project, we developed the StableFlame process, which we are now scaling through our spin-off, StableFlame. We are proud to offer our customers a technology that makes a significant contribution to achieving their climate goals."
-p-500.jpg)
StableFlame offers its customers a straightforward pricing model with two options. The first option includes the self-learning digital twin of the waste bunker, complete with a user interface for three-dimensional visualization of the bunker contents. Thus, crane operators receive information about the calorific value within the waste bunker and can decide how to carry out homogenization and with what to feed the hopper.
The second option encompasses the self-learning digital twin of the waste bunker, augmented with an algorithm that optimally plans the crane's actions, thereby ensuring optimal homogenization and feeding of the hopper. This solution takes into account the specific conditions of each plant and aims for full automation without the need for crane operators. This pricing model is designed to cater to different automation and control preferences within waste management facilities, offering our customers flexibility and efficiency. Our prices apply per boiler where the StableFlame process is implemented, regardless of the plant's size. Depending on the characteristics of the specific plants and their level of automation and IT infrastructure, adaptation work may be necessary. These one-time costs are calculated separately as part of a commissioning project.
The StableFlame process was developed as part of the groundbreaking project "AI Tool for Predictive Process Optimization and Control for Medium-Sized Process Plant Operators," supported by the Ministry of the Environment, Climate Protection, and the Energy Sector of Baden-Württemberg.
This project impressively demonstrates the effectiveness of our technologies under real-world conditions. Our pilot plant, equipped with the StableFlame process at Boiler 6 of the thermal waste treatment plant in Mannheim, is being tested and continuously optimized under real operating conditions. The StableFlame process exemplifies our ability to develop efficient, low-emission, and self-learning systems that autonomously adapt to different types of plants and waste materials.
StableFlame is an advanced technology revolutionizing the thermal waste-to-energy process through the use of Artificial Intelligence (AI). By employing self-learning AI, it creates a digital twin of the waste bunker, enabling transparent representation of the waste's calorific value. This digital twin supports operations in two modes: In semi-automatic mode, crane operators receive visualized information about the calorific values of the waste in the bunker through an intuitive user interface. In fully automatic mode, the entire process is conducted without human intervention, significantly improving waste homogenization and combustion efficiency. Existing thermal waste-to-energy plants can be retrofitted with StableFlame, and it should be considered in the planning of new facilities.
StableFlame utilizes AI that recognizes the calorific value of waste during the unloading process through images captured by cameras. From the moment of recognition, the waste is tracked throughout the bunker, as the AI receives information about crane coordinates and plant parameters. It considers changes such as storage time, mixing, and drying. The AI continuously calibrates itself based on combustion information to autonomously improve its precision for plants with different characteristics or previously unknown types of waste. In fully automatic operation, the optimization algorithm plans strategies to optimally utilize the energetic potential of the waste bunker content, taking into account the individual conditions of each waste-to-energy plant.
StableFlame enhances the efficiency of waste combustion, reduces the CO2 footprint, and optimizes the energy recovery process. It helps to lower operational costs by minimizing manual interventions and improving the overall performance of the facility.
The implementation of StableFlame in existing thermal waste-to-energy facilities follows a standardized scheme designed to seamlessly integrate the technology into operations and achieve immediate efficiency gains. The StableFlame expert team conducts a thorough preliminary analysis to determine the specific conditions and technical requirements of the facility. Together with the client, the project plan for commissioning is designed. Based on this analysis, necessary technical preparations are made, such as installing or adjusting camera systems and integrating into the existing IT infrastructure. Subsequently, the digital twin's training phase begins, equipped with advanced AI software to optimize the waste recovery process. Careful calibration and customized adjustment to the facility ensure high precision and efficiency. Comprehensive training of the operating personnel ensures a smooth transition and maximum utilization of the new technology. After commissioning, the StableFlame team provides continuous support and maintenance to ensure the long-term performance and reliability of the system. This thoughtful approach not only minimizes the impact on ongoing operations but also allows for significant improvements in terms of efficiency, sustainability, and economic viability of waste recovery.
StableFlame's software runs on-premise at the operators' thermal waste-to-energy facilities. A virtual machine is required, on which the StableFlame team installs and commissions all necessary systems. A VPN access for the StableFlame team is necessary to ensure monitoring and maintenance.
The StableFlame process is specifically designed for the standard construction of about 90% of all thermal waste-to-energy plants that have one waste bunker per boiler and do not have a mechanical mixing and homogenization facility. Depending on the current level of digitalization of the plant, the necessary work to create interfaces to existing systems, such as the crane or fire performance control, varies. The newer the systems of a facility are, the lower the effort for StableFlame integration. If camera systems are already installed, the StableFlame team will review their usability. If these are not usable or no cameras are installed, StableFlame will specify the required camera systems.
After implementation, StableFlame provides continuous support and maintenance to ensure the system's optimal performance. Our team is available for technical support, updates, and further optimizations.
