Intelligent cold production: an example of applied AI and digital twins at the Karlsruhe Stadtwerke

Karlsruhe Municipal Utilities: Intelligent systems for urban cooling

The need to use energy more efficiently is becoming increasingly important in times of climate change and rising energy costs. This applies not only to heating and electricity supply, but also to cooling, which is indispensable, especially in modern, densely populated urban areas and for specialized applications such as data centers or medical facilities. Municipal energy suppliers, such as Stadtwerke Karlsruhe, bear a special responsibility in this regard. They must not only guarantee security of supply, but also find innovative ways to reduce operating costs while simultaneously contributing to sustainability. An impressive example of how this can be achieved is the optimization of cooling for the GartenCarée site in Karlsruhe using advanced digital technologies.

As a municipal utility company with approximately 1,100 employees, Stadtwerke Karlsruhe reliably supplies well over 400,000 people in the region with essential energy sources such as electricity, natural gas, district heating, and drinking water. Part of this broad range of responsibilities includes the operation and maintenance of complex energy systems for large building complexes, such as the cooling system for the GartenCarée. This complex is a multifunctional building complex housing a mix of apartments, office space, medical practices, retail outlets, restaurants, and a critical data center. An uninterrupted and demand-driven cooling supply is essential for the smooth operation of all units.

The existing refrigeration system at GartenCarée is by no means simple in its design. It combines different technologies to enable the most flexible and potentially energy-efficient supply possible. At its core are three absorption chillers, which offer the advantage of being able to operate not with expensive electricity, but with environmentally friendly district heating or even waste heat from a nearby refinery. This system is complemented by a conventional compression chiller with two compressors, which is used when cooling demand is high or under specific operating conditions, as well as a cooling tower that dissipates excess heat into the environment.

Despite the fundamental possibility of flexibly responding to cooling demand and outside temperatures by combining different energy sources, the Karlsruhe municipal utility company faced a common challenge: a lack of detailed transparency regarding the actual energy consumption of the entire system and the interaction of the individual components. Before the implementation of a smart solution, insights into efficiency were essentially limited to the annual balance sheet. While it was possible to determine at the end of the year how many megawatt-hours of cooling were produced and how much electricity and district heating were used in total, a detailed analysis of which component consumed how much energy at what time and whether this was in optimal proportion to the cooling produced was not possible. Those responsible at the municipal utility company aptly described this situation as a kind of "flying blind," as the potential for targeted optimizations remained largely hidden.

Project goal: Greater transparency and significant efficiency gains

The clear goal of the Karlsruhe municipal utilities was therefore to end this lack of transparency. They wanted not only to identify potential savings but also to actively and automatically implement them. The solution lay in the implementation of an automated energy system management system. This system, developed by the specialist Factor4Solutions, is based on AI-supported software and utilizes the concept of the digital twin. The introduction of this system enabled a fundamentally new way of controlling the refrigeration system as well as highly detailed energy monitoring.

The installation of the software and the necessary sensors made all energy flows within the refrigeration system visible and measurable in detail for the first time. Parameters such as the volume flow rate of the chilled and cooling water, the temperatures at various points in the system, and the exact energy consumption (electricity and district heating) of each individual component – ​​from the absorbers to the compressors and pumps, right up to the cooling tower – are now recorded and analyzed in real time. This comprehensive data foundation is the key to optimization. The software can now continuously analyze which operating conditions are the most economical. A simple yet effective example of such optimization potential, revealed by transparent data acquisition, is pump control. Often, pumps in refrigeration systems run at 100 percent volume flow regardless of actual demand. However, detailed analysis shows that, depending on the required cooling capacity and system condition, 60 percent or even less would often suffice. Manually or invariably operating such units leads to significant, unnecessary electricity consumption.

A frequently overlooked problem in refrigeration systems, which only becomes apparent through detailed analysis, is the power consumption of auxiliary equipment. Pumps, fans in the cooling tower, or other auxiliary units can sometimes consume similar or even more power than the actual refrigeration units, especially if they are not optimally controlled. Furthermore, in many systems, optimization is limited to individual components or to an idealized operating state that is rarely achieved in reality – typically in less than three percent of all operating hours. This results in significant amounts of electrical energy being wasted during the majority of operating time. The targeted improvement of the interaction of all system components is precisely the specialty of the implemented solution.

After six months: Tangible successes and high potential

Factor4Solutions' intelligent software takes a holistic approach. It doesn't consider each component in isolation, but rather the entire refrigeration system as a dynamic unit. The energy consumption of all relevant units – chillers, pumps, fans, etc. – is evaluated together. This is made possible by the use of digital twins of the individual components and the entire system. A digital twin is essentially a virtual, high-precision model that mirrors the real system in real time. This model allows the AI ​​to calculate various operating scenarios and continuously determine how the individual components ideally need to work together to achieve maximum system efficiency.

The system is no longer controlled according to rigid, predefined schemes (e.g., one specific machine always runs as base load, another as peak load), but dynamically and based on demand. The selection and control of the units is now based on current weather conditions (outside temperature), the actual cooling demand, and the determined efficiency. Using standardized communication protocols, the intelligent system management system selectively activates and regulates the output of individual chillers and auxiliary units. The result is operation that achieves the highest possible efficiency at the system level in every conceivable load situation and under all environmental conditions.

The effectiveness of this approach became apparent at the Karlsruhe municipal utility company shortly after the completion of the test phase and the full commissioning of the system. Within the first six months of switching to intelligent control, the utility company recorded a significant reduction in annual electricity consumption – a saving of over 20 percent compared to the same period of the previous year. This efficiency gain is directly reflected in operating costs and has already resulted in five-figure financial savings within this first six months.

But according to experts, this is just the beginning. The first six months of operation under the new control system also included periods in which the system ran partially or entirely according to the old, conventional controls. The full potential of the intelligent solution unfolds over a complete annual cycle, as the optimal operating mode of the various chillers (absorber, compression, free cooling) depends heavily on the season, the outside temperature, and the specific cooling demand. Calculations by Factor4Solutions, based on initial operating data and simulations with the digital twin, predict potential energy savings of over 40 percent over a full operating year. This ambitious goal is now firmly in the sights of the Karlsruhe municipal utility company.

From local to global: SMEs conquer the global market with clever strategies - Image: Xpert.Digital

At a time when a company's digital presence determines its success, the challenge is how to make this presence authentic, individual and far-reaching. Xpert.Digital offers an innovative solution that positions itself as an intersection between an industry hub, a blog and a brand ambassador. It combines the advantages of communication and sales channels in a single platform and enables publication in 18 different languages. The cooperation with partner portals and the possibility of publishing articles on Google News and a press distribution list with around 8,000 journalists and readers maximize the reach and visibility of the content. This represents an essential factor in external sales & marketing (SMarketing).

More about it here:

• Authentic. Individually. Global: The Xpert.Digital strategy for your company

100 percent operational reliability through intelligent monitoring

Besides pure energy efficiency and cost reduction, the operational reliability of the system is of paramount importance to a utility company like Stadtwerke Karlsruhe. Especially when supplying a data center, where even brief outages can have serious consequences, the continuous availability of cooling is absolutely essential. The intelligent solution offers another crucial advantage here: The system is not only set up once and then managed independently, but is continuously monitored, tested, and optimized remotely by the provider, Factor4Solutions.

While the municipal utility company has its own monitoring systems and qualified personnel, the additional, specialized support from Factor4Solutions' expert team, with their in-depth know-how in refrigeration technology and AI-supported system optimization, represents a valuable complement. Through continuous remote monitoring using AI, the system can often detect anomalies or potential problems before they lead to a malfunction. Thanks to the digital twin and the real-time comparison of current operating data with the model's ideal state, deviations can be identified early, and customers can be proactively informed about impending failures or necessary maintenance measures. This enables predictive maintenance and avoids unplanned downtime. Ensuring operational readiness at all times is of fundamental importance to the Karlsruhe municipal utility company, not least because of its contractual obligation to supply cooling, particularly to the sensitive data center in GartenCarée.

Introduction and integration of the solution

From the municipal utility's perspective, the introduction and implementation of the smart solution went smoothly and without complications. Collaboration with the external service provider, who already manages the utility's measurement and control technology, was essential in this process.

Together, the data points required for the software setup were identified and approved. Factor4Solutions, which remotely manages the solution from its Berlin location, had clearly communicated in advance which specific data points from the plant's control system were needed. These included, for example, flow rates, temperatures, and the precise readings from the heat and cold meters.

Once this data was provided, the software was configured very quickly. Within a day, the system was basically set up and could begin data collection and initial analysis. After only four to six weeks, a first comprehensive analysis of the plant data was available. Based on the insights gained from this analysis, the system could be further fine-tuned. For example, it became clear which additional data points would be helpful for even more precise optimization, and the system was expanded and calibrated accordingly.

This relatively simple and quick integration makes the technology fundamentally interesting for a wide variety of applications and industries, wherever complex cooling systems are in use and their efficiency needs to be increased.

Rapid amortization as a strong argument for expanding cooperation

The positive effects of intelligent cooling optimization became apparent very quickly and convincingly for the Karlsruhe municipal utilities. A comparison of annual energy consumption and cooling production with the previous year, even though the new control system had only been fully operational for six months, already showed a significant increase in efficiency and measurable energy savings.

Particularly encouraging is the economic aspect: the system's amortization period is remarkably short. The annual costs for using the software and the ongoing support from Factor4Solutions' refrigeration specialists are offset by the energy savings achieved within the first five months of operation. This means that from the sixth month of the year onward, the system generates net cost savings that exceed the initial system costs. This rapid return on investment (ROI) is a compelling argument for implementing such technologies.

Given the positive experiences and proven efficiency gains, it's no surprise that Karlsruhe's municipal utility company is not only considering expanding its collaboration with Factor4Solutions, but is already implementing it. Another of the utility's cooling systems is now also being managed and optimized by the intelligent system. Furthermore, the utility operates a number of other cooling systems of varying sizes and complexity that are potentially suitable for optimization using the AI ​​solution. Since the technology has proven to be flexible and integrable into virtually all common cooling systems, there's a good chance that significant analyses, increased transparency, and further savings potential can be realized in these systems as well through the use of AI and digital twins.

The project at GartenCarée in Karlsruhe is a successful example of how modern digital technologies – especially artificial intelligence and digital twins – can revolutionize energy efficiency in a traditionally energy-intensive sector like refrigeration. By transforming operational facilities into transparent, intelligently controlled systems, energy suppliers can not only achieve significant savings in operating costs but also increase operational reliability and make a vital contribution to reducing energy consumption and thus protecting the climate. The success at the Karlsruhe municipal utility company demonstrates that investments in smart, data-driven optimization systems quickly pay for themselves and pave the way for a more sustainable and efficient energy supply. The initially impressive savings of over 20 percent and the projected potential of over 40 percent underscore the enormous potential inherent in the intelligent control and networking of energy systems.

☑️ SME support in strategy, consulting, planning and implementation

☑️ Creation or realignment of the digital strategy and digitalization

☑️ Expansion and optimization of international sales processes

☑️ Global & Digital B2B trading platforms

☑️ Pioneer Business Development

Konrad Wolfenstein

I would be happy to serve as your personal advisor.

You can contact me by filling out the contact form below or simply call me on +49 7348 4088 965 (Munich) .

I'm looking forward to our joint project.

Xpert.Digital is a hub for industry with a focus on digitalization, mechanical engineering, logistics/intralogistics and photovoltaics.

With our 360° business development solution, we support well-known companies from new business to after sales.

Market intelligence, smarketing, marketing automation, content development, PR, mail campaigns, personalized social media and lead nurturing are part of our digital tools.

You can find out more at: www.xpert.digital - www.xpert.solar - www.xpert.plus