Optimizing heat exchanger design of biomass boilers through CFD simulation
Biocurve manufactures condensing biomass boilers for central heating and hot water systems for domestic users. The design of such boilers is currently based on the experience of the Biocurve technical staff. A prototype is designed, and then the original design is improved through trial-and-error cycles. The only software used in the design process is computer-aided design (CAD) software (viz SolidWorks) and spreadsheets. Currently, the time needed to design a prototype of a condensing biomass boiler is close to 6 months. An additional 6 months are necessary to bring the prototype boiler to the market (construction, refinement, testing and official approval). With this current methodology, the capability of Biocurve to develop new boiler models or improve existing ones is very limited.
Goals of the experiment
The objective of this experiment has been the introduction of computational fluid dynamics (CFD) tools, integrated in a cloud environment, in the design cycle for the condensing biomass boilers of Biocurve. The CFD tools (automatic mesh generator, CFD solver and CFD post-processing) have been applied to the simulation of the heat exchanger, one of the main components of the boilers manufactured by Biocurve. One of the challenges of the experiment has been the creation of a user-friendly interface and workflow. This has been a main requirement of the experiment, since Biocurve staff has never employed CFD tools previously.
The resulting CloudFlow application allows Biocurve to design a heat exchanger not only in terms of thermal performance but also in terms of material use (fewer tubes or shorter tubes). This was not possible with the previously existing workflow. Also, this application enables Biocurve to research the performance of a number of possible modifications or improvements that, up to now, was also unfeasible because it would have required building physical prototypes. As a result, in terms of the product design and innovation, the design of the heat exchangers can be optimized, both in the flue gas side (number, total length, diameter, curvature and slope of flue gas pipes) and in the water side (water tank geometry).
During the experiment, the following technical improvements have been achieved:
- Development of a virtual model of the current 25 kW boiler model in which the number of pipes has been reduced from 10 to 3. The reduction of the number of pipes in this model represents a saving of 18 kg of stainless steel (a 32 percent of the original weight of the pipes of this boiler). This lower number of pipes implies savings of raw material, but also, savings in fabrication costs (lower hours of workforce required, smaller insulation needed, less paint) and transport costs (lower volume of the boiler). The total cost saved estimated for this model is around 400 euros per unit.
- Adapting an elliptical geometry for the tubes (currently the pipes have a circular geometry)
- The water side has been modified in order to get a more homogenous distribution of the water mass flow throughout the water tank
- As a result of these modifications, the volume of the 25 kW boiler model has been reduced by 30 percent. This reduction saves material costs and, also, allows the installation of these boilers in a greater number of houses, since space is often of essence.
Also, the CloudFlow application contributes to reduce the time (and costs) to bring a new boiler to the market. Thus, the time-to-market for a new boiler can be reduced from currently 1 year to 8 months (a time reduction of 33 percent).
The use of the CloudFlow application can save Biocurve around 23,000 euros in the design of a new boiler model, taking into account the reduction of the number of prototypes should be built and, as a consequence, the lower time-to-market required. The Cloud application would allow Biocurve to increase the number of new models developed per year (currently, Biocurve is able to design 1 new model per year). The reduction of the size of the boilers due to a better design can save around 15 percent of the total costs of a boiler.
This product (condensing biomass boilers) is positioned in the residential/commercial pellet boilers market. According to the European Bioenergy Outlook, edited by AEBIOM (http://www.aebiom.org/blog/category/publications/statistics/), the European potential market for residential pellet boiler (up to 50 kW) is expected to grow up to 2020 in more than 500,000 units, as well as some 30,000 commercial pellet boilers (more than 50 kW). Biocurve analyses Europe as a whole market and their boilers comply not only with European legislation, but also with local regulations (particularly British, German, French, Austrian, Italian and, of course, Spanish legislations). With the CloudFlow application, Biocurve will be enabled to speed up the maturity of some products (15 kW, 25 kW or 100 kW) even 1 year in advance and enter in some currently unaffordable markets thanks to reducing costs (that is, low outputs markets and Eastern Europe market). It is estimated that sales of current boilers will be increased by 80-100 units/year, new sales of lower output range boilers will reach 100-150 units/year and of higher output boilers, 50-80 units/year. Addressing a customer segment defined by the domestic and commercial (tertiary) heating sector, in a three-year horizon Biocurve estimates to face a market around 600,000 potential buyers, with a potential share reaching 0.05 percent and leading to incomes of 2.55 million euros. Between 3 to 6 new jobs would be created if this volume of sales will be reached.
From the point of view of nablaDot, this experiment has provided nablaDot with a new business model, through the development of CFD tools on the cloud. This business model can be offered both to SMEs and large companies. Currently, nablaDot’s business (approximately 90 percent of the turnover) is based on CFD consultancy using commercial CFD software and in-house computational resources. In the long term (5 years from now), it is expected that 40-50 percent of the turnover (around 150,000 euros) will be related to ad-hoc development and supporting of CFD tools used in the cloud. 3 new engineers are expected to be hired thanks to this line of business.
UNIZAR-BIFI provides computing power to mainly local manufacturing companies. In the medium term (3 years), the market size of HPC centers (companies demanding power computing) is expected to be twentyfold. This experiment will help to expand its services to a wider geographical range and to maintain its remarkable position as a HPC center. According to the expectations, this will represent an increase of incomes between 30,000 euros and 150,000 euros in the next 3 years and the creation of new jobs (between 2 and 5).
The results of this CloudFlow experiment are an excellent proof of the advantages of the cloud-based simulations. The access to HPC resources through the Cloud offers relevant benefits: obviously, simulations are run faster, more complex models can be implemented or alternative designs can be calculated in parallel; as a result, the design process is not only improved but accelerated, since best solutions can be find sooner. Thus, the tool developed in this experiment is expected to be offered to at least 15 manufacturers of heat exchangers (from different sectors, such as oil and gas industry, food industry, power sector or automotive sector).