Methods for AI-supported determination of site-specific, holistic energy concepts for heated buildings

Abstract

The invention relates to a method for AI-supported determination of site-specific, holistic energy concepts for heated buildings, characterized by the following steps Providing building-specific data via an input module (11) for at least one program module (12) and / or for an AI module (13), determining physical parameters of the building structure in at least one program module (12) and providing the parameters for an AI module (13), Providing parameters of technical solution components, legal frameworks, funding components and costs for solution components for an AI module (13), AI-supported determination of holistic optimized solutions according to physical logic in the AI ​​module (13) and Output of energy concepts according to physical logic or output of energy concepts according to national legislation, wherein the AI-supported determination of holistic optimized solutions in the AI ​​module (13) is based on state descriptions using unit vectors and observables from self-adjoint operators as isomorphically equivalent in Hilbert space. The invention further relates to a device, a computer device, a computer-readable storage medium and a computer program product for carrying out the method.

Description

[0001] The invention relates to an artificial intelligence (AI)-based method for the automated generation of site-specific, holistic energy concepts for heated buildings, as well as a device, a computer device, a computer-readable storage medium, and a computer program product for AI-supported determination of site-specific, holistic energy concepts for heated buildings.

[0002] Building operations in Germany are responsible for approximately 30 percent of greenhouse gas emissions. According to current German legislation, specifically the Building Energy Act (GEG), which implements the EU Energy Performance of Buildings Directive, the building sector must be climate-neutral by 2045. The European Union's climate targets, as outlined in the EU Energy Performance of Buildings Directive, also stipulate that the building sector must achieve climate neutrality by 2050.

[0003] For the upcoming transformation process, building owners require the support of energy consultants. These consultants must be proficient in legal requirements and possess the necessary cross-sectoral expertise in building physics, building services engineering, renewable energies, and energy storage. Continuous technological advancements and increasing regulatory and legal requirements are constantly increasing the complexity of the parameters to be considered. The method according to the invention offers the possibility of significantly improving the energy consulting process for heated buildings and developing site-specific energy concepts much more effectively and with higher quality in the future, thereby reducing energy consumption and lowering building operating costs.

[0004] Various methods are known in the art that use software to perform calculations for energy performance certificates according to DIN V18599. It should be acknowledged that these methods employ a holistic site analysis and partially integrate economic considerations into the process. However, a disadvantage is that they rely solely on key performance indicators and only partially represent real-world physical processes. The calculated results often deviate from actual consumption by a factor of 2 to 3, and no solutions are suggested. Only variants entered by the energy consultant are verified using this method.

[0005] Furthermore, simulation methods and corresponding software products are known that can perform complex simulations, particularly in the field of plant engineering. It should be acknowledged that these methods calculate energy consumption based on physical principles. However, a disadvantage is the lack of a holistic site analysis, as either the plant engineering or the building physics determines the essential parameters of the simulation software. Another drawback is the very high modeling effort required for each variant, and the results are only of limited use for comparing different variants. Additionally, the software requires in-depth technical knowledge from the operator, and the fact that only limited solutions are usually available for specific simulations is a disadvantage.

[0006] Open source simulation software provides partially and fully coupled analyses of heat networks, energy systems and buildings, some with high levels of detail.

[0007] A disadvantage is that extensive expert knowledge is required to operate the software. Creating simulation models requires significant effort and detailed knowledge, as well as high computational demands and consequently longer processing times.

[0008] In contrast to the aforementioned simulation software applications, publicly accessible configurators are also available that are relatively easy to use and deliver quick results. However, a disadvantage of these configurators is that they do not allow for a realistic analysis and generally only achieve insufficient depth in building data capture. Furthermore, they do not offer any suggested solutions.

[0009] As the overview described above shows, currently only selected, complex simulation systems are capable of actively generating proposed solutions. However, these are limited to individual aspects of the plant engineering. A more holistic site analysis is currently only offered by energy consulting software, whose calculation kernel, however, is based solely on the calculation method according to DIN V 18599.

[0010] Since software support is currently suboptimal, the creation of site-specific energy concepts and renovation roadmaps by energy consultants is currently very time-consuming. A successful transformation of buildings—in Germany alone, this affects 19 million residential buildings and 2 million non-residential buildings—is hardly achievable with such processing times and the existing capacity of energy consultants. In addition to the processing time, the quality of standard energy consulting also needs to be questioned. Due to the many complex interrelationships, it is virtually impossible for consultants to provide competent advice on all sensible solution options and combinations.

[0011] Furthermore, a method and a device for predicting energy consumption, a device and a readable storage medium are known in the prior art from EP4085387 A1.

[0012] The invention includes a method and a device for predicting electricity consumption, a device, and a readable storage medium. The method comprises acquiring a reference variable generated over a historical period; acquiring the predicted electricity consumption in a target period by inputting a variable property into an electricity consumption prediction model, wherein the target period and the historical period are related to each other; and obtaining the electricity consumption prediction model by training a reference variable sample characterized by a sample electricity consumption. The method involves acquiring the reference variable, which includes a discrete reference variable and a continuous reference variable within the historical period.In this process, a property of the recorded reference variable is taken into account, and an extracted variable property is entered into a variable prediction model to output the predicted electricity consumption in the target period.

[0013] From WO2021133253 A1, a method for predicting energy consumption is known, wherein the method is applicable on a computer and comprises: acquiring initial historical data, wherein the initial historical data comprise historical energy consumption data of a first object; acquiring a real energy consumption sample of the first object based on the initial historical data; acquiring a simulated energy consumption sample of the first object based on the real energy consumption sample, wherein the simulated energy consumption sample comprises an energy consumption sample acquired by extension based on the real energy consumption sample; and calculating the predicted energy consumption of the first object based on the real energy consumption sample and the simulated energy consumption sample by invoking an energy consumption forecasting model.

[0014] CN114943565 A discloses a method for predicting the spot price of electrical energy, which is based on an intelligent algorithm and is characterized by the fact that it comprises the following steps: Collecting historical resource exchange data and influencing factor data, wherein the historical resource exchange data are used to indicate the clearing price of the target energy in a historical period and the influencing factor data are used to indicate factors that influence the clearing price of the target energy in a forecast period; analyzing the historical resource exchange data and the influencing factor data to obtain a data impact analysis result, wherein the data impact analysis result is used to display the impact analysis result of the influencing factor data on the settlement price in the forecast period; Analyzing historical resource exchange data by combining the data impact analysis results to obtain a predicted settlement price for the target energy during the forecast period; Determining a transaction strategy for the target energy source during the forecast period based on a target yield function and the predicted clearing price, wherein the transaction strategy includes a transaction amount and a resource exchange amount, and the target yield function is used to specify a yield target during the forecast period.

[0015] Currently, there is hardly any active support for energy consultants in developing site-specific energy concepts; instead, they are offered only fragmented, individual solutions as working tools. A particular disadvantage is the lack of support in finding solutions. Even for a single residential building, there are more than ten categories of solution components that must be configured individually for each location, including: renovation options, PV output, PV utilization, battery capacity, battery type, heat generator, heat transfer, hot water generation, heat storage, ventilation concept, funding concept, and much more. Coupled with the increasing complexity of legal frameworks, it is therefore almost impossible for consultants to oversee the entire range of solutions or to find an optimal combination of measures. The quality of an energy concept currently depends significantly on the individual knowledge and experience of the energy consultant.

[0016] The object of the invention is therefore to provide the energy consultant with a technical aid to develop optimal solutions with regard to energy requirements and costs.

[0017] In contrast to state-of-the-art solutions, energy consultants should receive support in the form of site-specific, holistic solution proposals, which significantly reduces the processing time per consultation process and increases and stabilizes the quality of the consultation.

[0018] The problem is solved by a method and articles with the features according to the independent patent claims. Further developments are specified in the dependent patent claims.

[0019] The object of the invention is solved in particular by a method for AI-supported determination of site-specific, holistic energy concepts for heated buildings, which is characterized by the following steps: Providing building-specific data through an input module for at least one program module, whereby building-specific data is additionally or alternatively provided directly for an AI module. Determining the physical parameters of the building structure in at least one program module and providing these parameters to an AI module. Providing parameters for technical solution components, legal frameworks, funding options, and costs for solution components to an AI module. AI-supported determination of holistically optimized solutions according to physical logic in the AI ​​module and Output of energy concepts according to physical logic or Output of energy concepts according to national legislation, whereby the AI-supported determination of holistic optimized solutions in the AI ​​module is based on state descriptions using unit vectors and observables from self-adjoint operators as isomorphically equivalent in Hilbert space.

[0020] Unit vectors are defined as normalized vectors with the norm of one, and observables are measurable quantities with an associated operator. Self-adjoint operators, as defined by Schrödinger and Heisenberg, are linear operators with special properties.

[0021] In the context of functional analysis, isomorphic / equivalent uniting means that parts of one structure are mapped to equivalent parts of another structure.

[0022] The presented approach allows for the simplest possible representation of arbitrarily complex problems through the optimal choice of an individual Hilbert space.

[0023] AI-supported development of site-specific, holistic energy concepts for heated buildings creates a quasi-digital twin—a comprehensive representation of individual buildings across numerous parameters. This approach to solving the problem of energy concept development opens up a wide range of options for energy optimization, taking into account heterogeneous external factors. The method employed is similar to the representation of states and observables in quantum mechanics using Hilbert spaces.

[0024] As a result, the individually optimal energy measure is found by solving the multidimensional problem using machine learning as a subfield of artificial intelligence.

[0025] Advantageously, building-specific data is selected from data on the current state of the plant technology, consumption and load profiles, customer requirements, and site-specific boundary conditions and transmitted to the AI ​​module.

[0026] Preferably, building-specific data is transmitted to the program module as building-specific boundary conditions.

[0027] An advantageous embodiment of the method consists in the fact that the heat capacity of the building structure, transmission heat losses with current status and variants and / or photovoltaic yields with current status and variants are transmitted to the AI ​​module as physical parameters.

[0028] Preferably, parameters of technical solution components, legal frameworks and funding components and / or costs for solution components are transmitted from intelligent databases to the AI ​​module.

[0029] The AI ​​module advantageously calculates 10,000 to 50,000 simulated options per building.

[0030] The task is also solved by a device for AI-supported determination of site-specific, holistic energy concepts for heated buildings with an input module, a program module, intelligent databases, an AI module, and an output according to physical logic and an output according to national legislation, which is suitable for carrying out a procedure of the specified type.

[0031] The task is further solved by a computer device comprising a processor and a memory, wherein at least one program is stored in the memory and wherein the at least one program is loaded and executed by the processor to implement the procedure for AI-supported determination of site-specific, holistic energy concepts for heated buildings.

[0032] A computer-readable storage medium with at least one program code stored therein, wherein the at least one program code is loaded and executed by a processor to execute the method for AI-supported determination of the location of individual, holistic energy concepts for heated buildings solves the problem of the invention.

[0033] Finally, the object of the invention is solved by a computer program product that comprises a computer program which, when executed by a processor, implements a method for AI-supported determination of individual, holistic energy concepts for heated buildings according to the aforementioned method.

[0034] The invention's concept consists of providing an AI-supported method for the automated generation of site-specific, holistic energy concepts for heated buildings. The method utilizes an algorithm with an associated intelligent database.

[0035] The method according to the invention offers the possibility of significantly improving the energy consulting process and developing site-specific energy concepts more effectively and with higher quality. The AI-supported algorithm accesses a comprehensive database of solution modules and, during the consulting process, provides holistic technical recommendations for individual buildings or groups of buildings, incorporating technical parameters and solutions. This reduces the processing time for energy consulting concepts in residential buildings by 65% ​​and in non-residential buildings by as much as 75%. In addition to increasing efficiency, this approach also enables a more consistent quality of consulting, which then depends less on the individual experience and knowledge of the consultant.

[0036] Further details, features, and advantages of embodiments of the invention will become apparent from the following description of exemplary embodiments with reference to the accompanying drawings. These show: Fig. 1: Schematic AI algorithm and Fig. 2: Process flow diagram.

[0037] In Fig. Figure 1 schematically shows the AI ​​system 1 with the AI ​​algorithm as the central element of the procedure, along with the connected input parameters and databases.

[0038] Examples of technology components (2) include electricity generators such as photovoltaics or wind power, heat generators such as geothermal energy, heat pumps, furnaces, burners, insulation such as insulating materials, composite insulation systems, electrical storage technology and temperature management, or similar. In addition, the calculations incorporate customer requirements (3), building parameters (4), site parameters (5), the potential for renewable energies (6), the relevant laws (7), funding programs (8), installers (9), and a scan (10) of the building.

[0039] Customer requirements 3 include parameters such as maximum funding, maximum self-sufficiency and, if necessary, the exclusion of various technology components.

[0040] Building parameter 4 includes, for example, the dimensions of the building, the energy consumption of the current building, and the nature of the building's thermal envelope.

[0041] Site parameter 5 identifies the building site including the adjacent terrain, and any significant shading with regard to photovoltaics and wind energy.

[0042] The potential for renewable energies 6 is determined by the local statistics for solar, wind, and geothermal energy.

[0043] The relevant laws include, for example, the EU Building Directive, the Building Energy Act (GEG), state building codes, and other regional regulations and provisions.

[0044] The funding programs 8, including EU, federal and state funding, also influence the feasibility and economic viability of the projects.

[0045] The selection of installers for photovoltaics, wind energy, insulation, and similar systems is factored in depending on the building's location, with the aim of achieving the optimal combination of shortest travel time and best assessment. Finally, a drone scan of the building or the integration of GIS data to support building parameters and location parameters also influences the optimization process.

[0046] In Fig. Figure 2 is a schematic representation of the process flow diagram of the procedure for AI-supported determination of site-specific, holistic energy concepts for heated buildings.

[0047] The process is carried out using various modules. On-site data acquisition is input via an input module 11. The building-specific boundary conditions are fed into an external program module 12 with commercial software, which determines, for example, the thermal capacity of the building structure, the transmission heat losses in the current state and of variants, as well as photovoltaic yields in the current state and of variants, and forwards this information to the internal AI module 13 with the AI ​​algorithm.

[0048] The parameters of the plant technology in its current state, consumption and load profiles, customer requirements and site-specific boundary conditions are transferred directly from input module 11 to AI module 13.

[0049] In AI module 13, after the limit values ​​have been defined, AI-supported, holistic optimization takes place according to physical logic, including an energy balance for the building, including process energy. This is followed by results processing, after which output modules 14 and 15 present the results according to physical logic or with a focus on national legislation.

[0050] The AI ​​module 13 receives parameters from intelligent databases 16, for example, of technical solution components, legal frameworks and funding components, as well as the costs for the solution components, which are fed into the optimization process of the AI ​​module 13.

[0051] The solution to the multidimensional matrix problem in Hilbert space can be calculated as the optimal solution L from the scalar product of all relevant observables A in the state ψ with measured value α as follows: L=|<α|ψ>|2

[0052] Here, the observable A is described by the operator  and |α > is the eigenstate corresponding to the eigenvalue α of the operator Â.

[0053] Solution L can be understood as an optimization with regard to building energy consumption.

[0054] A sketched matrix structure of relevant observables is shown and described below. Reference symbol list 1 AI system 2 technology building blocks 3 customer requests 4 Building parameters 5 location parameters 6. Potential of Renewable Energy 7 laws 8 funding programs 9 installers 10 scans 11 Input module 12 External program module 13 Internal AI module with AI algorithm 14 Output module according to physical logic 15 Output module according to national legislation 16 Intelligent Databases QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] EP 4085387 A1

[0011] WO 2021133253 A1

[0013] CN 114943565 A

[0014]

Claims

[1] Methods for AI-supported determination of site-specific, holistic energy concepts for heated buildings, characterized by the following steps: - Provision of building-specific data by an input module (11) for at least one program module (12) and / or for an AI module (13), - Determining physical parameters of the building structure in at least one program module (12) and providing the parameters for an AI module (13), - Providing parameters of technical solution components, legal frameworks, funding components and costs for solution components for an AI module (13), - AI-supported determination of holistic optimized solutions according to physical logic in the AI ​​module (13) and - Output module (14) of energy concepts according to physical logic or - Output module (15) of energy concepts according to national legislation, wherein - the AI-supported determination of holistic optimized solutions in the AI ​​module (13) on state descriptions using unit vectors and observables from self-adjoint operators as isomorphically equivalent united in Hilbert space takes place. [2] Method according to claim 1, characterized by , that building-specific data are selected from data on the current state of the plant technology, consumption and load profiles, customer requirements, site-specific boundary conditions and are transmitted to the AI ​​module (13). [3] Method according to claim 1 or 2, characterized by , that building-specific data are transmitted to the program module (12) as building-specific boundary conditions. [4] Method according to claim 3, characterized by , that the heat capacity of the building structure, transmission heat losses with current status and variants and / or photovoltaic yields with current status and variants are transmitted to the AI ​​module (13) as physical parameters. [5] Method according to any one of claims 1 to 4, characterized by , that parameters of technical solution components, legal frameworks and funding components and / or costs for solution components are transmitted from Intelligent Databases (16) to the AI ​​module (13). [6] Method according to any one of claims 1 to 5, characterized by , that 10,000 to 50,000 simulated options per building are calculated using the AI ​​module (13). [7] Device for AI-supported determination of site-specific, holistic energy concepts for heated buildings with an input module (11), a program module (12), intelligent databases (16) as well as an AI module (13) and an output according to physical logic (14) and an output according to national legislation (15). [8] Computer device comprising a processor and a memory, wherein at least one program is stored in the memory and wherein the at least one program is loaded and executed by the processor to implement the method for AI-supported determination of site-specific, holistic energy concepts for heated buildings according to any one of claims 1 to 6. [9] Computer-readable storage medium with at least one program code stored therein, wherein the at least one program code is loaded and executed by a processor to implement the method for AI-supported determination of site-specific, holistic energy concepts for heated buildings according to any one of claims 1 to 6. [10] Computer program product comprising a computer program which, when executed by a processor, implements a method for AI-assisted determination of site-specific, holistic energy concepts for heated buildings according to any one of claims 1 to 6.

Images