Energy renovation process for a building, or part of a building, and associated computer programs

Abstract

The invention aims to provide a method for energy renovation of a building making it possible to reduce the energy requirement for its implementation and the negative ecological impact linked to its implementation.The solution proposed by the invention is a method for energy renovation of a building, or part of a building, remarkable in that it comprises the steps: a) Providing a substitution mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, as a function of one or more input variables, b) Acquiring, for one or more of the input variables, data relating to the building, or part of a building, to be renovated, c) Calculating, from the substitution mathematical model and the acquired data, the energy impact data, d) Determining, as a function of the energy impact data, the relevance of the energy renovation(s) for the building, or part of a building, to be renovated, e) If the relevance is proven, carrying out the energy renovation(s) of the building, or part of a building, to be renovated.

Description

Title of the invention: Method for energy renovation of a building, or part of a building, and associated computer programs Technical field of the invention

[0001] The present invention relates to the technical field of methods for renovating a building, or part of a building, and more particularly to renovation methods aimed at reducing the energy consumption of the building, or part of a building. State of the art

[0002] Current energy renovation processes use software for thermal modeling of a building. Energy renovation processes are for example described in documents US2012284124 (Harangozo & Mate) and EP2667332 (Calao Énergie Renouvelable).

[0003] As part of these energy renovation processes, a first step generally consists of bringing a diagnostician to the site to collect information on the building, as well as to carry out a certain number of measurements. In practice, these measurements and information collected correspond to around a hundred data. In addition to the complexity of collecting this data, the movement of a diagnostician on site generally requires the use of a means of transport, therefore generating energy consumption with a negative ecological impact.

[0004] A second step for the diagnostician is to enter the collected data into software. In addition to the tedious nature of entering a hundred or so data, the necessarily long time spent on said entry also generates energy consumption. Indeed, the more time the diagnostician spends on entry, the more energy he consumes, as a computer in operation consumes energy.

[0005] A third step consists of calculating an estimate of the building's energy consumption in its current state using thermal modeling software. Thermal modeling software is used to model the heat flows of a building. This thermal modeling is based on the theories of thermodynamic physics. In particular, these models generally involve the "Th CEex" or "3 CL" calculation methods that comply with current energy standards. To calculate an estimate of a building's energy consumption, one or other of these methods requires a large amount of input data. For example, for a 20 m2 building, the calculation requires between 1,000 and 3,000 input data. Thus, in addition to the hundred or so data entered by the diag nosticeur, many other data necessary for the calculation are then extrapolated. The execution of this calculation based on this large amount of data is extremely complex. It then requires powerful computer hardware, as well as a significant execution time. The provision of specific computer hardware as well as the significant execution time therefore generates significant energy consumption and a negative ecological impact. In addition, since physical models are generally imperfect, the result obtained remains an estimate of the building's energy consumption. Also, the large amount of input data and the complexity of identifying them constitute major sources of error in the results obtained.

[0006] Once the energy consumption of the building in its current state is calculated, a fourth step consists of calculating the energy consumption of the building after renovation. The diagnostician then enters new data corresponding to the energy renovations to be made to the building. As before, a large amount of data is entered, a large amount of data is then extrapolated by the software, and the latter finally calculates an estimate of the energy consumption of the building after renovation. If the client wishes to simulate the impact of different renovations on the energy consumption of the building, one calculation per renovation must be carried out. The greater the number of simulations, the greater the energy consumption.

[0007] Based on all of these estimates of the building's energy consumption, an additional step is for the client to make a decision on the renovations to be carried out by balancing the benefit and the cost.

[0008] Finally, a final stage of renovation of the building is then implemented on the basis of this arbitration.

[0009] If taken independently and on the scale of a diagnosis, these energy consumptions and these negative ecological impacts may seem moderate, they are in reality significant when all the stages of the process are taken into account and the number of diagnoses carried out per day, in the world, is taken into account.

[0010] Therefore, the invention aims to provide a method for energy renovation of a building making it possible to reduce the energy requirement necessary for its implementation, as well as the negative ecological impact linked to this implementation. Disclosure of the invention

[0011] The solution proposed by the invention is a method for energy renovation of a building, or part of a building, remarkable in that it comprises the steps: • a) Provide a substitute mathematical model configured to calculate energy impact data associated with one or more renovations energy of a building, or part of a building, to be renovated, based on one or more input variables, • b) Acquire, for one or more of the input variables, data relating to the building, or part of the building, to be renovated, • c) Calculate, from the mathematical substitution model and the acquired data, the energy impact data, • d) Determine, based on the energy impact data, the relevance of the energy renovation(s) for the building, or part of the building, to be renovated, • e) If the relevance is proven, carry out the energy renovation(s) of the building, or part of the building, to be renovated.

[0012] The use of a substitution mathematical model may seem contrary to common sense for those skilled in the art, since the latter always strives to comply with the standards and therefore favors the use of the “Th CEex” or “3 CL” calculation methods which, in turn, comply with current standards. However, it turns out that the use of a substitution mathematical model makes it possible, compared to current physical models, to simplify the calculation, to reduce the number of input data necessary to obtain a result, but also to obtain a result whose precision is sufficient for the customer to make a decision. The reduction in the number of input data makes it possible to simplify the input and therefore to reduce the time required for said input. In addition, it also makes it possible to make the use of the energy renovation process accessible to everyone and therefore to avoid the need for a diagnostician to travel.Also, the simplicity of the mathematical substitution model as well as the reduced number of input data required make it possible to reduce the computing power requirement of the computer hardware and the execution time of the calculation in comparison with current energy renovation processes. Thus, the energy consumption and the negative ecological impact of the energy renovation process proposed by the invention are greatly reduced compared to current energy renovation processes.

[0013] According to an advantageous characteristic of the invention allowing the provision of a mathematical substitution model adapted to the energy renovation method which is the subject of the invention, step a) comprises the sub-steps: • al) Provide a database comprising several sets of energy diagnostic data of different renovated buildings, or different renovated building parts, each data set being associated with a different renovated building, or a different renovated building part, each data set being associated with a variable, said variables comprising an energy impact variable associated with one or several energy renovations of a building, or part of a building, and several other variables, • a2) Generate a mathematical substitution model from the data of the database by selecting the energy impact variable as output variable, and the other variables as input variables.

[0014] According to another advantageous characteristic of the invention making it possible to adjust the mathematical substitution model, step a) comprises the sub-step: • a3) Adjust the mathematical substitution model from a set of database data including data different from the data in the dataset used for generating the substitution mathematical model.

[0015] According to another advantageous characteristic of the invention making it possible to improve the precision of the generated mathematical substitution model, the number of variables is less than or equal to 50.

[0016] According to yet another advantageous characteristic of the invention making it possible to improve the energy efficiency of the energy renovation process, the number of input variables for which data is acquired is less than or equal to 10.

[0017] According to yet another advantageous characteristic of the invention making it possible to improve the precision of the calculation of the potential energy impact data, the input variables comprise: • A variable representing a habitable surface area of the building, or part of the building, to be renovated, and / or • A variable representing a construction period of the building, or part of the building, to be renovated, and / or • A variable representing an energy performance diagnostic score of the building, or part of the building, to be renovated, before energy renovation, and / or • A variable representing a geographical location of the building, or part of the building, to be renovated, and / or • A variable representing a type of heating used in the building, or part of the building, to be renovated, before energy renovation, and / or, • A variable representing the age of the heating system used in the building, or part of the building, to be renovated, before energy renovation, and / or, • For each type of energy renovation, a variable representing a selection of the type of energy renovation as energy renovation of the building, or part of the building, to be carried out, and / or • For each type of energy renovation, one or more variables re presenting the content of the energy renovation of the building, or part of the building, to be carried out, • A variable representing the energy consumption of the building, or part of the building, to be renovated, before energy renovation.

[0018] According to yet another advantageous characteristic of the invention making it possible to improve the precision of the calculation of the potential energy impact data, the mathematical substitution model is a model of the decision tree forest type.

[0019] According to yet another advantageous characteristic of the invention making it possible to improve the precision of the calculation of the potential energy impact data, step c) comprises the sub-steps: • cl) Determine, for the input variable(s) for which no data has been acquired, a default data item, • c2) Calculate, from the mathematical substitution model, the acquired data and the default data(s) determined, the energy impact data.

[0020] Another aspect of the invention relates to a computer program for implementing an energy renovation method according to the invention. This computer program is remarkable in that, when executed by a microprocessor, said computer program is configured to implement the steps: • a) Provide a substitute mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, based on one or more input variables, • b) Acquire, for one or more of the input variables, data relating to the building, or part of the building, to be renovated, • c) Calculate, from the mathematical substitution model and the acquired data, the energy impact data.

[0021] Such a computer program allows the generation and exploitation of a substitute mathematical model for the implementation of the energy renovation method according to the invention, and thus, to greatly reduce the energy consumption and the negative ecological impact associated with the implementation of said method compared to current energy renovation methods.

[0022] Yet another aspect of the invention relates to a computer program for providing a substitute mathematical model of an energy renovation method according to the invention. This providing computer program is remarkable in that, when executed by a microprocessor, said computer program is configured to implement the steps: • al) Provide a database comprising several sets of energy diagnostic data of different renovated buildings, or different renovated parts of buildings, each data set being associated with a different renovated building, or a different renovated part of building, each data set being associated with a variable, said variables comprising an energy impact variable associated with one or more energy renovations of a building, or part of building, and several other variables, • a2) Generate a mathematical substitution model from a set of data from the database by selecting as output variable, the energy impact variable, and as input variables, the other variables.

[0023] Thus generated, the mathematical substitution model has the advantage of being a simpler model than traditional physical models while making it possible to obtain a precise estimate of the energy impact of an energy renovation. It thus makes it possible to greatly reduce the energy consumption and the negative ecological impact associated with the implementation of the energy renovation method which is the subject of the invention compared to current renovation methods. Description of figures

[0024] Other features and advantages of the invention will emerge from reading the description given below of particular embodiments of the invention, given for information purposes, but not as a limitation, with reference to the appended drawings in which: • [Fig.l]: Figure [Fig.l] is a schematic representation of an example of an embodiment of an energy renovation method according to the invention, • [Fig.2]: Figure [Fig.2] is a schematic representation of an example of implementation of a method for providing a mathematical substitution model according to the invention. Detailed description

[0025] The invention relates to a method for the energy renovation of a building, or part of a building. The term "energy renovation method" means any method aimed at modifying, and / or adding, and / or replacing, one or more elements of a building, or part of a building, so as to reduce the energy consumption of a building, or part of a building. In particular, the term "energy renovation" means any work aimed at modifying, and / or adding, and / or replacing, one or more elements of a building, or part of a building, so as to reduce this energy consumption.

[0026] “Energy consumption” means the energy consumption of users of a building, or part of a building, into energy sources with constant habits. The energy source(s) may be electricity and / or gas and / or oil and / or wood and / or any other energy source suitable to the person skilled in the art. The energy source(s) are used for the production of heat or cold, the production of domestic hot water, active ventilation, water pumping, lighting the building and the operation of household appliances or any other equipment requiring an energy supply. Energy consumption can be calculated by taking into account only the dwelling, in which case we speak of "final energy consumption".Energy consumption can be calculated by taking into account the entire value chain, that is, not only the dwelling itself, but also the entire energy production and transport chain; this is known as "primary energy consumption." The latter has the advantage of taking into account energy losses throughout the energy production and transport process. In the following description, the term "energy consumption" refers indifferently to final energy consumption or primary energy consumption.

[0027] The building, or part of a building, may be for commercial use or for residential use. The building may be a house, a villa, a building, a shed or any other building suitable to the person skilled in the art. The part of the building may be an apartment, a commercial premises, or any other premises suitable to the person skilled in the art. When the building, or part of a building, is for residential use, the users may, for example, be a household.

[0028] The term "element" of a building, or part of a building, means any constituent element of the building, or part of a building, and / or any element installed inside and / or outside the building, or part of a building. In particular, these elements may be of the type interior or exterior insulation, roof, door, window, floor, ceiling, heat or cold production system, domestic hot water production system, ventilation, electricity production system, for example, of the solar panel type, electrical appliances, in particular household appliances, or any other element of a building, or part of a building suitable to a person skilled in the art.

[0029] As shown schematically in figure [Fig. 1], the process of energy renovation of a building comprises the step: • a) Provide a substitute mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, based on one or more input variables.

[0030] In practice, a computing device may be configured to provide the model substitution mathematics. The computing device may, among other things, include a microprocessor and a memory in which a computer program is stored. Execution of the computer program by the microprocessor may provide the substitution mathematical model.

[0031] Generally speaking, the term "surrogate mathematical model" (or "surrogate model" in English) refers to a mathematical model generated from data contained in a database, to which one or more input variables and one or more output variables are associated. Thus, the surrogate mathematical model is not constructed to simulate the laws of physics, but on the contrary to simulate input-output behavior. This is referred to as a bottom-up approach based on data. In this case, a surrogate mathematical model constructed to calculate energy impact data proves to be simpler, in terms of calculation, than traditional physical models. The execution of calculations by computer is then faster with a surrogate mathematical model than with traditional physical models conforming to current energy standards.

[0032] The mathematical substitution model of the energy renovation method according to the invention may then comprise one or more output variables. The output variable(s) comprise an energy impact variable associated with one or more energy renovations of the building, or part of the building, to be renovated.

[0033] According to a first embodiment of the energy impact variable, the latter may represent an estimate, over a given period, of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s). The period may be a year, or a month, or any other period suitable to a person skilled in the art. When the period is a year, the energy impact variable may then be expressed in kilowatt-hours per year (kW.h / year). In a similar embodiment, the energy impact variable may be expressed in joules per year (J / year).

[0034] According to a second embodiment of the energy impact variable, the latter may represent an estimate, over a given period, of the cost of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s). The period may be a year, or a month, or any other period suitable to a person skilled in the art. When the period is a year, the energy impact variable may then be expressed in currency per year, for example in euros per year (€ / year).

[0035] According to a third embodiment of the energy impact variable, the latter can represent an estimate, over a given period, of the quantity of greenhouse gas emitted in connection with the energy consumption of the building, or part thereof of building, after having carried out the energy renovation(s). The period can be one year, or one month, or any other period suitable to the person skilled in the art. When the period is one year, the energy impact variable can then be expressed in CO2 equivalent per year (CO2-eq / year).

[0036] According to a fourth embodiment of the energy impact variable, the latter may represent an estimate of the energy performance diagnostic score of the building, or part of the building, after having carried out the energy renovation(s). The energy impact variable may then be expressed with a letter between A and G.

[0037] According to a fifth embodiment of the energy impact variable, the latter may represent the difference between an estimate, over a given period, of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s) and an estimate, over said given period, of the energy consumption of the building, or part of the building, before energy renovation. The period may be a year, or a month, or any other period suitable to a person skilled in the art.

[0038] According to a first exemplary embodiment of the fifth embodiment of the energy impact variable, the period may be one year and the energy impact variable may be expressed in kilowatt-hours per year (kW.h / year). The energy impact variable is then equal to the difference between an estimate, over one year, of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s), and an estimate, over said year, of the energy consumption of the building, or part of the building, before renovation.

[0039] According to a second exemplary embodiment of the fifth embodiment of the energy impact variable, the period may be one year and the energy impact variable may be expressed in joules per year (J / year). The energy impact variable is then equal to the difference between an estimate, over one year, of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s), and an estimate, over said year, of the energy consumption of the building, or part of the building, before renovation.

[0040] According to a third exemplary embodiment of the fifth embodiment of the energy impact variable, the period may be one year and the energy impact variable may be expressed in currency per year, for example in euros per year (€ / year). The energy impact variable is then equal to the difference between an estimate, over one year, of the cost related to the energy consumption of the building, or part of the building, after having carried out the energy renovation(s), and an estimate, over said year, of the cost related to the energy consumption of the building, or part of the building, before energy renovation.

[0041] According to a fourth exemplary embodiment of the fifth embodiment of the energy impact variable, the period may be one year and the energy impact variable may be expressed in CO2 equivalent per year (CO2-eq / year). The energy impact variable is then equal to the difference between an estimate, over one year, of the quantity of greenhouse gas emitted in connection with the energy consumption of the building, or part of the building, after having carried out the energy renovation(s), and an estimate, over said year, of the quantity of greenhouse gas emitted in connection with the energy consumption of the building, or part of the building, before energy renovation.

[0042] In a preferred embodiment of the fifth embodiment of the energy impact variable, the energy impact variable may be expressed as a percentage (%) and be equal to the difference between an estimate, over a given period, of the energy consumption of the building, or part of the building, after having carried out the energy renovation(s), and an estimate, over said given period, of the energy consumption of the building, or part of the building, before energy renovation, the whole divided by said estimate, over said given period, of the energy consumption of the building, or part of the building, before energy renovation. The energy impact variable, thus calculated, makes it possible to more accurately compare the results obtained for different buildings, or parts of the building, for which the surface area, and / or the number of users, differ.Therefore, the mathematical substitution model obtained on the basis of this variant of realization of the energy impact variable is more precise compared to mathematical substitution models obtained on the basis of other variants of realization of the energy impact variable.

[0043] In other embodiments of the energy impact variable, the latter may be expressed in any other form suitable to those skilled in the art and making it possible to estimate the relevance of the energy renovation(s) for the building, or part of the building, to be renovated.

[0044] The substitution mathematical model may comprise several output variables and be configured to calculate data associated with each of the output variables. These output variables may comprise one or more of the previously described variants of the energy impact variable. These output variables may also comprise several energy impact variables, each associated with a different energy renovation combination. An energy renovation combination is understood to mean a combination comprising one or more energy renovations. An energy renovation may, for example, be of the type renovation of the heating system, renovation of the heating system control device, renovation of the air conditioning system, renovation of the system production of domestic hot water, renovation of the ventilation system, renovation of wall insulation, renovation of roof insulation, renovation of floor insulation, renovation of ceiling insulation, renovation of openings, installation or renovation of an electricity production system, or any other energy renovation suitable to the person skilled in the art.

[0045] The input variables may include a variable representing a habitable surface area of the building, or part of the building, to be renovated. This variable is called a “habitable surface area variable”.

[0046] The input variables may include a variable representing a construction period of the building, or part of the building, to be renovated. This variable is called a “construction period variable”. The construction period may be the year of construction. Preferably, the construction period is expressed in intervals of years, for example per period of 20 years.

[0047] The input variables may include a variable representing an energy performance diagnostic score of the building, or part of the building, to be renovated, before energy renovation. This variable is called a “diagnostic score variable”. This energy performance score may vary depending on the country. For example, in France, this score varies from A, extremely efficient housing, to G, extremely poor-performing housing.

[0048] The input variables may include a variable representing a geographic location of the building, or part of the building, to be renovated. This variable is called a "geographic location variable." In practice, the geographic location may take the form of GPS coordinates. The geographic location may take the form of a character string designating a geographic region. For example, each character string may refer to a state and / or a region and / or a department and / or a canton, and / or a city.

[0049] The input variables may include a variable representing a type of heating used in the building, or part of the building, to be renovated, before energy renovation. This variable is called a “heating type variable”. In practice, the heating may be of the electric heating type, air-to-air heat pump, air-to-water heat pump, gas boiler, oil boiler, open fireplace, insert, wood stove, or any other type of heating suitable to those skilled in the art.

[0050] The input variables may include a variable representing the age of the heating system used in the building, or part of the building, to be renovated, before energy renovation. This variable is called the "heating age variable". In practice, the heating age variable may be expressed in years, and / or months, or even in tens of years.

[0051] The input variables may include, for each type of renovation energy, a variable representing a selection of the type of energy renovation as energy renovation of the building, or part of the building, to be carried out. This variable is called "energy renovation type selection variable".

[0052] The input variables may include, for each type of energy renovation, one or more variables representing the content of the energy renovation of the building, or part of the building, to be carried out. This type of variable is called "energy renovation content variable". For example, this variable may indicate whether the energy renovation must be total or partial, or how many % or m2 of the element of the building, or part of the building, must be renovated. When the energy renovation is of the type renovation of the wall insulation, renovation of the roof insulation, renovation of the floor insulation, renovation of the ceiling insulation, the energy renovation content variable may indicate the type of material used for said energy renovation or the thickness of the insulation after renovation.When the energy renovation is of the heating system renovation type, the energy renovation content variable can, for example, indicate the type of heating system after renovation.

[0053] The input variables may include a variable representing, over a given period, an energy consumption of the building, or part of the building, to be renovated, before energy renovation. This variable is called “energy consumption variable before energy renovation”. The period may be a year, or a month, or any other period suitable to a person skilled in the art. When the period is a year, this variable may, for example, be expressed in kilowatt-hours per year (kW.h / year).

[0054] In an alternative embodiment of the energy consumption variable before energy renovation, said variable can be expressed in joules per year (J / year).

[0055] In another variant embodiment of the energy consumption variable before energy renovation, said variable can be expressed in currency per year, for example in euros per year (€ / year), and represent an estimate, over one year, of the cost of the energy consumption of the building, or part of the building, before energy renovation.

[0056] In another variant embodiment of the energy consumption variable before energy renovation, said variable can be expressed in CO2 equivalent per year (CO2-eq / year) and represent an estimate, over one year, of the quantity of greenhouse gases emitted in connection with the energy consumption of the building, or part of the building, before energy renovation.

[0057] The mathematical substitution model may, advantageously, be a model of the decision tree forest type (or “random decision forest” in English). The latter is a mathematical substitution model known from the prior art and based on the use of decision trees as a predictive model.

[0058] In alternative embodiments, the surrogate mathematical model may be of the type polynomial response surfaces, kriging, gradient-enhanced kriging, radial basis function, support vector machines, spatial mapping, artificial neural networks and Bayesian networks, Fourier surrogate modeling, multiple linear regression, or any other surrogate mathematical model suitable to those skilled in the art.

[0059] In a first embodiment of the provision of the substitution mathematical model, step a) may comprise the sub-step: • al) Provide a database comprising several sets of energy diagnostic data of different renovated buildings, each set of data being associated with a different renovated building, each data set being associated with a variable, said variables comprising an energy impact variable associated with one or more energy renovations of a building, or part of a building, and several other variables, • a2) Generate a mathematical substitution model from a set of data from the database by selecting as output variable, the energy impact variable, and as input variables, the other variables.

[0060] In practice, the computing device may be configured to provide the database. The execution of the computer program by the microprocessor may make it possible to provide the database. In particular, step a1) may comprise the following sub-step: • garlic) Import the database.

[0061] Execution of the computer program by the microprocessor may allow the database to be imported, for example by downloading it from remote computer hardware, or by copying it from computer hardware physically connected to the computing device, and saving it in a memory of the computing device.

[0062] Databases containing several sets of energy diagnostic data concerning different renovated buildings, or renovated parts of buildings, exist and are freely accessible on the internet. For example, in France, the French Environment and Energy Management Agency (ADEME), otherwise known as the Ecological Transition Agency, makes available to third parties a database containing more than 50,000 sets of energy diagnostic data for different renovated buildings, or renovated parts of buildings. The data in each data set is associated with several hundred variables. Other databases are freely accessible in other countries. These databases containing a Large amounts of data associated with a large number of variables are generally referred to as raw databases. Some databases contain, for each renovated building, or part of a building, a dataset relating to a simulation before renovation and a dataset relating to a simulation after renovation. These databases can be modified so as to combine, for each renovated building, or part of a building, the dataset relating to a simulation before renovation and the dataset relating to a simulation after renovation. It is also possible to aggregate several raw databases in order to increase the number of potentially exploitable datasets.

[0063] The database, thus imported and saved, can be used as is as a basis for the generation of the mathematical substitution model of step a2). If the generation of the mathematical substitution model makes it possible to obtain satisfactory results as is, the precision and simplicity of the mathematical substitution model can be greatly improved by selecting only qualitative data from the database.

[0064] To do this, a selection can be made on the variables associated with the data. Thus step a1) can include the sub-step: • al2) Select a specific number of variables.

[0065] In practice, the execution of the computer program by the microprocessor can make it possible to select the variables. As described previously, these variables can be of the input variable type or of the output variable type.

[0066] The output variable(s) may be selected according to the desired result. In particular, and as explained previously, the output variable(s) may comprise an energy impact variable, or one or more energy impact variables, each associated with a type of energy renovation. When the desired variable is not associated with any data in the database, but can be obtained by calculation from one or more variables associated with data in the database, new data relating to the desired variable may be added to the database. Step a1) may then comprise the sub-steps: • alO 1) For each data set, calculate data corresponding to a determined variable based on one or more data from the data set relating to variables from which it is possible to calculate the determined variable, • al02) Save the calculated data in the database.

[0067] In practice, the execution of the computer program by the microprocessor can make it possible to calculate these data and to record the calculated data in the memory of the computer device.

[0068] Once the output variable(s) are defined, all other variables associated with data in the database, with the exception of the variables possibly used in step a10l), are usable input variables. The input variable(s) can be selected according to their impact on the output variable(s). To do this, step a13) can comprise the sub-steps: • a121) Measure, for each input variable, the impact on the variable(s) exit, • al22) Select the input variables for which the measured impact is below a threshold.

[0069] In practice, the execution of the computer program by the microprocessor can make it possible to measure the impact of the input variables and select the input variables for which the measured impact is lower than the threshold. For example, for an output variable representing an energy impact expressed in kilowatt-hours per year (kW.h / year), the threshold used can be - 1 kW.h / year. Also, for an output variable representing an energy impact expressed as a percentage (%), the threshold used can be - 0.1%. In alternative embodiments of the impact measurement, the latter can consist of calculating a correlation coefficient, for example a variance analysis coefficient of the Pr(>F) type. For such a coefficient Pr(>F), the threshold used can be 0.05.

[0070] Advantageously, the number of selected variables is less than or equal to 50. Preferably, the number of selected variables is less than or equal to 40.

[0071] Once the variables have been selected, it is possible to simplify the database in order to improve the speed of query execution. Step al3) can then include the sub-step: • al23) Delete the data associated with the variables from the database entry excluded from the selection.

[0072] In practice, the execution of the computer program by the microprocessor can make it possible to delete this data associated with the input variables excluded from the selection.

[0073] In an alternative embodiment, all of the data can be retained, but only the data associated with the selected variables will be used to generate the mathematical substitution model in step a2).

[0074] A selection can also be made on the data sets. Thus, step a1) can include the sub-step: • al3) Select a specific number of data sets.

[0075] In practice, the execution of the computer program by the microprocessor can make it possible to select the data sets. The data sets can, in particular, be selected according to their relevance for calculating the result research.

[0076] For example, a data set for which one or more of the data associated with the output variables are not provided is useless for generating the model and may be a source of inaccuracies. Thus, step a1) may include the sub-step: • al4) Delete the data sets for which at least one of the data associated with the output variable(s) is missing.

[0077] Also, a data set for which one or more of the data associated with the input variables are not provided may be a source of inaccuracies in the generation of the model. Thus, step a1) may comprise the sub-step: • al5) Delete the datasets for which at least one of the data associated with the input variable(s) is missing.

[0078] In an alternative embodiment of the selection mode, step a1) may comprise the sub-step: • al5') Delete the datasets for which a determined number of data associated with the input variable(s) is missing.

[0079] It is thus possible to retain more data, while deleting data sets for which a lot of data is missing. This number can be less than 10, advantageously less than 5, preferably less than 2.

[0080] In an alternative embodiment of the selection mode, step a1) may comprise the sub-step: • al5”) Delete data sets for which at least one of the data associated with one or more specific input variables is missing.

[0081] This variant allows to delete only the data sets for which missing data are associated with the input variables having the greatest impact on the output variable(s).

[0082] In practice, the execution of the computer program by the microprocessor can make it possible to delete the data sets at steps al5) or al5') or al5”).

[0083] Also, the data sets may be associated with an energy renovation combination, said combination possibly comprising only a single energy renovation. Some energy renovation combinations may be highly recurrent in the database while others may be weakly recurrent. It is therefore possible to eliminate the data sets associated with the weakly recurrent energy renovation combinations. Step a1) may comprise the sub-step: • a 16) Identify energy renovation combinations in the database, • al7) Measure the recurrence of each of the energy renovation combinations, • al8) Delete the data sets associated with energy renovation combinations whose recurrence is lower than a threshold.

[0084] In practice, the execution of the computer program by the microprocessor can make it possible to identify the energy renovation combinations, to measure their recurrences and to delete the data sets associated with the energy renovation combinations whose recurrence is lower than the threshold. For example, the threshold can be 10 recurrences.

[0085] By applying one or more of the steps, or sub-steps, al2), al21), al22), al23, al3), al4), al5), al5'), al5"), al6), al7), al8), or any other selections of variables or data sets suitable to those skilled in the art, it is possible to significantly reduce the size of the database so as to obtain a database whose data are more qualitative than the raw database. This is then referred to as an optimized database. For example, by applying several of these selections to the ADEME database, it is possible to obtain an optimized database comprising fewer than 5,000 data sets and a number of associated variables less than 50.

[0086] It is also possible to import in step a1 1) an already optimized database. Advantageously, the number of variables associated with the data of the optimized database is less than or equal to 50. Preferably, the number of variables is less than or equal to 40.

[0087] In practice, to implement step a2), the computer device may be configured to generate the substitution mathematical model. The execution of the computer program by the microprocessor may enable this generation. Computer subroutines enabling the generation of a substitution mathematical model exist and are freely accessible within the framework of the tool libraries of various programming languages. For example, the “sklearn.ensemble” module available in the tool library of the “Python” programming language enables the generation of a substitution mathematical model of the decision tree forest type. Such subroutines may be used as part of the computer program of the invention.

[0088] The set of data in the raw or optimized database used to generate the substitution mathematical model is called the “generational data set”. The generational data set may comprise all the data in said raw or optimized database. Advantageously, and in order to allow adjustment of the substitution mathematical model, the generational data set comprises only part of the data in the raw or optimized database. raw or optimized data. The substitution mathematical model may in particular be generated from at least 60% of the datasets in the raw or optimized database. Preferably, the substitution mathematical model may in particular be generated from at least 80% of the datasets in the raw or optimized database. In practice, the greater the number of datasets in the database, the smaller the portion of datasets selected from the database to form the generational dataset may be.

[0089] Thus, as represented in the exemplary embodiment of figure [Fig.2], step a) can then comprise a sub-step: • a3) Adjust the mathematical substitution model from a set of database data including data different from the data in the dataset used for generating the substitution mathematical model.

[0090] The set of data different from the data in the data set used for the generation of the substitution mathematical model, that is to say, the set of data different from the data in the generational data set is generally called the adjustment data set. The substitution mathematical model may in particular be adjusted from at least 10% of the data sets in the raw or optimized database. Preferably, the substitution mathematical model may in particular be adjusted from at least 20% of the data sets in the raw or optimized database. In practice, the adjustment may consist of verifying the accuracy of the generated substitution mathematical model, by testing the latter with the data sets in the adjustment data set.For example, when, for a data set, a difference is observed between the result obtained with the mathematical substitution model and the result already recorded in the database, and this difference is greater than a threshold, the mathematical substitution model can then be modified so as to reduce this difference. In the context of a cross-adjustment, steps a2) and a3) can be repeated several times by selecting a generational data set and a different adjustment data set each time. In practice, to implement step a3), the computing device can be configured to adjust the mathematical substitution model. The execution of the computer program by the microprocessor can make it possible to adjust the mathematical substitution model by implementing step a3) alone, or by repeating the implementation of steps a2) and a3) several times.The computer routines for generating a mathematical substitution model previously described generally allow adjustment of the mathematical substitution model.

[0091] In alternative embodiments of step a), it is possible to verify the per- formation of the mathematical substitution model before exploiting it. Thus, and as represented in the example of realization of the figure [Fig.2], step a) can also include a sub-step: • a4) Check the validity of the mathematical substitution model.

[0092] In practice, to implement step a4), the computing device may be configured to verify the validity of the substitution mathematical model. The execution of the computer program by the microprocessor may allow this verification. The validity of the mathematical model may in particular be verified by means of one or more performance indicators. To do this, step a4) may comprise the sub-steps: • a41) Calculate one or more performance indicators, • a42) Determine the validity of the mathematical substitution model in depending on the performance indicator(s).

[0093] In practice, to implement step a4), the computer device may be configured to calculate one or more performance indicators, and possibly, determine the validity of the substitution mathematical model as a function of the performance indicator(s). The execution of the computer program by the microprocessor may allow this calculation, and possibly this determination. The computer subroutines allowing the generation of a substitution mathematical model previously described may also allow the calculation of performance indicators.

[0094] For example, commonly used performance indicators for mathematical model verification are bias, root mean square error (RMSE), mean square error (MSE), weighted average absolute error (MAE), Nash-Sutcliffe criterion, standard deviation ratio (RSR), relative volume error (RVE) criterion, correlation coefficient between estimated and observed values. Other performance indicators may also be the percentage of values whose absolute error is less than 5%, or the percentage of values whose absolute error is less than 10%.

[0095] The validity of the surrogate mathematical model may be determined by comparing each of the performance indicators to a threshold value associated with it. The execution of the computer program by the microprocessor may allow this comparison. For example, a surrogate mathematical model may be considered valid when the bias is less than 0.001 in absolute value, and / or the root mean square error (RMSE) is less than 0.06, and / or the mean square error (MSE), the weighted average absolute error (MAE) is less than 0.0001, and / or the Nash-Sutcliffe criterion is greater than 0.8 and / or the standard deviation ratio (RSR) less than 0.4 and / or the relative volume error (RVE) criterion is less than 0.000001, and / or the correlation coefficient between the estimated values and the observed values is less than 1, and / or the percentage of values, the error in absolute value of which is less than 5%, is greater than 75, and / or the percentage of values, the error in absolute value of which is less than 10%, is greater than 90.

[0096] When the determination of the validity of the substitution mathematical model, based on the performance indicator(s), is not implemented by the computer program, the computing device may be configured to display the performance indicator(s) on a screen. The execution of the computer program by the microprocessor may allow the display of the performance indicator(s) on the screen. The determination of the validity of the substitution mathematical model is then left to the discretion of the user based on the information displayed on the screen.

[0097] When the validity of the substitution mathematical model is proven, the valid substitution mathematical model can then be used for the implementation of step b).

[0098] When the validity of the substitution mathematical model is invalidated, and as in the exemplary embodiment shown in figure [Fig.2], the raw database can be re-modified by selecting another set of variables in accordance with step a12); and / or another set of data sets in accordance with step a13), before generating a new substitution mathematical model in accordance with step a2), possibly adjusting the new substitution mathematical model in accordance with step a3) and verifying the validity of the new substitution mathematical model in accordance with step a4). This loop of steps can be reproduced until a valid substitution mathematical model is obtained.

[0099] In a second embodiment of providing the substitution mathematical model, step a) may comprise the sub-step: • al') Import an existing substitution mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, based on one or more input variables.

[0100] Execution of the computer program by the microprocessor may allow the substitution mathematical model to be imported, for example by downloading it from remote computer hardware, or by copying it from computer hardware connected to the computing device, and by saving it in a memory of the computing device.

[0101] Once the mathematical substitution model has been provided, it is possible to use it to carry out calculations relating to buildings, or parts of buildings, to be renovated.

[0102] The process of energy renovation of a building, or part of a building, then includes the step: • b) Acquire, for one or more of the input variables, data relating to a building, or part of a building, to be renovated.

[0103] In practice, the computer device may comprise a means for acquiring data relating to the building, or part of a building, to be renovated, entered on a user interface. The latter may comprise a computer and / or a screen, which may be touch-sensitive, and / or a keyboard, and / or a mouse and / or any other user interface suitable for those skilled in the art. The user interface is generally remote, but may, in certain embodiments, be directly connected to the computer device. The means for acquiring the data may, for example, be in the form of a network card and / or a computer bus and / or any other data acquisition means suitable for those skilled in the art. The execution of the computer program by the microprocessor may allow the acquisition of data relating to a building, or part of a building, to be renovated.

[0104] In order to facilitate the use of the method which is the subject of the invention, the number of input variables, for which data is acquired, may advantageously be less than or equal to 20, preferably less than or equal to 10. This is then referred to as a “simplified mode”. In particular, these variables may include: • A habitable surface area variable, and / or • A construction period variable, and / or • A diagnostic note variable, and / or • A geographic location variable, and / or • A heating type variable, and / or • A variable of heating age, and / or • For several types of energy renovation, a variable for selecting the type of energy renovation, and / or • For one or more of the said types of energy renovation, one or more energy renovation content variables, and / or • A variable energy consumption variable before energy renovation.

[0105] In an alternative embodiment, the number of input variables, for which data is acquired, may advantageously be greater than 80% of the input variables. This is then referred to as “expert mode”. In practice, the number of input variables, for which data is acquired, may be greater than or equal to 40. This alternative embodiment is particularly suitable for professionals, and in particular diagnosticians, architects, building professionals, or any other professional suitable for those skilled in the art.

[0106] In another embodiment, the user can have the choice between the "simplified mode" and "expert mode". The computing device may then include a means for selecting "simplified mode" or "expert mode" via the user interface. Execution of the computer program by the microprocessor may enable this selection.

[0107] The process of energy renovation of a building, or part of a building, includes the step: • c) Calculate, from the mathematical substitution model and the acquired data, the energy impact data.

[0108] In practice, the computing device may be configured to calculate said energy impact data. The execution of the computer program by the microprocessor may enable this calculation.

[0109] When, for one or more input variables, no data is acquired, the energy impact data can be calculated without taking these input variables into account.

[0110] In an alternative embodiment of the calculation, step c) may comprise the sub-steps: • cl) Determine, for the input variable(s) for which no data has been acquired, a default data item, • c2) Calculate, from the mathematical substitution model, the acquired data and the default data(s) determined, the energy impact data.

[0111] In particular, when no data associated with variables for selecting a type of energy renovation is acquired, the data associated with these variables can be entered by default so as to select one or more types of energy renovation and to deselect the remaining types of energy renovation. In practice, the computing device can be configured to determine the default data(s). The execution of the computer program by the microprocessor can determine the default data(s). The default data associated with an input variable can correspond to the average value associated with said variable in the database. The default data associated with an input variable can correspond to the most recurring value associated with said variable in the database.

[0112] The process of energy renovation of a building, or part of a building, then comprises the step: • d) Determine, based on the energy impact data, the relevance of the energy renovation(s) for the building, or part of the building, to be renovated.

[0113] In practice, the computing device may be configured to determine in depending on the energy impact data, the relevance of the energy renovation(s) for the building, or part of the building, to be renovated. The execution of the computer program by the microprocessor can determine this relevance. For example, the relevance of the energy renovation(s) can be considered proven when the energy impact data exceeds a threshold. Thus, when the energy impact variable is expressed as a percentage (%) as previously described, the relevance of the energy renovation(s) can, for example, be considered proven when the energy impact data is less than - 10%. When the energy renovation(s) are considered not relevant, it is always possible to redo the calculation for another combination of energy renovation, by repeating steps b), c) and d) as illustrated in the figure [Fig. 1].To do this, a new energy renovation combination must be selected in step b). The input data that had already been entered in the previous sequence and that have not been changed are retained.

[0114] When the mathematical substitution model includes different energy impact variables, each associated with a different energy renovation combination, the relevance of the energy renovation(s) can be considered proven only for the energy renovation combination with which the most qualitative energy impact data is associated. For example, when the energy impact variable is expressed as a percentage (%) as previously described, the most qualitative energy impact data is the data with the largest absolute value.

[0115] Automatic determination of relevance may be particularly suitable for professionals or local authorities seeking a rapid and objective means of determining the buildings for which energy renovations must be undertaken.

[0116] In embodiments particularly suited to individuals, the computing device may be configured to display the energy impact data on a screen. The execution of the computer program by the microprocessor may allow the energy impact data to be displayed on the screen. The determination of relevance is then left to the user's discretion based on the information displayed on the screen. When the mathematical substitution model comprises several output variables, the computing device may be configured to display, on the screen, one or more of the calculated data associated with the output variables. The execution of the computer program by the microprocessor may allow this display.Thus, when different energy impact variables correspond to different energy renovation combinations, the user can make a comparison between the said combinations in order to choose. the most relevant. The computing device may be configured to display other data on the screen. In particular, when the energy impact variable represents an estimate of the energy consumption of the building, or part of the building, after carrying out the energy renovation(s), the computing device may be configured to display data relating to the energy consumption of the building, or part of the building, before energy renovation. The user can thus compare the values before and after energy renovation. The data relating to the energy consumption of the building, or part of the building, before energy renovation, may be acquired data associated with an input variable, i.e. the energy consumption variable before energy renovation.Alternatively, the data relating to the energy consumption of the building, or part of the building, before energy renovation, can be calculated using the substitution mathematical model. The execution of the computer program by the microprocessor can allow this display and / or calculation.

[0117] The process of energy renovation of a building, or part of a building, also includes the step: • e) If the relevance is proven, carry out the energy renovation(s) of the building, or part of the building, to be renovated.

[0118] Energy renovations are implemented using standard renovation methods. For example, in the case of a renovation of the heating system, or the air conditioning system, or the heating system control device, the renovation may include the removal of the old system or device, and the installation of a new system or device. In the case of a renovation of the domestic hot water production system, the renovation may include the removal of the old system and the installation of a new system, or the renovation of the insulation of the old system. The renovation of the ventilation system may include the removal of the old system and the installation of a new system, including the installation of new ventilation ducts and the installation of new ventilation devices.Renovating wall insulation may include installing insulation on the external walls of the building's walls, and / or removing the old layer of insulation on the internal walls and installing a new layer of insulation on said internal wall. Renovating roof, floor, or ceiling insulation may include removing the old layer of insulation, where it exists, and installing a new layer of insulation. Renovating openings may include removing one or more windows and / or doors and installing new windows and / or doors. Installing or renovating an electricity generation system may include removing the old electricity generation system, if existing, and installing a new electricity generation system, for example, installing . photovoltaic panels.

[0119] In alternative embodiments, the energy renovation method may be dedicated to a particular type of building, or part of a building. For example, the energy renovation method may be dedicated to a building, or part of a building, for residential use. In this case, the substitution mathematical model is configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, for residential use, to be renovated, as a function of one or more input variables. Therefore, to generate such a substitution mathematical model, the database preferably comprises energy diagnostic data sets relating to different renovated residential buildings, or different renovated parts of residential buildings.In other examples of implementation, the energy renovation process is dedicated to a building, or part of a building, used as an office, or to a building, or part of a building, of the bakery type, or to a building, or part of a building, of the industrial warehouse type.

[0120] The invention also relates to a computer program for implementing an energy renovation method according to the invention. Several characteristics of this program have previously been described throughout the description. This computer program is particularly remarkable in that, when executed by a microprocessor, said computer program is configured to implement the steps: • a) Provide a substitute mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, based on one or more input variables, • b) Acquire, for one or more of the input variables, data relating to the building, or part of the building, to be renovated, • c) Calculate, from the mathematical substitution model and the acquired data, the energy impact data.

[0121] The computer program may also, when executed by a microprocessor, be configured to implement one or more of the steps, or sub-steps, al), alOl), al02), ail), al2), al21), al22), al23), al3), al4), al5), al5'), al5”), al6), al7), al8), a2), a3), a4), a41), a42), al'), cl), c2), d).

[0122] The invention also relates to a computer program for providing a mathematical model for substituting an energy renovation method according to the invention. This providing computer program is remarkable in that, when executed by a microprocessor, said providing computer program is configured to implement the steps: • al) Provide a database comprising several sets of energy diagnostic data of different renovated buildings, or different renovated parts of buildings, each data set being associated with a different renovated building, or a different renovated part of building, each data set being associated with a variable, said variables comprising an energy impact variable associated with one or more energy renovations of a building, or part of building, and several other variables, • a2) Generate a mathematical substitution model from the data of the database by selecting the energy impact variable as output variable, and the other variables as input variables.

[0123] The providing computer program may also, when executed by a microprocessor, be configured to implement one or more of the substeps, al), alOl), al02), ail), al2), al21), al22), al23), al3), al4), al5), al5'), al5”), al6), al7), al8).

[0124] The providing computer program may also, when executed by a microprocessor, be configured to implement the step: • a3) Adjust the mathematical substitution model from a set of database data including data different from the data in the dataset used for generating the substitution mathematical model.

[0125] The providing computer program may also, when executed by a microprocessor, be configured to implement the step: • a4) Check the validity of the mathematical substitution model.

[0126] The supply computer program may also, when executed by a microprocessor, be configured to implement one or more of the sub-steps a41), a42).

Claims

Claims

1. Method for energy renovation of a building, or part of a building, characterized in that it comprises the steps: • a) Providing a substitution mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, as a function of one or more input variables, • b) Acquiring, for one or more of the input variables, data relating to the building, or part of a building, to be renovated, • c) Calculating, from the substitution mathematical model and the acquired data, the energy impact data, • d) Determining, as a function of the energy impact data, the relevance of the energy renovation(s) for the building, or part of a building, to be renovated, • e) If the relevance is proven, carrying out the energy renovation(s) of the building, or part of a building, to be renovated.

2. Energy renovation method according to claim 1 characterized in that step a) comprises the sub-steps: • al) Providing a database comprising several sets of energy diagnostic data of different renovated buildings, or different renovated building parts, each data set being associated with a different renovated building, or with a different renovated building part, each data set data being associated with a variable, said variables comprising an energy impact variable associated with one or more energy renovations of a building, or building part, and several other variables, • a2) Generating a substitution mathematical model from a set of data from the database by selecting as output variable, the energy impact variable, and as input variables, the other variables.

3. Energy renovation method according to claim 2 characterized in that step a) comprises the sub-step: • a3) Adjusting the substitution mathematical model from a set of data from the database comprising data different from the data from the set of data used for the generation of the substitution mathematical model.

4. Energy renovation method according to one of claims 2 or 3, characterized in that the number of variables is less than or equal to 50.

5. Energy renovation method according to one of the preceding claims, characterized in that the number of input variables for which data is acquired is less than or equal to 10.

6. Energy renovation method according to one of the preceding claims, characterized in that the input variables comprise a variable representing a habitable surface area of the building, or part of the building, to be renovated, and / or a variable representing a construction period of the building, or part of the building, to be renovated, and / or a variable representing an energy performance diagnostic score of the building, or part of the building, to be renovated, before energy renovation, and / or a variable representing a geographical location of the building, or part of the building, to be renovated, and / or a variable representing a type of heating used in the building, or part of the building, to be renovated, before energy renovation, and / or a variable representing the age of the heating system used in the building, or part of the building, to be renovated, before energy renovation, and / or, for each type of energy renovation,a variable representing a selection of the type of energy renovation as energy renovation of the building, or part of the building, to be carried out, and / or, for each type of energy renovation, one or more variables representing the content of the energy renovation of the building, or part of the building, to be carried out, and / or a variable representing an energy consumption of the building, or part of the building, to be renovated, before energy renovation.,

7. Energy renovation method according to one of the preceding claims, characterized in that the mathematical substitution model is a model of the decision tree forest type.

8. Energy renovation method according to one of the preceding claims, characterized in that step c) comprises the sub-steps: • cl) Determine, for the input variable(s) for which no data has been acquired, a default data item, • c2) Calculate, from the mathematical substitution model, of the acquired data and the determined default data, the energy impact data.

9. Computer program for implementing an energy renovation method according to one of claims 1 to 8, characterized in that, when executed by a microprocessor, said computer program is configured to implement the steps: • a) Provide a substitute mathematical model configured to calculate energy impact data associated with one or more energy renovations of a building, or part of a building, to be renovated, based on one or more input variables, • b) Acquire, for one or more of the input variables, data relating to the building, or part of the building, to be renovated, • c) Calculate, from the mathematical substitution model and the acquired data, the energy impact data.

10. Computer program for providing a mathematical model for substituting an energy renovation process according to one of claims 1 to 8, characterized in that, when executed by a microprocessor, said computer program is configured to implement the steps: • al) Provide a database comprising several sets of energy diagnostic data of different renovated buildings, or different renovated parts of buildings, each data set being associated with a different renovated building, or a different renovated part of building, each data set being associated with a variable, said variables comprising an energy impact variable associated with one or more energy renovations of a building, or part of building, and several other variables, a2) Generate a mathematical substitution model from the database data by selecting the energy impact variable as the output variable and the other variables as the input variables.

Images