Method, device and system for determining emissions of a territory

Abstract

The present invention relates to a method, device and system for determining emissions of a territory. To this end, computing means receive first data from measurements of concentrations of emissions associated with the territory, each of the measurements being associated with a location in the territory, obtain second data representative of maps of atmospheric flows associated with the territory, and determine third data representative of maps of emission sources (5) in the territory, on the basis of the first data and second data.

Description

[0001] Description

[0002] Title: Method, device and system for determining emissions from a territory

[0003] technical field

[0004] The present invention relates to the monitoring, quantification and identification of emission sources.

[0005] The present invention relates more particularly to the monitoring of greenhouse gas emissions in a territory.

[0006] The present invention relates more particularly to a method for determining emissions from a territory.

[0007] The present invention also relates to a device implementing such a method, as well as a system incorporating such a device.

[0008] The invention will therefore find many advantageous applications in the fight against climate change and in the search for opportunities to reduce emissions.

[0009] Previous art

[0010] In the fight against climate change, the most common and frequently monitored greenhouse gas is carbon dioxide, or CO2. Currently, several methods exist for measuring CO2. These include the use of ground-based sensors installed at specific locations: carbon dioxide sensors, infrared gas analyzers, and atmospheric monitoring stations. These systems allow for the evaluation of CO2 levels at a precise location in ppm (parts per million), or the mass concentration of CO2 in the air (for example, in mg / kg).

[0011] Other methods can also be applied, such as satellite remote sensing systems or computer modeling. These systems have the advantage of being able to monitor much larger areas of CO2 emissions. However, in the case of remote sensing, resolution problems can arise, preventing the measurement of emission sources within a city. Computer simulations incorporate estimates (not geolocated and precisely dated) rather than measurements, which limits their actual accuracy.

[0012] Of course, it is also possible to use such technologies in a complementary way to obtain a more complete picture of CO2 emissions in an urban area. The use of these tools thus makes it possible to monitor and assess the impact of human activities on the environment, and to take measures to reduce CO2 emissions and improve air quality.

[0013] Various studies have been conducted to track CO2 emissions in large cities. These studies are carried out over a period of one to two years and rely on a single sensor positioned at a height of between 25 and 190 meters. This allows for the measurement of various events impacting CO2 emissions within a city, without distinguishing their location or origin. In particular, these studies cannot differentiate between CO2 emissions originating outside the city and affecting the measurements.

[0014] It is also known to study a city's CO2 emissions using inverse Bayesian estimation, over a period of approximately one month at a 6-hour resolution, based on data collected from six different sites. Such a study limits the ability to quantify the extent of the bias introduced by the interaction between the chosen network of sites and the true emission field.

[0015] Another known solution uses a Lagrangian inversion system to quantify a region's emissions over a season, based on observations of CO2 and CO (carbon monoxide) emitted from a network of six ground stations. However, this solution remains imprecise and lacks information about the quantified emissions.

[0016] Thus, the various known works of the prior art are limited both by the number of sensors used, the periodicity of measurements, and the total time of the study.

[0017] In particular, none of these solutions appears to adequately account for the dispersion of emissions over time, that is, the movement of these emissions between their source and their measurement point. This limitation greatly reduces the usefulness of these studies, preventing the clear identification, location, and quantification of emission sources.

[0018] The Applicant therefore submits that there is currently no satisfactory alternative emissions monitoring solution capable of providing accurate quantification of emissions at their source level.

[0019] Summary of the invention

[0020] The present invention aims to improve the current situation described above.

[0021] The present invention aims more particularly to overcome the above drawbacks by providing a method, a device, and a system for determining emissions from a territory that takes into account the mechanisms of dispersion and advection of these emissions over time. To this end, the object of the present invention relates, in a first aspect, to a method for determining emissions from a territory, the method being implemented by at least one processor, the method comprising the following steps:

[0022] - receipt of initial data on emission concentration measurements associated with the territory, each measurement being associated with a location on the territory;

[0023] - obtaining second representative data sets of atmospheric flow maps associated with the territory; and

[0024] - determination of third data representative of emission source maps on the territory, from the first and second data.

[0025] Preferably, the emissions correspond to carbon dioxide emissions.

[0026] It is understood here that the initial data allows for the measurement of emission levels (particularly CO2) at multiple locations. This initial data can be received through communication with fixed sensors placed at specific locations or via a network of mobile sensors, with each location corresponding to the position of a sensor. The initial data can also include satellite measurements, which correspond to a "point cloud," with each point representing a discrete location. It is also understood that the second and third data sets can include multiple maps. For example, several flow maps are planned, depending on meteorological conditions, as described below.Similarly, several emission source maps are planned, each associated with a set of measurements at a given time, or, as described below, a source map associated with each location, the different maps then being combined to determine an overall profile associated with the territory.

[0027] The process includes, for example, the use of and / or communication with a Geographic Information System (GIS) to enable and facilitate the processing of data directly related to their location within the territory. The scale used, that is, the granularity of the data, is selected appropriately for processing atmospheric flows and the distribution of sources. In particular, the territory is preferably on the scale of a city, that is, an area on the order of several kilometers or tens of kilometers on each side.

[0028] For the purposes of this invention, "flow" refers to the impact of wind, weather phenomena, or terrain on the movement of emissions across a given area. This flow results in a displacement of emissions between their source, i.e., their origin, and the location where they are measured. The method according to this invention aims to reverse the impact of such flow, thereby tracing emissions back to their source through localized measurements.

[0029] In particular, emission sources are quantified; that is, emission measurements, which normally correspond to a concentration (in ppm or mg / kg), are converted into a volume or quantity of emissions associated with a specific area of ​​the territory, the area corresponding to the source of that quantity of emissions. Source mapping includes, for example, emission volumes per m³ 2of the territory, that is to say, a quantity of emissions relative to the surface area of ​​the associated source. This allows us to transform a concentration measured at a given location into a quantity and location of emissions produced.

[0030] A person skilled in the art also understands that, within the framework of source mapping—that is, the determination of third-party data associated with emission sources rather than their mere presence—it becomes possible to eliminate emissions originating outside the territory itself, or to distinguish them as external sources. For example, one can thus separate CO2 emitted at different points within the territory from CO2 originating from neighboring areas at varying distances and transported via runoff.

[0031] Preferably, the determination of the third data, i.e. the inversion between the measured concentration and the volume produced, is carried out via a non-Bayesian renormalization and inversion algorithm.

[0032] A person skilled in the art understands that a Bayesian model is a probabilistic model that treats parameters independently. A Bayesian model requires a very large amount of input data to achieve an acceptable level of certainty. In contrast, a non-Bayesian, or frequency-domain, model according to the invention requires less input data to achieve the same level of certainty. However, the computing power required by a non-Bayesian model is greater, and the present invention also aims to minimize this computing power as described below.

[0033] Thanks to the present invention, it is therefore possible, starting from conventional measurements of emission concentrations, to take into account flows across the territory in order to identify emission sources, as well as to distinguish emissions external to the territory. Emission source maps are thus more representative of reality and usable, particularly for the purpose of reducing emissions.

[0034] It is also possible to integrate this process into a broader process, particularly one that includes further processing to analyze third-party data. This process could thus include determining the total volume of emissions associated with the territory, identifying high-contributing areas, predicting future emissions, or identifying emission reduction solutions. The process could also include rendering visual content representative of the third-party data, notably through communication with a human-machine interface for displaying the visual content.

[0035] In an advantageous embodiment of the present invention, the method further comprises receiving a set of parameters including:

[0036] - information representative of land use patterns in the area; and / or

[0037] - information representative of habitat characteristics in the area; and / or

[0038] - information representative of activity associated with the territory; and / or

[0039] - information representative of a road transport network associated with the territory, the process further including a step of selecting a set of locations on the territory at least according to the set of parameters, the first data being received with respect to the set of locations.

[0040] The characteristics of the area's housing include, for example, its density and / or type of housing. The activity associated with the area includes, for example, the proportion of tertiary and / or industrial activity. Information representing the road transport network includes, for example, information on its structure, geometry, and the type of roads (motorways, expressways, ring roads, etc.).

[0041] The selection of the location set corresponds, for example, to the provision of sensors according to the selected locations, to the control of a network of mobile sensors to match them to the locations, or to the selection of data from more comprehensive satellite data. It is understood here that the selection of the location set corresponds to determining a representative set of locations in relation to the territory and its composition, so as to obtain a relevant number of measurements without multiplying unnecessary calculations.

[0042] For example, for an area with sides of 12 to 15 km, 75 locations are planned, corresponding to, for example, 75 sensors, with the locations spaced between 1 and 2.7 km apart. The Applicant submits in particular that, for the same area, it would be possible to limit this to 35 distinct locations.

[0043] In an additional embodiment, the process further comprises the following steps:

[0044] - receipt of fourth data representing a mapping of the territory;

[0045] - determination of a plurality of zones of the territory according to the fourth data, each zone of the plurality of zones being associated with a distinct emission potential, the process further comprising a step of selecting a set of locations on the territory at least according to the plurality of zones, the first data being received with respect to the set of locations.

[0046] The fourth set of data includes, for example, topographic and / or satellite data. This fourth set of data is received, for example, through communication with a database server such as the IGN (French National Geographic Institute) and / or Copernicus. A preliminary mapping step of the territory is typically planned to obtain this fourth set of data.

[0047] Determining the plurality of zones within the territory involves, for example, classifying the fourth set of data using a homogeneous polygon class algorithm and then defining a plurality of zones based on the classified data. For instance, the zones are planned to be grouped into seven categories according to their emission potential. Emission potential refers to a zone's propensity to generate more or less significant emissions, taking into account the activities assumed to be associated with that zone, without directly measuring these emissions. This step allows, for example, the identification and differentiation of the various types of roads most likely to generate emissions. The data are then aggregated by zone after classification, in order to establish a limited number of distinct profiles, each associated with an emission potential.

[0048] The zones allow, for example, obtaining one or more of the above parameters; the classified data allows, for example, distinguishing different land uses, including a type of road or associated building.

[0049] It is understood here that this design makes it possible to determine key points for the selection of locations, for example for the placement of sensors, in particular in order to carry out measurements as close as possible to potential emission sources, or in order to distribute the measurements between the different determined areas.

[0050] In an additional embodiment, the first data is received at intervals during a given time period.

[0051] It is clear here that this design allows for the inclusion of a temporal element in the execution of the process and in the monitoring of emissions. For example, emissions are monitored in real time, with updates at each new time interval.

[0052] It is also possible to perform further processing on data received over a longer period, such as calculating averages, cumulative totals over specific periods, and so on. The longer the time period, the more additional analyses can be performed, allowing for the identification of hourly, daily, seasonal, and other fluctuations. Ideally, a time period of at least one year allows for the consideration of seasonal variations. Similarly, selecting a sufficiently short interval for receiving the first data allows for the detection of fluctuations over a shorter period. For example, receiving the first data every 10 minutes can be done to detect hourly fluctuations.

[0053] The Applicant submits in particular that monitoring third-party data month by month, with a time period of at least one year, provides an appropriate temporal resolution.

[0054] Preferably, the method further includes a selection, within the time period, of a subset of emission concentration measurements based on the first data, the subset having concentrations above a reference value, the third data being obtained from the subset of measurements.

[0055] It is understood here that this design limits calculations by determining the third data points only periodically. The reference value corresponds, for example, to an average value of the measurements, or to a minimum measurement percentile for each location. In other words, the first data points are selected for which the measurements at a significant number of locations are situated in the upper percentiles of their statistics. This design therefore makes it possible to distinguish between background and measurement noise measurements, and to concentrate the execution on the periods when emissions are highest.

[0056] In a particular embodiment, the process further includes obtaining representative wind information over the territory, the second data being obtained at least as a function of the representative wind information.

[0057] Wind representation information includes, for example, wind direction and / or intensity over the territory, for example a wind rose associating wind direction and intensity for the entire territory.

[0058] Wind information, for example, is derived from numerical meteorological data, such as satellite data. The process might involve receiving meteorological data over a one-hour resolution window, with the wind information being derived from this meteorological data. In a specific design, the initial data includes, alongside measurements, meteorological data associated with each location.

[0059] It is understood here that flow maps are determined at least partially based on wind characteristics, the characteristics useful for determining flow maps corresponding in particular to speed, direction, intensity, and vertical profile. For example, a wind map of the territory is obtained, that is to say, detailed wind fields across the territory, combining the aforementioned characteristics.

[0060] Determining the flow mapping therefore corresponds to determining the advection and diffusion mechanisms, depending on the wind intensity and direction. As stated previously, the flows correspond to the movements of emissions; winds cause such movements but are not identical.

[0061] In an embodiment that can be combined with the previous embodiment, the method further includes obtaining representative information of neighborhoods associated with the territory, each neighborhood being associated with a level of roughness with respect to atmospheric flows, the second data being obtained at least as a function of the representative neighborhood information.

[0062] It is understood here that neighborhoods allow us to model the effects of building friction on water flow. Representative information for neighborhoods is, for example, determined from satellite data of the territory and / or is part of the set of parameters described above. Satellite data includes, for example, representative information for urban blocks, from which neighborhoods can be defined.

[0063] It is understood here that such a design allows for consideration of the influence of buildings on atmospheric flows, while requiring less computing power than a precise model of the territory, i.e., the entire set of buildings. The Applicant submits in particular that a more precise model does not improve the accuracy of the flows, and therefore unnecessarily increases the computational burden.

[0064] In yet another embodiment, the method further comprises obtaining a set of reference meteorological situations, the second data being further determined according to each meteorological situation in the set, the method further comprises receiving information representative of instantaneous meteorological situations, the third data being further determined according to the instantaneous meteorological situations.

[0065] In other words, the process includes the determination of a plurality of flow maps, each flow map being associated with a reference meteorological situation and corresponding to the atmospheric flows resulting from that meteorological situation.

[0066] Reference weather situations, for example, correspond to a set of meteorological data covering the most common situations associated with the territory, with the aim of grouping calculations and limiting computing power requirements. Reference and instantaneous weather situations include, for example, representative wind information as defined above, as well as data on the time of day, season, etc. The calculation of reference weather situations is carried out, for example, based on preferred wind directions over the territory for establishing these situations; that is, in such a way that the majority of reference weather situations include wind behavior corresponding to the most common conditions.Alternatively, a portion of the reference scenarios may include less common wind directions to accurately represent all possible situations. Other parameters used to distinguish weather conditions include time of day, differentiating between daytime and nighttime periods, seasons, particularly winter, when heating-related emissions are highest, specific times associated with peak activity and / or traffic, and so on.

[0067] It is understood that, in contrast, instantaneous weather situations correspond, for example, to actual hourly weather conditions, such as those included in or obtained alongside the initial data. Determining these third data points involves, for example, assigning the instantaneous weather situation to the nearest reference weather situation, that is, selecting the appropriate flow map.

[0068] In other words, the second set of data comprises several atmospheric flow maps, each dependent on the meteorological situation. Determining the third set of data involves selecting the appropriate reference meteorological situation based on the current weather conditions. In one particular example, the process involves determining a source map for each location and reference meteorological situation; the received measurements then allow for the selection of appropriate source maps and their quantification.It is thus understood that the establishment of these reference situations makes it possible, for a large number of measurements over time, to limit the total number of calculations by determining in advance the typical flow maps, and by extension the sources associated with each type of measurement, obtaining a source map over the whole territory then corresponding to a combination and weighting of pre-established source maps.

[0069] In one embodiment, the process further includes obtaining a first mesh representative of the territory, with the second and / or third data points being obtained at least partially via the first mesh. It is understood here that the first mesh is, for example, determined from fourth data points as described above. The process uses a modeling tool to generate the first mesh, or communicates with a separate modeling tool. For example, the first mesh can be generated using a Fluidyn-PANACHE type modeling tool.

[0070] The use of a mesh thus corresponds to a discretization of the calculations, in order to simulate the flows for the determination of the second data and / or in order to trace the flows for the determination of the third data.

[0071] Preferably, the process also includes obtaining information representative of potential sources of emissions in the territory, the first grid corresponding to an irregular grid determined according to the potential sources.

[0072] In other words, the process here involves determining the initial grid, so as to establish it in a way that is suitable for subsequent processing. Potential emission sources correspond to elements of the territory that are a priori likely to generate significant emissions, for example, major roads. Thus, in a specific example, the process includes receiving mapping data of the territory, determining the territory's road axes and / or road traffic density from the mapping data, and then determining potential emission sources within the territory.

[0073] It is understood here that the first mesh is irregular, resulting in smaller, and therefore more precise, cells at the potential sources. This design allows for higher-resolution calculations to create flow and / or emission source maps, thereby improving the accuracy of the second and / or third data points without increasing the computational load in these areas. Simultaneously, the larger cells outside these potential sources reduce the overall computing power required for the entire mesh. This achieves a balance between the accuracy of the results and the constraints on the calculations.

[0074] In an additional embodiment, obtaining the second dataset includes an initial simulation, using the first mesh, of atmospheric flows over the area. As stated previously, this initial simulation corresponds to a discretized calculation of atmospheric flows with respect to the first mesh. The calculations are based, for example, on conditions at predefined points of the first mesh, such as those obtained from the data listed above, including wind measurements or other meteorological conditions. The simulation is based, for example, on solving the Navier-Stokes equations of fluid mechanics. A well-known method for solving these equations is the Reynolds-averaged Navier-Stokes (RANS) method. Specifically, the atmospheric flows here correspond to low-level flows, for example, in an area between 0 and 300 meters in altitude.For example, a k-epsilon (or ks) turbulence model is used.

[0075] According to a particular example, obtaining the second set of data includes, from initial conditions, calculating detailed wind fields over the territory, via the first mesh, and then calculating atmospheric dispersion matrices of emissions from the wind fields, i.e. a matrix representing the propagation of emissions relative to their sources.

[0076] In an additional embodiment, the determination of the third data includes a second simulation, via the first mesh, of adjoint fields to the locations of the measurements as a function of the second data, the third data being determined as a function of the adjoint fields and the first data.

[0077] By "adjoined fields," we mean here a link between a measurement performed on a location and a source map, which is a function of the measurement. In other words, determining the third data points involves, for each location, determining a source map based on the second data point, and then determining the third data points based on the source map and the first data point.

[0078] Determining adjoint fields is advantageously based on inverting the equations associated with diffusion and convection to trace the wind direction. This is best achieved by solving an adjoint ADE equation (Advection-Dispersion Equation), where advection corresponds to the transport of an element by the movement of the surrounding medium—in this case, the transport of emissions by atmospheric flows. The adjoint ADE equation is thus the inverse of a standard ADE equation, tracing the transport phenomenon back to determine the source of the measured emissions.

[0079] It is understood here that the determination of the adjoint fields results in obtaining a database which allows the sources of emissions to be reconstructed based on the measurements at each location.

[0080] In particular, the various preceding embodiments can be combined. In a preferred embodiment, starting from the initial mesh, the process simulates atmospheric flows under a plurality of reference meteorological conditions. From these atmospheric flows, the process then simulates advection mechanisms to obtain a source map associated with each location and each reference meteorological condition, thus generating a usable database corresponding to a set of source maps. Subsequently, based on the measurements obtained and the instantaneous meteorological conditions, the process determines a final source map from the database, corresponding to a weighted combination of the source maps in the database.

[0081] In one specific embodiment, the process further includes generating a second regular grid of the territory and assigning the third data to the second grid.

[0082] It is understood here that assigning third data to a second regular grid corresponds to an aggregation of irregular third data, particularly data associated with a first irregular grid, in order to facilitate their subsequent use. Assigning third data to the second grid corresponds, for example, to the transition from a three-dimensional grid, closer to a territorial model, to a two-dimensional grid, specifically a hexagonal grid based on a geospatial indexing system such as H3. Assigning third data to the second grid thus makes it easier to exploit this third data, particularly by describing precise characteristics of each cell according to the intended use.

[0083] Preferably, the process also includes the following steps:

[0084] - obtaining fifth data points representative of a set of business sectors associated with the cells of the second grid; and

[0085] - Assignment of an emission level to each sector of activity based on the aforementioned fifth and third data points.

[0086] The fifth data point is obtained, for example, from satellite data, or from the fourth data point described above. The fifth data point includes, for example, the proportion of each grid cell occupied by each sector of activity, i.e., the land occupancy rate of the sector of activity in question.

[0087] It is understood here that sectors can be defined in a variety of ways according to the needs of the person skilled in the art and satellite data, and correspond to different activities likely to generate emissions, for example road traffic, construction, generic or specific industrial activities, etc.

[0088] Thus, since each cell in the second grid is associated with an emission level derived from the third set of data, the fifth set of data allows this emission level to be distributed among the different sectors of activity, and by extension, an emission level to be determined for each sector of activity. This design therefore allows for a more precise quantification of emission sources.

[0089] According to a second aspect, the present invention relates to a device for determining emissions from a territory, the device comprising a memory associated with at least one processor configured for implementing the steps of the process according to the first aspect of the present invention.

[0090] According to a third aspect, the present invention relates to a computer program which includes instructions adapted for carrying out the steps of the process according to the first aspect of the present invention, in particular when the computer program is executed by at least one processor.

[0091] According to a fourth aspect, the present invention relates to a computer-readable recording medium on which is recorded a computer program comprising instructions for carrying out the steps of the process according to the first aspect of the invention.

[0092] On the one hand, the recording medium can be any entity or device capable of storing the program. For example, the medium can include a storage means, such as a ROM, a CD-ROM or a microelectronic circuit-type ROM, or a magnetic recording means or a hard drive.

[0093] On the other hand, this recording medium can also be a transmissible medium such as an electrical or optical signal, such a signal being able to be transmitted via an electrical or optical cable, by conventional or radio frequency, by self-directing laser beam, or by other means. The computer program according to the present invention can, in particular, be downloaded from a network such as the Internet.

[0094] Alternatively, the recording medium may be an integrated circuit in which the computer program is incorporated, the integrated circuit being adapted to execute or to be used in the execution of the process in question.

[0095] According to a fifth aspect, the present invention relates to a system for determining emissions from a territory, the system comprising:

[0096] - a plurality of sensors associated with the territory, each sensor being associated with a location within the territory; and

[0097] - an emissions determination device according to the second aspect of the present invention, the device being configured to communicate with a plurality of sensors so as to receive the initial data, each measurement of the initial data being associated with one of the sensors. It is also understood that, according to the design of the device, it is configured to receive additional data, in particular data relating to the mapping of the territory and / or the meteorological conditions of the territory, from the sensors or from another remote device, for example from a satellite database.

[0098] Thus, through the various functional and structural technical characteristics above, the Applicant proposes a process, a device and a system for determining emissions from a territory capable of taking into account the mechanisms of dispersion and advection in order to determine, from local concentration measurements, the actual sources of emissions in a territory.

[0099] Description of the figures

[0100] Other features and advantages of the present invention will become apparent from the description of the specific and non-limiting embodiments of the present invention below, with reference to the attached Figures 1 to 9, in which:

[0101] [Fig l]

[0102] Figure 1 illustrates a map of a territory on which emissions are likely to be measured, according to a particular and non-limiting embodiment of the present invention;

[0103] [Fig-2]

[0104] Figure 2 schematically illustrates a device configured to determine emissions from the territory of Figure 1;

[0105] [Fig.3]

[0106] Figure 3 schematically illustrates a flowchart of the different stages of a process for determining emissions from the territory of Figure 1;

[0107] [Fig.4]

[0108] Figure 4 illustrates a first representative graph of measurements of emission concentrations and winds over the territory of Figure 1;

[0109] [Fig. 5]

[0110] Figure 5 schematically illustrates a map of emission sources in the area of ​​Figure 1;

[0111] [Fig.6]

[0112] Figure 6 illustrates a second representative graph of winds over the territory of Figure 1; [Fig-7]

[0113] Figure 7 illustrates a third representative graph of neighborhoods associated with the territory of Figure 1; [Fig. 8]

[0114] Figure 8 schematically illustrates a first and second mesh of the territory shown in Figure 1; and

[0115] [Fig.9]

[0116] Figure 9 schematically illustrates a map of wind fields over the territory of Figure 1.

[0117] Detailed description

[0118] A method, a device, and a system for determining emissions will now be described in what follows, with joint reference to Figures 1 through 9. The same elements are identified with the same reference symbols throughout the description. As indicated in the preamble to this description, current solutions for determining emissions in a given area are largely imprecise and generally rely on passive monitoring of measurements from sensors scattered across the territory, without any further analysis of the origin of these emissions.

[0119] One of the objectives of the present invention is to propose a determination of emissions which takes into account the movement of these emissions over time, in particular under the effect of winds and other atmospheric phenomena.

[0120] This is made possible in the example described below.

[0121] As illustrated in Figure 1, the present example considers the use of the method according to the invention in the study of emissions from a city, which corresponds to territory 1.

[0122] It will be understood here that this example is not limiting and that the method according to the invention can be applied to the study of a wide variety of territories, at scales of varying sizes. It will also be understood that, since the method takes atmospheric phenomena into account, its appropriate application is to a territory 1 of sufficiently large surface area to account for such phenomena.

[0123] Furthermore, emissions preferably correspond to CO2 emissions, particularly in the context of carbon accounting or greenhouse gas reduction. However, it is understood that such a process also applies to a variety of other emissions for which monitoring may prove useful to those skilled in the art.

[0124] As stated previously, computer means are planned configured for the implementation of a method for determining emissions from territory 1, for example method 3 in Figure 3. Such a method 3 may aim to identify the main sources of emissions from territory 1, to quantify the actual amount of emissions produced in territory 1, or for a variety of further processing.

[0125] As illustrated in Figure 2, the computing resources configured for implementing the process are advantageously grouped together in an electronic device 2, for example, a computer (hereinafter referred to as the "computer"). The computer 2 is configured, for example, to transmit and receive data within a communication network. The elements of the computer 2, individually or in combination, can be integrated into a single integrated circuit, into several integrated circuits, and / or into discrete components. The computer 2 can be implemented as electronic circuits, software (or computer) modules, or a combination of electronic circuits and software modules.

[0126] The computer 2 comprises one (or more) processor(s) 21 configured to execute instructions for carrying out the steps of the process 3 and / or for executing instructions from the software embedded in the computer 2. The processor 21 may include integrated memory, an input / output interface, and various circuits known to those skilled in the art. The computer 2 further comprises at least one memory 20, corresponding, for example, to volatile and / or non-volatile memory, and / or includes a memory storage device that may comprise volatile and / or non-volatile memory, such as EEPROM, ROM, PROM, RAM, DRAM, SRAM, flash, magnetic disk, or optical disk.

[0127] The computer code of the embedded software(s) including the instructions to be loaded and executed by the processor is for example stored on memory 20 of computer 2.

[0128] According to one embodiment, the calculator 2 is configured to implement a method for determining emissions from territory 1, which is part of a larger method comprising one or more steps for analyzing emissions from territory 1. In this larger method, the quantities calculated during the method according to the invention, in particular the third data points, are used as input data for subsequent analyses. The method 3 according to the invention can also, by way of another example, be carried out iteratively over time with respect to territory 1, for example by determining a set of flow maps and / or source maps 5 during initial iterations, which are recorded and reused in successive iterations based on input data.

[0129] In a first step 31 of the emissions determination process, the computer 2 receives initial measurement data 4 of emission concentrations associated with territory 1. In particular, each measurement 4 is associated with a location 11 within territory 1, as illustrated in Figure 1. The first graph in Figure 4 thus illustrates a plurality of measurements 4, each associated with a location 11. In particular, the first graph illustrates the evolution of the CO2 concentration in ppm 42 as a function of time 41.

[0130] The first data is for example received from a set of 200 sensors associated with territory 1, so that each of the 200 sensors is associated with a location 11 on territory 1. The 200 sensors form for example a fixed or mobile network of sensors, the network being configured to place the 200 sensors at the locations 11, themselves fixed or mobile.

[0131] The initial data is received, for example, by a beacon unit 22 from the computer 2, which is communicating with the sensors 200. The computer 2 establishes, for example, a communication network, such as a multiplexed communication network, in which data is transmitted via a wireless or wired connection. The computer 2 thus establishes communication with the sensors 200, which serve as data acquisition devices, and with the beacon unit 22 to enable data exchange.

[0132] In one particular embodiment, an emissions determination system for area 1 is provided, comprising, on the one hand, a plurality of sensors 200 associated with area 1, such that each sensor is associated with a location 11, and on the other hand, the emissions determination device 2. Within the system, the device 2 and the plurality of sensors 200 are configured to communicate, so that the device 2 receives the initial data.

[0133] In another example, the initial data is received from a server communicating with the beacon unit 22, for example, a server configured to centralize data from sensors 200 or a server from another database. In a particular example, the initial data is received from a satellite database server, with locations 11 corresponding to specific points from which the data is obtained. The system described above can also integrate a server for centralizing data from sensors 200.

[0134] Optionally, the computer 2, for example the beacon unit 22, receives fourth data representing a map of territory 1. This fourth data corresponds, for example, to topographic and / or satellite data, or more generally to any type of data that allows for the representation of a ground plan of territory 1, and in particular to one or more of the parameters described below. In a specific example, the fourth data is obtained from a map of territory 1. From this fourth data, the computer 2, for example the processor 21, determines a plurality of zones within territory 1, such that each zone is associated with a distinct emission potential. The fourth data is then classified, for example, using a homogeneous polygon class algorithm. The classes form, for example, the zones.Preferably, calculator 2 groups the classes into several categories, for example 7 categories, based on their emission potential, with the categories forming the zones. In other words, the data from the classes are aggregated within each category. Grouping the classes allows for faster data processing. For example, the following categories can be distinguished:

[0135] - green space;

[0136] - watercourse;

[0137] - low-density urban area;

[0138] - high-density urban area; and

[0139] - road axis.

[0140] Finally, the computer 2 selects a set of locations 11 within the territory 1 based on the plurality of zones, so that the first data are received with respect to the selected locations 11. The selection of the set of locations 11 corresponds, for example, to the selection of specific points from satellite data, to the selection of a subset from a network of 200 fixed sensors, or to the control of a network of 200 mobile sensors so that the positions of the 200 sensors correspond to the locations 11. Thus, the locations 11 are associated with key points in the territory 1, selected according to their emission potential, for example, in order to measure emissions as close as possible to their potential sources, particularly near major roads.

[0141] Alternatively or complementarily, the computer 2, for example the beacon unit 22, receives a set of parameters, for example at least partially comprised of, or derived from, the fourth data point. This set of parameters includes:

[0142] - information representative of land use patterns in the territory 1; and / or

[0143] - information representative of habitat characteristics in territory 1, for example density and / or type of habitat; and / or

[0144] - information representative of an activity associated with territory 1, for example a service sector or an industrial activity; and / or

[0145] - information representative of a road transport network associated with territory 1, for example the layout of roads and / or the type of road, i.e. the presence of highways, expressways, ring roads, etc. As before, depending on the set of parameters, the computer 2, for example the processor 21, selects a set of locations 11 on territory 1. The set of locations 11 is, for example, determined jointly depending on the set of parameters and the plurality of zones.

[0146] In the example described here, territory 1 has dimensions between 12 and 15km on each side, and approximately 75 locations 11 are selected on this territory 1, the locations 11 being spaced between 1km and 2.7km apart in order to distribute the measures 4 over territory 1.

[0147] Advantageously, and in accordance with the first graph in Figure 4, the initial data are received at intervals over a given time period. In other words, measurements 4 are taken over time 41. Preferably, measurements 4 are taken over a long period, for example, at least one year. The interval between two measurements can correspond to any interval considered sufficient by a person skilled in the art to account for variations, and in particular to detect any emission peaks, for example, a 10-minute interval.

[0148] Advantageously, the computer 2, for example the processor 21, selects a subset of emission concentration measurements within the given time period, based on the initial data. In other words, the processor 21 selects one or more time intervals and the associated measurements 4. The subset is advantageously selected so that the concentration measurements are above a reference value. In one particular example, the selected subset shows concentration measurements in their upper percentiles for a plurality of locations 11. In other words, the subset is selected to study the time periods in which a significant number of measurements are elevated. The subsequent steps of the process 3 are then performed with respect to this subset of measurements.This design makes it possible to reduce the total number of calculations, while distinguishing measurements from background noise or measurement noise.

[0149] In a second step 32, the computer 2 obtains further representative data from atmospheric flow maps over territory 1. The atmospheric flows here are intended to represent the impact of wind, and more generally meteorological conditions, on the movement of emissions between their source and their measurement. Specifically, the computer 2 implements and / or communicates with a GIS to process the data, taking into account its integration within territory 1. Preferably, and as illustrated in Figures 4 and 6, the computer 2, particularly the beacon unit 22, receives representative wind information 6, 6' over the territory. This representative wind information 6, 6' includes, for example, a wind measurement 6, such as a wind speed measurement in m / s 43 along different directions or at different points within territory 1.According to the example in Figure 6, the representative wind information 6, 6' includes at least one wind rose 6', that is, information combining the direction, frequency, and intensity of the winds for the entire territory 1 or for a given location 11. Generally, the representative wind information 6, 6' includes information on the speed 61 (in km / h), intensity, direction, and vertical profile of the winds at one or more points within the territory 1.

[0150] For example, 200 sensors are planned, configured on the one hand to measure an emission concentration (e.g., a CO2 level), and on the other hand a wind speed at a given location 11, in which case the first data received by the computer 2 also include a wind measurement 6. Information representative of winds 6, 6' can also be obtained by communication with a satellite database, for example the ERA5 reanalysis data repository provided by the Copernicus Climate Change Service, which provides digital meteorological data over a one-hour resolution window.

[0151] The representative wind information 6, 6' is then used by the computer 2, for example by the processor 21, to determine the second set of data. The computer 2 determines, for example, wind fields over territory 1. Figure 9 illustrates an example of wind fields 9 over territory 1, combining their direction and speed in m / s 91. The atmospheric flows thus correspond to the movements of emissions over territory 1, particularly under the influence of the wind fields 9.

[0152] Additionally, and as illustrated in Figure 7, the calculator 2 also obtains representative information from districts 7 associated with territory 1. These districts 7 are defined, for example, from map data or received directly, such as through communication with a Copernicus Urban Land Cover service, which defines urban blocks. The districts 7 thus correspond to different types of land use, particularly different types of buildings, in order to represent the friction between the terrain and atmospheric flows. Each district 7 is associated, for example, with a specific roughness level, depending on the type of district 7. In particular, this design allows such friction to be modeled with comparable accuracy and lower computational requirements than a complete model of territory 1 and its associated buildings.Calculator 2 then uses representative information from districts 7, and in particular the associated roughness, to determine atmospheric flows. It is advantageous to determine atmospheric flows by combining winds 6 and districts 7 across the territory, so as to take into account, as precisely as possible, the dispersion and advection mechanisms of emissions from their source.

[0153] Advantageously, the second set of data is obtained by calculation on a mesh, that is, a discretized calculation of atmospheric flows. According to the example in Figure 8, the computer 2 thus obtains a first mesh 81 representative of territory 1, via which the second set of data is calculated. The computer 2 communicates with, or integrates, for example, a dedicated modeling tool, such as a Fluidyn-PANACHE tool, to generate the first mesh 81. In this same example, the first mesh 81 corresponds to an irregular mesh, that is, a mesh whose cells have a variable size. The irregular mesh is determined here based on a set of potential emission sources in territory 1; that is, the first mesh 81 has smaller cells at the level of potential emission sources, so that the calculations performed there are more precise.Calculator 2 determines, for example, potential sources of emissions from map data as described above, in particular by taking into account major road axes as sources of emissions, or based on representative information of road traffic density.

[0154] Calculator 2 therefore obtains the second data via the first mesh 81, in particular by simulating atmospheric flows on territory 1, that is to say by discretized calculation of atmospheric flows on each mesh.

[0155] Computer 2, for example processor 21, is advantageously configured to solve the Navier-Stokes equations, using a RANS approach, with a k-eps turbulence model. According to one embodiment, processor 21 applies modified similarity laws for velocity, temperature, and atmospheric turbulence, in particular the following laws: [Math. l]

[0156] [Math.2]

[0157] Where u corresponds to the wind speed, 9 to the temperature, and K to the turbulent diffusivity. Once performed on the entire first mesh 81, the calculations allow us to obtain a flow map. As stated previously, the flow map can be obtained from, among other things, representative wind information 6, 6'. The representative wind information 6, 6' defines, for example, the initial conditions on one or more cells of the first mesh 81, from which the calculator 2 determines the flows on all the cells.

[0158] According to an advantageous variant, the second calculator obtains a set of reference weather conditions, and the second set of data is determined based on each weather condition within that set. In other words, each flow map is associated with a reference weather condition. This design corresponds to the production of a set of reference flow maps, each associated with a given weather condition, rather than the calculation of a specific flow map for every instantaneous weather condition.

[0159] Thus, calculator 2 establishes a set of scenarios, forming the reference meteorological situations and corresponding to a set of cases from which flow maps can be created. For example, the most common situations associated with territory 1 are selected, such as the preferred wind direction lobes illustrated by the wind rose 6', here winds blowing from the South-Southwest and North-Northeast. In a variant, situations are also selected where the winds blow along less preferred lobes, here East-West direction situations. The reference meteorological situations are also established, for example, according to the time of day, so as to cover daytime and nighttime periods and peak traffic situations, as well as according to the season and / or temperature, in particular to cover winter periods and urban heating peak situations.Calculator 2 can thus receive information representative of the time and / or date and / or local temperature in order to establish reference weather situations.

[0160] It is thus understood that, for each of the established reference meteorological situations, the computer 2 determines a flow map, notably from the representative wind information 6, 6' and / or the first grid 81, in order to produce a set of flow maps, for example stored in the memory 20 of the computer 2. In a third step 33, the computer 2, for example the processor 21, determines third data points representing emission source maps 5 from the first and second data points, notably map 5 in Figure 5, which illustrates the emission sources in kg / m³ 2 / s 51 on territory 1. In particular, the processor 21 processes the flow maps in such a way as to reverse the dispersion of emissions in the atmosphere and trace the flows back to their source. The measurements 4 in ppm are then converted into a volume, or surface volume, of emissions on territory 1, the location 11 of the measurements 4 being combined with the flow map to determine the position of the emission sources. Notably, this design makes it possible both to distinguish the locations 11, where the emissions are measured, from their sources where they are produced, and to distinguish emissions produced within or outside territory 1, by identifying the impact of the flows on the emissions.

[0161] According to an advantageous variant, calculator 2 receives representative information on the current weather situation. The third set of data is then determined by combining the reference weather situations and the current weather situation, that is, by selecting and / or weighting the reference weather situation(s), and consequently the flow maps, corresponding to the current weather situation. The matching of weather situations is performed, for example, using the various pieces of information listed above, namely winds, temperature, time, date, season, etc. Thus, the calculation requirements are limited, avoiding the need to calculate flow maps for each current weather situation.

[0162] As described below, the calculation is also planned, for each location 11, of adjoint fields, corresponding to the emission sources of a measure 4 associated with that location 11. The source mapping 5 then corresponds to a combination of the adjoint fields of each location 11, for example weighted by the value of their measure 4.

[0163] In other words, it is possible to determine adjoint fields for each combination of a location 11 and a reference meteorological situation. The determination of the third data corresponds to a selection and weighting of the locations 11 based on measurements 4, and of the reference meteorological situations based on instantaneous meteorological situations. The adjoint fields are, for example, determined beforehand and stored in the memory 20 of the computer 2, thus forming a database for reconstructing the emission source map 5. Each iteration of the process 3 then implements a selection and weighting of the adjoint fields based on measurements 4, in order to obtain the emission source map 5.

[0164] As before, the third data are advantageously determined via the first mesh 81, in order to discretize the calculations. The calculator 2 thus simulates the fields annexed to the locations 11 via the first mesh and as a function of the second data.

[0165] In the classical framework of a known source dispersion model, the mass fraction %(X,t) of a chemical species is obtained by solving an ADE equation, a solution of: [Math.3]

[0166] With u the wind velocity vector, G the spatial distribution of the source and Ç the turbulent diffusion such that:

[0167] [Math.4]

[0168] With K being the turbulent diffusivity.

[0169] The measurement g, at location i is given by integration over a volume and an acquisition time of the mass fraction with the sampling function 7ti such that: [Math.5]

[0170] Thus, a dual, inverse version of this construction can be obtained by adjoint method by solving an adjoint ADE, that is to say an ADE going back in time and flow: [Math.6]

[0171] The adjoint field of a location i is then linked to the measurement by an equivalent integral operating on the spatial distribution of the emission sources. The following integral describes the projection of the source G onto the adjoint functions n to obtain the set of measurements 4 at locations 11:

[0172] [Math.7]

[0173] The relationship between the source, the adjoint functions, and the measurements 4 is invertible by exploiting the inverse of the Gram matrix of the adjoint functions H = (n, q). Calculator 2 thus uses the following renormalization function, which gives the real region in which the sources s can be reconstructed from the measurements 4:

[0174] [Math.8] Thus, from local measurements of emission concentrations, we obtain a quantity of emissions produced at sources distributed within or outside territory 1. Sources outside territory 1 may or may not be taken into account when establishing the emission source map 5, in particular depending on whether we seek to identify all the sources of emissions measured in territory 1, or exclusively to identify the emission sources in territory 1. In the example of figure 5, the emission source map 5 is thus limited to territory 1, that is to say that emissions whose source is located outside territory 1 have been excluded.

[0175] It is understood here that, once the emission source map 5 has been obtained, further processing can be carried out to analyze or render the third data points in a specific way. For example, the computer 2 transmits the third data points via the beacon unit 21 to a human-machine interface for displaying graphic content representative of the emission source map 5, or any other appropriate data. In the example in Figure 5, the data have also been processed to remove a background value, by removing, for each location 11, the minimum value observed across all the initial data points, in order to obtain third data points free of background noise.

[0176] Calculator 2 can also analyze a plurality of emission source maps 5 obtained from different measurements 4 over time. Calculator 2 determines, for example, a total volume of emissions associated with territory 1 or a part thereof, corresponding to the sum of the emission sources within it. Calculator 2 can also identify high-contribution areas, corresponding to sources for which the emission volume exceeds a threshold value. Calculator 2 can also forecast emissions for a given period, based on emission source maps 5 associated with similar periods and, for example, based on weather forecasts for a future period.

[0177] Preferably, and as illustrated in Figure 8, the computer 2 generates a second mesh 82 of territory 1. The second mesh 82 is regular, meaning that all cells have a roughly identical size, to facilitate the processing, analysis, and rendering of the third data. For example, hexagons of approximately 63m per side are used. The computer 2 also assigns the third data to the second mesh 82 by aggregating the third data from the first mesh 81. In other words, the third data is converted between the first mesh 81 and the second mesh 82 based on the correspondence between the cells.

[0178] Optionally, computer 2 implements a source allocation algorithm to assign emissions to a plurality of source sectors. In this algorithm, computer 2 obtains fifth data points representative of a set of activity sectors, corresponding, for example, to the key source sectors of territory 1. These fifth data points are obtained, for example, from satellite data, or from the fourth data points, notably from a narrow spectroscopic classification of satellite images of territory 1. These fifth data points allow for an estimation of a land cover rate for each cell of the second grid 82. For example, a normalized grid of the proportions of each activity sector is obtained.

[0179] Calculator 2 then assigns an emission level to each sector of activity based on the fifth and third data points. In other words, since each cell in the second grid is associated with an emission level and distributed across several sectors of activity, emissions can be proportionally allocated among these sectors, thus distributing the emissions of territory 1 across all sectors. This design makes it possible, at the territorial level, to identify the sectors of activity that contribute the most to emissions.

[0180] Thus, it will be understood that the present invention provides a method, a device, and a system for determining emissions over a territory, which makes it possible to obtain a map of emission sources across the entire territory based on local concentration measurements, taking into account the movement of emissions over time to clearly identify the sources. This method thus makes it possible to visualize concretely the impact of activities in a territory on emission production, particularly CO2. It will be understood, however, that such a method is also applicable to a variety of molecules, particles, or other substances that can be measured locally and that move via atmospheric flows.

[0181] It should be noted that this detailed description relates to a particular embodiment of the present invention, but in no way does this description limit the scope of the invention; on the contrary, its purpose is to remove any possible inaccuracy or misinterpretation of the following claims.

[0182] Of course, the present invention is not limited to the embodiments described above but extends to a method for determining emissions from a territory that would include secondary steps without falling outside the scope of the present invention. The same would apply to a system and / or device configured for implementing such a method.

[0183] It should also be noted that the reference signs placed in parentheses in the following claims are in no way intended to be limiting; these signs are solely intended to improve the intelligibility and understanding of the following claims and the scope of the protection sought.

Claims

Demands 1. A method (3) for determining emissions from a territory (1), said method (3) being implemented by at least one processor, said method (3) comprising the following steps: - receipt (31) of initial measurement data (4) of emission concentrations associated with said territory (1), each of said measurements (4) being associated with a location (11) on said territory (1); - obtaining (32) second representative data from atmospheric flow maps associated with said territory (1); and - determination (33) of third data representative of emission source maps (5) on said territory (1), from said first and second data.

2. A method (3) according to claim 1, further comprising receiving a set of parameters including: - information representative of land use patterns in said territory (1); and / or - information representative of habitat characteristics of said territory (1); and / or - information representative of activity associated with said territory (1); and / or - information representative of a road transport network associated with said territory (1), said process further comprising a step of selecting a set of locations (11) on said territory (1) at least according to said set of parameters, said first data being received with respect to said set of locations (11).

3. A method (3) according to claim 1 or 2, further comprising the following steps: - receipt of fourth data representative of a mapping of said territory (1); - determination of a plurality of zones of said territory (1) according to said fourth data, each zone of said plurality of zones being associated with a distinct emission potential, said process further comprising a step of selecting a set of locations (11) on said territory (1) at least according to said plurality of zones, said first data being received with respect to said set of locations (11).

4. Method (3) according to any one of claims 1 to 3, wherein said first data are received at intervals during a given time period.

5. A method (3) according to claim 4, further comprising selecting, within said time period, a subset of emission concentration measurements based on said first data, said subset having concentrations greater than a reference value, said third data being obtained from said subset of measurements.

6. Method (3) according to any one of claims 1 to 5, further comprising obtaining representative wind information (6, 6') over said territory (1), said second data being obtained at least as a function of said representative wind information (6, 6').

7. Method (3) according to any one of claims 1 to 6, further comprising obtaining representative information of districts (7) associated with said territory (1), each district being associated with a level of roughness with respect to atmospheric flows, said second data being obtained at least as a function of said representative district information (7).

8. Method (3) according to any one of claims 1 to 7, further comprising obtaining a set of reference meteorological situations, said second data being further determined according to each meteorological situation of said set, said method (3) further comprising receiving information representative of instantaneous meteorological situations, said third data being further determined according to said instantaneous meteorological situations.

9. Method (3) according to any one of claims 1 to 8, further comprising obtaining a first mesh (81) representative of said territory (1), said second data and / or said third data being obtained at least partially via said first mesh (81).

10. Method (3) according to claim 9, which comprises in one part obtaining information representative of potential sources of emissions on said territory (1), said first grid (81) corresponding to an irregular grid determined according to said potential sources.

11. Method (3) according to claim 9 or 10, wherein said obtaining (32) of said second data comprises a first simulation, via said first mesh (81), of said atmospheric flows over said territory (1).

12. Method (1) according to any one of claims 9 to 11, wherein said determination (23) of said third data comprises a second simulation, via said first mesh (81), of fields adjoint to said locations (11) of said measurements as a function of said second data, said third data being determined as a function of said adjoint fields and said first data.

13. Method (3) according to any one of claims 1 to 12, further comprising a generation of a second regular mesh (82) of said territory (1) and an assignment of said third data to said second mesh (82).

14. A method (3) according to claim 13, further comprising the following steps: - obtaining fifth data points representative of a set of business sectors associated with the meshes of said second mesh (82); and - Assignment of an emission level to each sector of activity based on the aforementioned fifth and third data points.

15. Device (2) for determining emissions from a territory (1), said device comprising a memory (20) associated with at least one processor (21) configured for carrying out the steps of the process according to any one of claims 1 to 14.

Images