Calculating unit, shade determination device, fluid distribution system and associated distribution process

Abstract

Calculation unit, hue determination device, fluid distribution system and associated distribution method The present invention relates to a calculation unit (50) for a device (14) comprising: at least one light source (22) configured to emit light in the direction of a fluid; a light sensor (24) configured to receive a light signal representative of a hue of the fluid flowing in a fluid distribution system (10), the light sensor (24) being further configured to emit quantities representative of a light intensity of the light signal for a given color channel;the computing unit comprising an estimation module (52) configured to determine a variable representing a color of the fluid via an artificial intelligence model, each quantity representing a light intensity being an input variable of the model, an output variable of the artificial intelligence model being the variable representing a color of the fluid. Figure for the abstract: Figure 1;

Description

Title of the invention: Calculation unit, shade determination device, fluid distribution system and associated distribution method

[0001] The present invention relates to a calculation unit, a shade determination device, a fluid distribution system and an associated distribution method.

[0002] In the context of a coating application system, it is known to determine the tint of a fluid, for example, paint or cleaning fluid, circulating in a fluid distribution system, in order to determine whether the correct tint of paint is circulating in the system, or whether the cleaning fluid is pure, in order to determine the cleanliness of the fluid distribution system. For example, during a tint change operation, it is known to perform a visual inspection of the fluid sprayed by the distribution system to verify that the tint change operation is complete and that the new tint of paint is circulating in the system without being mixed with the old paint.

[0003] It is also known, for example from FR 3 127 281 A, to use light sensors to measure the color of the fluid. For this purpose, it is known to use an optical fiber, which carries a light signal to a light sensor, the light signal being representative of the fluid's color. However, the use of the optical fiber causes attenuation of the light signal. The value of the attenuation depends on the length of the optical fiber and the wavelengths composing the light signal. Thus, the light sensor must be manually calibrated to take into account the attenuation caused by the optical fiber in order to correctly determine the fluid's color, which is a lengthy and complex process.

[0004] The aim of the invention is therefore to resolve these drawbacks and to determine the color of the fluid in a simpler and more reliable way.

[0005] To this end, the invention relates to a calculation unit for a color determination device for a fluid distribution system, the device comprising:

[0006] - at least one light source configured to emit light polychromatic or monochromatic in the direction of the fluid in a measurement zone;

[0007] - a light sensor configured to receive a light signal that has been reflected on the fluid or transmitted through the fluid, the light signal corresponding to an optical reflection or, respectively, to an optical transmission, by the fluid, of polychromatic or monochromatic light emitted in the direction of the fluid, the the light signal being representative of a color of the fluid circulating in the fluid distribution system, the light sensor being further configured to emit quantities, each quantity being representative of a luminous intensity of the light signal for a given color channel; and

[0008] - an acquisition optical fiber configured to carry the light signal to light sensor;

[0009] the computing unit comprising an estimation module configured to receive the quantities emitted by the light sensor.

[0010] According to the invention, the estimation module is further configured to determine a variable representing a hue of the fluid via a previously trained artificial intelligence model, each quantity representing a luminous intensity of the light signal for a given color channel being an input variable of the model, an output variable of the artificial intelligence model being the variable representing a hue of the fluid.

[0011] Thanks to the invention, determining the fluid's color does not require manual calibration of the light sensor. Furthermore, the use of the artificial intelligence model makes it possible to obtain reliable results without requiring additional light signal processing steps and by limiting the complexity of the calculations, which are costly in terms of resources and time. Thus, determining the fluid's color is simpler for users.

[0012] According to other advantageous aspects of the invention, the calculation unit comprises one or more of the following features, taken individually or in all technically possible combinations:

[0013] - the variable representing the color of the fluid is an index of fluid purity circulating in the fluid distribution system; - the purity index is a decimal number, preferably between 0 and 1;

[0014] - the purity index is a binary number,

[0015] the purity index then preferably being a class indicator among a first class corresponding to the case where the fluid is clean, and a second class corresponding to the case where the fluid is dirty;

[0016] - the variable representing the color of the fluid is a class indicator among a plurality of classes, each class corresponding to a predefined shade of fluid; - the fluid is a coating liquid or a cleaning liquid or an aerosol composed of solid colored particles suspended in a gas;

[0017] The invention also relates to a color determination device for a fluid distribution system, comprising:

[0018] - at least one light source configured to emit light polychromatic or monochromatic in the direction of the fluid in a measurement zone;

[0019] - a light sensor configured to receive a light signal that has been reflected on the fluid or transmitted through the fluid, the light signal corresponding to an optical reflection or respectively, to an optical transmission, by the fluid, of polychromatic or monochromatic light emitted in the direction of the fluid, the light signal being representative of a hue of the fluid circulating in the fluid distribution system, the light sensor being further configured to emit quantities, each quantity being representative of a luminous intensity of the light signal for a given color channel;

[0020] - an acquisition optical fiber configured to carry the light signal to light sensor; and

[0021] - a computing unit as described above.

[0022] Advantageously, the given color channels are red, green, blue and clear channels, the quantities being representative of a luminous intensity of the light signal respectively for the red channel, the green channel, the blue channel and the clear channel.

[0023] The invention also relates to a fluid distribution system comprising a fluid flow circuit, and a color determination device as described above, connected in series to the fluid flow circuit.

[0024] The invention also relates to a method for determining the color of a fluid circulating in a fluid distribution system, implemented by a color determination device described above, the method comprising the following steps:

[0025] - emission of polychromatic or monochromatic light in the direction of the fluid by at least one light source;

[0026] - reception, by the light sensor, of the light signal reflected on the fluid or transmitted through the fluid;

[0027] - reception, by the estimation module, of quantities representative of an intensity luminous intensity of the light signal for the given color channels; and

[0028] - determination by the estimation module of the variable representing the hue fluid.

[0029] According to other advantageous aspects of the invention, the color determination method comprises one or more of the following features, taken individually or in all technically possible combinations:

[0030] - the process further comprises the following steps:

[0031] - detection of a fluid change operation when the variable the representative color of the fluid is altered;

[0032] - a fluid change operation having been detected, determination of an end of the fluid change operation, when the variable representing the fluid color is substantially constant for a predetermined period,

[0033] the fluid change operation being preferably a system flush or a change of color of fluid circulating in the system;

[0034] - the fluid is a coating liquid or a cleaning liquid.

[0035] The invention will become clearer upon reading the following description, given solely by way of non-limiting example, and made with reference to the drawings in which: - [Fig. 1] [Fig. 1] is a diagram of a fluid distribution system according to the invention; and - [Fig.2] [Fig.2] is a perspective and cross-sectional view of part of the fluid distribution system of [Fig.1], corresponding to detail A in this figure; - [Fig.3] [Fig.3] is a logic diagram of a process according to the invention; - [Fig. 4] [Fig. 4] is a perspective view of part of a system of fluid distribution according to a second embodiment of the invention, corresponding to detail A in [Fig.1].

[0036] The invention is described below in the context of a coating fluid distribution system, for example, for paint, for use in a coating application installation, particularly by spraying. The paint may be in the form of a liquid or an aerosol, composed, for example, of solid colored particles suspended in a gas. However, this does not limit the invention to this particular application; the invention can be implemented in any fluid distribution system suitable for conveying a fluid, in which it is advantageous to identify the color of the fluid, particularly for monitoring operations or phenomena involving a change in the fluid's color.The fluid can be liquid or gaseous, and in particular, in the following description, the term fluid also includes solid particles transported by a gas flow, or suspended in a gas, especially an air flow. Solid particles are, for example, colored powders.

[0037] In the following description, the expression "approximately equal to" defines a relationship of equality to plus or minus 10%, preferably even to plus or minus 5%.

[0038] Figure 1 is a diagram of a fluid distribution system 10. The fluid distribution system 10 is, for example, a paint distribution system, and includes at least one paint component reservoir, a projector 11, also designated as a sprayer, and a fluid flow circuit 12.

[0039] In the example of [Fig. 1], the fluid distribution system 10 comprises three tanks 1, 2, and 3 of paint in different shades and a tank 4 of cleaning fluid, for example, aqueous, also called solvent. Tanks 1 to 4 are grouped together in a dedicated compartment 6, often referred to as a grinding room. The number and arrangement of tanks 1 to 4 in the grinding room are not limiting. They are adapted according to the intended use of the fluid distribution system 10.

[0040] A color-changing device 8 is supplied by the different reservoirs 1 to 4 and allows selection of which product, paint or cleaning fluid, circulates in the fluid flow circuit 12 to supply the projector 11.

[0041] The projector can, for example, consist of a manual spray gun, as shown in [Fig. 1]. In an alternative not shown, it is an automatic projector, of pneumatic or rotary type, mounted on the arm of a robot, of multi-axis or reciprocating type.

[0042] Advantageously, the projector 11 can be of the electrostatic type.

[0043] The fluid flow circuit 12 includes pipes 122 and 124 and connects the color-changing device 8, i.e. one of the reservoirs 1 to 4, with the projector 11 via pipes 122 and 124.

[0044] The fluid distribution system 10 further includes a shade determination device 14.

[0045] Part of the shade determination device 14 is connected in series to the fluid flow circuit 12, between the pipes 122 and 124.

[0046] Advantageously, the color-determining device 14 comprises a flange 13, advantageously opaque, which defines a chamber 15, visible in [Fig. 2], through which a fluid flow channel 16 is provided. The flange 13 is arranged to be connected on both sides to the pipes of the fluid flow circuit 12. The fluid flow channel 16 is represented by an arrow in [Fig. 2].

[0047] The shade determination device 14 includes at least one light source 22. It further includes a light sensor 24.

[0048] In the example of [Fig. 1], the hue-determining device 10 comprises only one light source 22. The light source 22 is configured to emit polychromatic light, i.e., light comprising at least two different wavelengths, or monochromatic light, i.e., light comprising a single wavelength. Polychromatic light preferably consists of white light. The light source 22 is, for example, formed of several monochromatic sources, each with a different respective wavelength, for example a A red laser source, a green laser source, and a blue laser source, or several light-emitting diodes (LEDs) of different colors. Alternatively, the light source 22 is a light source emitting a continuous light spectrum, such as an incandescent lamp. Furthermore, the light source(s) can be configured to emit pulsed light.

[0049] Alternatively, in the case where the light source 22 is configured to emit monochromatic light, the light source 22 is formed for example of one or more monochromatic sources of the same wavelength, for example a laser or a monochromatic LED.

[0050] The light source 22 is configured to emit light in the direction of the fluid, here paint, in a measurement area 23, shown in dotted lines in [Fig.2] and formed within the chamber 15. In particular, the light illuminates the fluid present within the flow vein 16, in the measurement area 23.

[0051] The light sensor 24 is configured to receive a light signal and to output quantities, each quantity representing a luminous intensity of the light signal for a given color channel. The given color channels are, for example, a red channel, a green channel, and a blue channel. Alternatively, the given color channels are a red channel, a green channel, a blue channel, and a clear channel, the latter channel representing the luminous intensity of the light signal as a whole. In this case, the quantities output by the light sensor 24 represent a luminous intensity of the light signal for the red channel, the green channel, the blue channel, and the clear channel, respectively. In other words, the light sensor 24 is configured to output a quantity representing a luminous intensity in the RGBC, or Red Green Blue Clear, system.

[0052] The luminous intensity is, for example and in a way known per se, represented by a number between 0 and 255, 0 corresponding to a zero intensity for the channel considered, and 255 corresponding to a maximum intensity for the channel considered.

[0053] It is possible to position the light sensor 24 and / or the light source 22 near the measurement area 23. However, the quality of the dispensed fluid or its environment can often impose constraints requiring compliance with regulations for use in explosive atmospheres, such as ATEX in Europe (for "ATmosphère EXplosive"), as in the context of the paint dispensing system. Using a light sensor 24 and / or a light source 22 compatible with this constraint can be costly, and its integration into the color measurement device can be complex or even impossible.

[0054] Thus, the light source 22 and the light sensor 24 are preferably arranged at a distance from the fluid flow circuit 12, for example by being grouped in a treatment box 26 which also belongs to the shade determination device 14.

[0055] In order for the light emitted by the light source 22 to illuminate the fluid in the measuring zone 23, the device 14 comprises one or more optical fibers 34, referred to as illumination optical fibers. The illumination optical fibers 34 are advantageously four in number, as shown in [Fig. 2]. The illumination optical fibers 34 are generally several meters long, for example at least ten meters, and preferably up to twenty-five meters. The illumination optical fibers 34 are advantageously connected to the light source 22 at one of their ends. The other end of the illumination optical fibers 34 is advantageously located in the flange 13.

[0056] In an alternative not shown, when the device 10 includes several light sources 22, each illuminating optical fiber 34 is connected to one of the light sources 22.

[0057] The device 14 further comprises an acquisition optical fiber 36, which is advantageously connected to the light sensor 24 by one of its ends, the other end being located in the flange 13. Thus, a light signal from the measurement area 23 is routed to the light sensor 24 via the optical fiber 36. The acquisition optical fiber 36 is of the same length as the illuminance optical fibers 34.

[0058] Advantageously, means for retaining the end or ends of the illumination and acquisition optical fibers 34 and 36, such as for example a cable gland 38, are disposed in the flange 13.

[0059] Advantageously, the device 14 further comprises an observation window 40 disposed tangentially to the fluid flow channel 16 in the flange 13. The observation window 40 may, in particular, be shaped like a porthole, as seen in [Fig. 2]. In the example of [Fig. 2], the measuring area 23 corresponds to the portion of the fluid flow channel 16 visible through the observation window 40.

[0060] The device 14 advantageously includes a retaining element 42 mounted in the flange 13, so as to hold the observation window 40 in position and to ensure the sealing of the fluid flow channel 16. The retaining element 42 may, for example, include a sealing gasket surrounding the observation window 40.

[0061] The device 14 further includes a computing unit 50. The computing unit 50 is connected to the light source 22 and the light sensor 24, and is formed, for example, of a memory and a processor associated with the memory, not shown. The calculation unit 50 is advantageously located inside the processing box 26.

[0062] The computing unit 50 includes an estimation module 52, as shown in [Fig. 1]. In the example of [Fig. 1], the estimation module 52 is implemented as software, or a software component, executable by the processor. The memory of the computing unit 50 is then capable of storing the estimation software, and the processor is then capable of executing the estimation software.

[0063] The estimation module 52 is configured to receive the quantities emitted by the light sensor 24, and then to determine a variable representing a fluid hue via a pre-trained artificial intelligence model. The model has, as input variables, each quantity representing a luminous intensity of the light signal for a given color channel and, as output variables, a variable representing a fluid hue. The variable representing a fluid hue represents a color of the fluid, or alternatively or in addition, an opacity of the fluid. In particular, in the case where the light emitted by the light source 22 is monochromatic, the variable representing a fluid hue represents the opacity of the fluid.In the case where the light emitted by the light source 22 is polychromatic, the variable representing a hue of the fluid is representative of the opacity and / or hue of the fluid.

[0064] The variable representing the color of the fluid is, for example, a purity index of the fluid circulating in the fluid distribution system 10. According to this example, the purity index is a decimal number, typically between 0 and 1; or a binary number, typically equal to 0 or 1. When the purity index is a binary number, the purity index is advantageously a class indicator among a first class when the fluid 10 is clean, and a second class when the fluid 10 is dirty.

[0065] According to an example, when the purity index is a decimal number, it is directly proportional to the opacity of the fluid.

[0066] Alternatively, the variable representing the color of the fluid is a class indicator among a plurality of classes, each class corresponding to a predefined color of fluid.

[0067] In an alternative not shown, the estimation module 52 is implemented as a programmable logic component, such as an FPGA (Field Programmable Gate Array), or as an integrated circuit, such as an ASIC (Application Specified Integrated Circuit).

[0068] When the computing unit 50 is implemented in the form of one or more software programs, i.e. in the form of a computer program, also called a computer program product, it is also capable of being stored on a medium, not shown, Computer-readable media. A computer-readable medium is, for example, a medium capable of storing electronic instructions and being connected to a bus of a computer system. Examples of readable media include optical discs, magneto-optical discs, ROM, RAM, any type of non-volatile memory (e.g., FLASH or NVRAM), or magnetic cards. A computer program containing software instructions is then stored on the readable medium.

[0069] The device 14 advantageously includes a human-machine interface 54, connected to the computing unit 50, for displaying information to a user. Alternatively or in addition, the human-machine interface 54 is used by the user to control the device 14. The human-machine interface 54 then typically includes a display screen, and optionally an input device, such as a keyboard and mouse, which are not shown.

[0070] The artificial intelligence model is pre-trained, advantageously by machine learning, for example via supervised learning, as known in itself.

[0071] The artificial intelligence model is typically trained using training datasets, each training dataset comprising input training data corresponding to the model's input variables, namely the representative quantities of light intensity for the given color channels, and output training data corresponding to the expected, i.e., target, representative variable of fluid hue for this input training data. The model is then trained, for example, via backpropagation of an error gradient, the error being calculated from the difference between the output training data and the output variable estimated by the model from the input training data, that is, from the difference between the target representative variable and the representative variable estimated for this input training data.

[0072] The artificial intelligence model is for example a random forest, a support vector machine (SVM) model, or an artificial neural network (ANN).

[0073] In the case of a neural network, it comprises an ordered succession of layers of neurons, each of which takes its inputs from the outputs of the previous layer.

[0074] More precisely, each layer comprises neurons taking their inputs from the outputs of the neurons of the previous layer, or from the input variables for the first layer.

[0075] Alternatively, more complex neural network structures can be envisaged with a layer that can be linked to a layer further away than the immediately preceding layer.

[0076] Each neuron is also associated with an operation, that is to say a type of processing, to be carried out by said neuron within the corresponding processing layer.

[0077] Each layer is connected to the other layers by a plurality of synapses. A synaptic weight is associated with each synapse, and each synapse forms a link between two neurons. It is often a real number, which takes on both positive and negative values. In some cases, the synaptic weight is a complex number.

[0078] Each neuron is designed to perform a weighted sum of the value(s) received from the neurons of the preceding layer, each value being multiplied by the respective synaptic weight of each synapse, or connection, between said neuron and the neurons of the preceding layer, and then to apply an activation function, typically a non-linear function, to said weighted sum, and to deliver at the output of said neuron, in particular to the neurons of the next layer connected to it, the value resulting from the application of the activation function. The activation function introduces non-linearity into the processing performed by each neuron. The sigmoid function, the hyperbolic tangent function, and the Heaviside function are examples of activation functions.

[0079] As an optional complement, each neuron is also capable of applying, in addition, a multiplicative factor, also called bias, to the output of the activation function, and the value delivered at the output of said neuron is then the product of the bias value and the value from the activation function.

[0080] A method for determining the color of the fluid flowing in the flow circuit 12 is now described. The method is implemented by the device 14.

[0081] The fluid, for example paint or aerosol, flows in the distribution system 10, in particular, in the fluid flow channel 16 of the device 14. The fluid is colored, opaque or possibly translucent, or even transparent.

[0082] During an emission step 100, the light source 22 emits polychromatic or monochromatic light towards the fluid, in the measurement zone 23. In particular, the light is emitted through the illuminating optical fibers 34 to their ends located in the flange 13. The fluid flowing in the measurement zone 23 is thus illuminated by the light through the observation window 40.

[0083] The light reflected by the fluid forms a light signal, which corresponds to the optical reflection by the fluid of the light emitted by the light source 22. The light signal is representative of the color of the fluid.

[0084] The light signal travels from the measurement zone 23 through the acquisition optical fiber 36 to the light sensor 24, which receives the light signal during a reception step 102. The light signal received by the light sensor 24 is attenuated due to the optical fibers 34 and 36, which cause attenuation dependent on their length and on the wavelengths composing the light and the light signal. More specifically, the light is attenuated when it travels through the illumination optical fibers 34, and the light signal is attenuated when it travels through the acquisition optical fiber 36.

[0085] The opaque flange 13 advantageously allows for a significant limitation of any light pollution from the outside environment, and thus ensures that the light signal received by the light sensor 24 during the reception stage 102 is as reliable as possible, with the least amount of external light pollution.

[0086] The light sensor 24 converts the received light signal into quantities, each quantity representing the light intensity of the signal for a given color channel. In the example, the light sensor 24 converts the light signal into four numbers between 0 and 255, corresponding to the light intensity for each channel in the RGBC system.

[0087] The calculation unit 50, in particular the estimation module 52, receives the quantities during a reception step 104.

[0088] The estimation module 52 determines the variable representing a fluid color via the artificial intelligence model during a determination step 106. Each quantity representing the light intensity of the light signal for a given color channel is a respective input variable of the artificial intelligence model. The output variable of the artificial intelligence model is the variable representing the fluid color.

[0089] Advantageously, once the variable representing the color of the fluid has been determined by the estimation module 52, a message representing the variable representing the color of the fluid is displayed on the human-machine interface 54, in visual form, and / or possibly in audible form, during a display step 108. Alternatively, during the display step 108, the computing unit 50 sends a message representing the variable representing the color of the fluid to a remote terminal, for example a computer or a mobile phone of a user which then constitutes the human-machine interface 54.

[0090] Advantageously, the process is implemented by the device 14 to perform any one of the following three tasks.

[0091] A first task is to estimate the purity of the fluid circulating in the fluid distribution system 10. Such a task is performed, for example, during a rinsing of the system 10 with the cleaning fluid. In this case, the variable The value representing the fluid's color is the fluid purity index, and the fluid purity index is expressed as a decimal number. For example, the fluid purity index is a decimal number between 0 and 1, or between 0 and 100. The fluid purity index corresponds to the purity of the fluid circulating in system 10. During a rinsing operation, system 10 is initially dirty, and the variable representing the fluid color will be, for example, approximately zero. Then, as the rinsing progresses, if the color determination process is repeated over time, the value of the variable representing the fluid color increases, becoming approximately 1 or 100, for example, when the fluid is pure, corresponding to a perfectly clean system 10.

[0092] A second task is a binary classification task for fluid purity. In this case, the variable representing a fluid color is the fluid purity index, this index being in the form of a binary number. Preferably, the purity index is then the class indicator between the first class, corresponding to the case where the fluid is clean, and the second class, corresponding to the case where the fluid is dirty. This task is performed, for example, during a rinse of system 10, in order to determine whether system 10 is clean or not.

[0093] The fluid purity index makes it possible to determine whether the system 10 is clean or dirty, for example during a rinsing operation. Indeed, if the fluid is considered dirty, in other words, if the purity index determined by the model indicates the class corresponding to the dirty fluid, then the system 10 is considered dirty, and if the purity index determined by the model indicates the class corresponding to the clean fluid, then the system 10 is considered clean.

[0094] Advantageously, in the case of the first and second tasks, the variable representing the color of the fluid is representative of the opacity of the fluid, the fluid being considered dirty when it is opaque, and considered clean when it is transparent.

[0095] A third task is to determine the fluid's hue from a palette of predetermined hues. In this case, the variable representing a fluid hue is representative of the fluid's color. The variable representing a fluid hue is then the class indicator among the plurality of classes, each class corresponding to a predefined fluid hue. This task is performed, for example, to verify that the hue of the fluid circulating in system 10 is indeed the expected hue.

[0096] Advantageously, if the computing unit 50 is configured to perform several of the three tasks, separate, separately trained models are used for each task. For example, the user indicates which task they want the device 14 to perform, and the computing unit 50 then uses the associated model.

[0097] Advantageously, if a user wishes to perform only the first and / or second task, the light source 22 is configured to emit only monochromatic light. In particular, monochromatic light is sufficient to evaluate the opacity of the fluid, and thus for the model to reliably determine the variable representing the fluid's color. The model training is then also performed using monochromatic light.

[0098] Advantageously, the determination process is carried out continuously, and the successive values ​​of the variable representing the color of the fluid are stored in the memory of the calculation unit 50.

[0099] Optionally, the processing unit 50 detects a fluid change operation when the fluid color representative variable is modified relative to the fluid color representative variable previously determined during a detection step 110, in particular when the change in the fluid color representative variable exceeds a predetermined threshold. If the fluid color representative variable is a class indicator, the fluid color representative variable is considered to have changed when the class indicated by the fluid color representative variable is modified. If the fluid color representative variable is a binary number, the fluid color representative variable is considered to have changed when the value of the binary number changes; or alternatively, in the case of a decimal number, when the value of the number changes by at least 10%.Advantageously, the user chooses the variation necessary for the variable representing the fluid's color to be considered modified. The fluid change operation is advantageously a rinsing or a color change of the fluid circulating in system 10.

[0100] Advantageously, a fluid change operation having been detected at the detection step 110, when the successive representative variables of the fluid color are substantially constant for a predetermined time, the calculation unit 50 determines that a color change operation or a rinsing operation is completed, during a determination step 112.

[0101] By substantially constant, in the case where the variable representing the color of the fluid is a decimal number, we mean that a variation in the variables representing the color of the fluid, determined successively by the artificial intelligence model, is less than 10% over a predetermined period, preferably less than 5% over the predetermined period. In the case where the variable representing the color of the fluid is a binary number or a class indicator, we mean substantially constant that the variables representing the color the fluid determined successively by the artificial intelligence model are identical, i.e. unchanged.

[0102] When the variable representing the fluid color remains substantially constant for a predetermined period, this means that the color of the fluid circulating in the system 10 has stabilized. In the case of a color change, this means that the desired color has been achieved, and in the case of a flush, this means that the system has been completely flushed. Advantageously, the computing unit 52 controls a specific display on the human-machine interface 54, or the sending of a specific message to the remote terminal when the fluid change operation is complete.

[0103] The predetermined duration used to consider that a variable is stable is for example between 0.5 and 5 seconds.

[0104] Alternatively or in addition, the device 14 is configured to transmit the variable representing the fluid color to a control unit configured to control a valve or a set of valves of the distribution system 10 based on the variable representing the fluid color. Thus, it is possible, for example, to optimize color-changing operations or to adjust the fluid composition to obtain the desired color.

[0105] In addition to the above description relating to the learning, i.e., training, of the model, the model training is, for example, carried out in the form of supervised learning. The model receives as input data comprising series of quantities representing the light intensity of the light signal for a given color channel. The series correspond to data provided by a light sensor belonging to a hue determination device comprising optical fibers for illumination and acquisition of a predefined length. Each series is associated with a class, or a decimal number, depending on the task on which the model is trained. The model then learns to make a prediction and is trained until it achieves satisfactory performance. Performance validation operations are advantageously performed on the model.

[0106] The model is then integrated into the estimation unit of the device 14, whose optical fibers 34 and 36 advantageously have the same length as those used for training the model. This makes it possible to obtain more reliable predictions from the artificial intelligence model.

[0107] Alternatively, the artificial intelligence model is trained by reinforcement learning. Thus, while the model is being used to determine the color of a fluid, a user indicates, for example, whether the model's predictions are correct or not, and the user's feedback is taken into account by the model in its future predictions. Reinforcement learning is advantageously performed on an already trained model and allows for refining the model's predictions.

[0108] Figure 4 shows a flange 13 of a color-determining device 114 according to a second embodiment of the invention. The color-determining device 114 is configured to perform a color measurement by transmitting the light signal through the fluid. In this case, the flange 13 of the color-determining device 114 advantageously comprises two recesses 138 for inserting, on the one hand, the ends of the illumination optical fibers 34, and on the other hand, the acquisition optical fiber 36, and two observation windows 140 arranged tangentially to the fluid flow path 16 that passes through a measurement zone 23, each observation window being at the bottom of one of the recesses 138. Advantageously, the observation windows 140 are opposite each other.

[0109] The method for determining the fluid is unchanged, except that the light signal corresponds to the optical transmission, by the fluid, of the light emitted towards the fluid by the light source 22.

[0110] Any feature described for an embodiment or variant in the foregoing may be implemented for the other embodiments and variants described above, provided that it is technically feasible.

Claims

Demands

1. Calculation unit (50) for a hue determination device (14; 114) for a fluid distribution system (10), the device (14; 114) comprising: - at least one light source (22) configured to emit polychromatic or monochromatic light in the direction of the fluid in a measurement area (23);- a light sensor (24) configured to receive a light signal having been reflected on the fluid or transmitted through the fluid, the light signal corresponding to an optical reflection or respectively, to an optical transmission, by the fluid, of polychromatic or monochromatic light emitted in the direction of the fluid, the light signal being representative of a hue of the fluid circulating in the fluid distribution system (10), the light sensor (24) being further configured to emit quantities, each quantity being representative of a luminous intensity of the light signal for a given color channel; and - an acquisition optical fiber (36) configured to carry the light signal to the light sensor (24);the computing unit (50) comprising an estimation module (52) configured to receive the quantities emitted by the light sensor (24), characterized in that the estimation module (52) is further configured to determine a variable representing a hue of the fluid via a previously trained artificial intelligence model, each quantity representing a luminous intensity of the light signal for a given color channel being an input variable of the model, an output variable of the artificial intelligence model being the variable representing a hue of the fluid.

2. Calculation unit (50) according to claim 1, wherein the variable representing the color of the fluid is a purity index of the fluid circulating in the fluid distribution system (10).

3. Calculation unit (50) according to claim 2, wherein the purity index is a decimal number, preferably between 0 and 1.

4. Calculation unit according to claim 2, wherein the purity index is a binary number, the purity index then preferably being a class indicator among a first class corresponding to the case where the fluid (10) is clean, and a second class corresponding to the case where the fluid (10) is dirty.

5. Calculation unit (50) according to claim 1, wherein the variable representing the color of the fluid is a class indicator among a plurality of classes, each class corresponding to a predefined color of fluid.

6. Calculation unit (50) according to any one of the preceding claims, wherein the fluid is a coating liquid or a cleaning liquid or an aerosol composed of solid colored particles suspended in a gas.

7. A hue determination device (14; 114) for a fluid distribution system (10), comprising: - at least one light source (22) configured to emit polychromatic or monochromatic light in the direction of the fluid in a measurement area (23); - a light sensor (24) configured to receive a light signal having been reflected from or transmitted through the fluid, the light signal corresponding to an optical reflection or, respectively, to an optical transmission, by the fluid, of the polychromatic or monochromatic light emitted in the direction of the fluid, the light signal being representative of a hue of the fluid flowing in the fluid distribution system (10), the light sensor (24) being further configured to emit quantities, each quantity being representative of a light intensity of the light signal for a given color channel;- an optical acquisition fiber (36) configured to carry the light signal to the light sensor (24); and; - a calculation unit (50) according to any one of the preceding claims.

8. Device (14; 114) according to claim 7, wherein the given color channels are red, green, blue and clear channels, the quantities being representative of a luminous intensity of the light signal respectively for the red channel, the green channel, the blue channel and the clear channel.

9. Fluid distribution system (10) comprising a fluid flow circuit (12), and a shade determination device (14; 114) according to claim 7 or 8, connected in series to the fluid flow circuit (12).

10. A method for determining the color of a fluid circulating in a fluid distribution system (10), implemented by a color determination device (14; 114) according to claim 7 or 8, the method comprising the following steps: - emission (100) of polychromatic or monochromatic light towards the fluid by at least one light source (22); - reception (102), by the light sensor (24), of the light signal, reflected on the fluid or transmitted through the fluid; - reception (104), by the estimation module (52), of quantities representing a luminous intensity of the light signal for the given color channels; and - determination (106) by the estimation module (52) of the variable representing the color of the fluid.

11. A method for determining the color of fluid according to claim 10, further comprising the following steps: - detection (110) of a fluid change operation when the variable representing the color of the fluid is changed; and - once a fluid change operation has been detected, determination (112) of the end of the fluid change operation when the variable representing the color of the fluid is substantially constant for a predetermined period. 19 the fluid change operation being preferably a system flush or a change of color of fluid circulating in the system (10).

12. Method of determination according to claim 10 or 11, wherein the fluid is a coating liquid or a cleaning liquid.