Computational unit and associated hue determination device, fluid ejection system, and ejection method - Patents.com

Abstract

To determine the color of a fluid in a simpler and more reliable manner. [Solution] The present invention includes a computing unit (50) for an apparatus (14), the apparatus comprising at least one light source (22) configured to emit light toward a fluid, and an optical sensor (24) configured to receive an optical signal representing the hue of the fluid circulating in the fluid distribution system (10), the optical sensor (24) being further configured to emit a parameter representing the light intensity of the optical signal for a given color channel, the computing unit comprising an estimation module (52) configured to determine a variable representing the hue of the fluid via an artificial intelligence model, each parameter representing a light intensity that is an input variable of the model, and an output variable of the artificial intelligence model being a variable representing the hue of the fluid.

Description

[Technical Field]

[0001] The present invention relates to a computing unit and associated hue determining device, fluid ejection system, and ejection method. [Background technology]

[0002] In the context of coating product application equipment, it is known to determine the hue of a fluid circulating through a fluid dispensing system, such as the hue of a paint or cleaning fluid, to determine whether the correct hue of paint is circulating through the system or whether the cleaning fluid is pure, and to determine the cleanliness of the fluid dispensing system. For example, it is known to perform a visual check of the fluid sprayed by the dispensing system during a color change operation to ensure that the color change operation is complete and that the new color of paint is circulating through the system without mixing with the old color.

[0003] It is also known, for example from US Pat. No. 5,699,239 to use an optical sensor to measure the hue of a fluid. For this purpose, it is known to use an optical fiber that carries an optical signal to the optical sensor, the optical signal representing the hue of the fluid. However, the use of an optical fiber causes attenuation of the optical signal. The attenuation value depends on the length of the optical fiber and the wavelengths that make up the optical signal. Therefore, in order to accurately determine the hue of the fluid taking into account the attenuation caused by the optical fiber, it is necessary to manually calibrate the optical sensor, which is a long and complicated process. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] French Patent Application Publication No. 3127281 Summary of the Invention [Problem to be solved by the invention]

[0005] The object of the present invention is to overcome these drawbacks and to more simply and reliably determine the color of a fluid. [Means for solving the problem]

[0006] To this end, the present invention provides a computing unit for a hue determination device of a fluid ejection system, said device comprising: at least one light source configured to emit polychromatic or monochromatic light toward the fluid in the measurement region; an optical sensor configured to receive optical signals reflected on or transmitted through the fluid, the optical signals corresponding to the polychromatic light or the monochromatic light emitted towards the fluid reflected or transmitted by the fluid, the optical signals representing a hue of the fluid circulating within the fluid ejection system, the optical sensor further configured to emit parameters, each parameter representing a light intensity of the optical signal for a given color channel; an acquisition optical fiber configured to carry the optical signal to the optical sensor; The computing unit includes an estimation module configured to receive the parameters emitted by the optical sensor.

[0007] According to the present invention, the estimation module is further configured to determine a variable representative of the hue of the fluid via a pre-trained artificial intelligence model, each parameter representing a light intensity of the light signal for a given color channel being an input variable of the model, and an output variable of the artificial intelligence model being the variable representative of the hue of the fluid; The variable representing the hue of the fluid is: a purity index of the fluid circulating within the fluid delivery system; or Class indicators for a number of classes, each class corresponding to a given fluid hue.

[0008] According to the present invention, determining the color of a fluid does not require manual calibration of the optical sensor. Furthermore, by using an artificial intelligence model, reliable results can be obtained by limiting the complexity of resource- and time-consuming calculations performed without requiring additional optical signal processing steps. In this way, determining the color of a fluid becomes simpler for the user.

[0009] According to other advantageous aspects of the invention, the computing unit comprises one or more of the following characteristics, taken individually or in all technically possible combinations: The purity index is a decimal number, preferably between 0 and 1. The purity index is binary. In that case, the purity index is preferably a class indicator with a first class corresponding to when the fluid is clean and a second class corresponding to when the fluid is dirty. The fluid is a coating or cleaning liquid or an aerosol consisting of solid colored particles suspended in a gas.

[0010] The present invention also provides a hue determination apparatus for a fluid ejection system, the hue determination apparatus comprising: at least one light source configured to emit polychromatic or monochromatic light toward the fluid in the measurement region; an optical sensor configured to receive optical signals reflected on or transmitted through the fluid, the optical signals corresponding to the polychromatic light or the monochromatic light emitted towards the fluid reflected or transmitted by the fluid, the optical signals representing a hue of the fluid circulating within the fluid ejection system, the optical sensor further configured to emit parameters, each parameter representing a light intensity of the optical signal for a given color channel; an acquisition optical fiber configured to carry the optical signal to the optical sensor; The present invention also relates to a hue determining device comprising the above-mentioned arithmetic unit.

[0011] Advantageously, the given color channels are a red channel, a green channel, a blue channel, and a clear channel, and the parameters represent the light intensities of the light signal for the red channel, the green channel, the blue channel, and the clear channel, respectively.

[0012] The present invention also relates to a fluid ejection system, the fluid ejection system comprising a fluid flow circuit and a hue determining device as described above connected in series with the fluid flow circuit.

[0013] The present invention also provides a method for determining the hue of a fluid circulating in a fluid ejection system implemented by the above-described hue determination apparatus, the method comprising: emitting polychromatic or monochromatic light toward the fluid by the at least one light source; receiving, by the optical sensor, the optical signal reflected on or transmitted through the fluid; receiving, by the estimation module, the parameter representing the light intensity of the light signal for the given color channel; determining, by said estimation module, said variable representing said hue of said fluid.

[0014] According to another advantageous aspect of the invention, the method for determining a hue comprises one or more of the following characteristics, taken individually or in all technically possible combinations: said determining method comprising: detecting a fluid alteration operation when the variable representing the hue of the fluid changes; determining an end of the fluid change operation when a fluid change operation is detected and the variable representing the hue of the fluid has remained substantially constant for a predetermined period of time; The fluid change operation is preferably a rinsing operation of the system or a change in the color of the fluid circulating in the system. the fluid is a coating or cleaning liquid;

[0015] The invention will appear more clearly on reading the following description, given only as a non-limiting example and made with reference to the drawings in which: [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a diagram of a fluid ejection system according to the present invention. [Figure 2] FIG. 2 is a cross-sectional perspective view of a portion of the fluid ejection system of FIG. 1 corresponding to Detail A of that figure. [Figure 3] FIG. 3 is a flow chart of the method according to the present invention. [Figure 4] FIG. 4 is a perspective view of a portion of a fluid ejection system according to a second embodiment of the present invention, the portion corresponding to Detail A of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0017] The present invention will be described below in the context of a coating fluid distribution system, such as a paint, used in particular in spray coating product application equipment. The paint may be, for example, presented in liquid form or in the form of an aerosol, for example, consisting of solid colored particles suspended in a gas. However, this does not constitute a limitation of the present invention to a particular application; the present invention can be implemented in any fluid distribution system adapted to convey a fluid, where it is advantageous to identify the color of the fluid, and in particular to supervise an operation or phenomenon involving a change in the color of the fluid. The fluid may be liquid or gaseous; in particular, in the following description, the term fluid includes solid particles transported by a gas flow and suspended in the gas, in particular in an air flow. The solid particles may be, for example, colored powders.

[0018] In the following description, the expression "substantially equal" defines a relationship equal to + / - 10%, more preferably + / - 5%.

[0019] 1 is a diagram of a fluid distribution system 10, such as a paint distribution system, that includes at least one paint component reservoir, an injector 11, also called an atomizer, and a fluid flow circuit 12.

[0020] In the example of Figure 1, the fluid distribution system 10 comprises three reservoirs 1, 2 and 3 of paint of different hues and a reservoir 4 of a cleaning liquid, for example water-based, also called solvent. The reservoirs 1 to 4 are grouped together in a dedicated compartment 6, often called a paint compartment. The number and distribution of the reservoirs 1 to 4 in the paint compartment are not limiting; they are adapted according to the intended use of the fluid distribution system 10.

[0021] The color change device 8 is supplied by different reservoirs 1-4 and allows selection of which product, paint or cleaning fluid is circulated in the fluid flow circuit 12 to supply the injector 11.

[0022] The injector can be constituted, for example, by a manual spray gun, as shown in Figure 1. In a variant not shown, it is an automatic pneumatic or rotary injector mounted on the arm of a multi-axis or reciprocating robot.

[0023] Advantageously, the injector 11 may be of the electrostatic type.

[0024] The fluid flow circuit 12 comprises pipes 122 and 124 and places the colour modifier 8, and thus one of the reservoirs 1-4, in fluid communication with the injector 11 via the pipes 122 and 124.

[0025] The fluid dispensing system 10 further includes a hue determiner 14 .

[0026] A portion of the color determining device 14 is connected in series to the fluid flow circuit 12 between the pipes 122 and 124 .

[0027] Advantageously, the hue determining device 14 comprises a flange 13, which is advantageously opaque, as can be seen in Figure 2, defining a chamber 15 within which is located a fluid flow path 16. The flange 13 is arranged to be connected on both sides to the pipes of the fluid flow circuit 12. The fluid flow path 16 is indicated in Figure 2 by an arrow.

[0028] The hue determining device 14 comprises at least one light source 22. It further comprises a light sensor 24.

[0029] In the example of FIG. 1, the hue determining device 10 includes only one light source 22. The light source 22 is configured to emit polychromatic light, i.e., light containing at least two different wavelengths, or monochromatic light, i.e., light containing a single wavelength. Polychromatic light preferably consists of white light. The light source 22 may, for example, be formed from several monochromatic light sources of different wavelengths, e.g., from a red laser source, a green laser source, and a blue laser source, or from several light-emitting diodes or LEDs of different colors. Alternatively, the light source 22 may be a light source emitting a continuous light spectrum, such as a filament lamp. Furthermore, the light source may be configured to emit pulsed light.

[0030] Alternatively, if light source 22 is configured to emit monochromatic light, light source 22 may be formed, for example, from one or more monochromatic light sources of the same wavelength, for example, from lasers or monochromatic LEDs.

[0031] The light source 22 is configured to emit light toward the fluid, in this case, toward the paint, within a measurement region 23 formed within the chamber 15, shown by the dotted line in Figure 2. In particular, the light illuminates the fluid present within the flow path 16 within the measurement region 23.

[0032] The optical sensor 24 is configured to receive the optical signal and generate parameters, each parameter representing the light intensity of the optical signal for a given color channel. The given color channels may be, for example, a red channel, a green channel, and a blue channel. Alternatively, the given color channels may be a red channel, a green channel, a blue channel, and a clear channel, with the clear channel representing the light intensity of the optical signal taken as a whole. In this case, the parameters generated by the optical sensor 24 represent the light intensity of the optical signal for each of the red channel, the green channel, the blue channel, and the clear channel. In other words, the optical sensor 24 is configured to generate parameters representing the light intensity in an RVBC or red-green-blue-clear system.

[0033] The light intensity is, for example, indicated in a known manner by a number between 0 and 255, 0 corresponding to zero intensity for the channel considered and 255 corresponding to maximum intensity for the channel considered.

[0034] It is possible to position the optical sensor 24 and / or light source 22 close to the measurement area 23. However, the quality of the fluid being dispensed or its environment may often lead to constraints of compliance with regulations for use in explosive atmospheres, such as in the context of paint dispensing systems, of the ATEX type in Europe ("ATmosphere EXplosive"). However, the use of an optical sensor 24 and / or light source 22 that meets this constraint may prove costly, and furthermore, their incorporation into a color measurement device may be complicated or even impossible.

[0035] Thus, the light source 22 and the light sensor 24 are preferably positioned at a distance from the fluid flow circuit 12, for example by being grouped together in a processing box 26 that also belongs to the color determining device 14.

[0036] To ensure that the light emitted by the light source 22 illuminates the fluid in the measurement region 23, the device 14 comprises one or more optical fibers 34, called illumination optical fibers. The illumination optical fibers 34 are advantageously four in number, as can be seen in FIG. 2. The illumination optical fibers 34 are generally several meters long, for example at least 10 meters, and preferably they are up to 25 meters long. The illumination optical fibers 34 are advantageously connected at one of their ends to the light source 22. The other end of the illumination optical fiber 34 is advantageously arranged in the flange 13.

[0037] In a variant not shown, if the device 10 comprises several light sources 22 , each illumination optical fiber 34 is connected to one of the light sources 22 .

[0038] The device 14 further comprises an acquisition optical fiber 36, which is advantageously connected by one of its ends to the optical sensor 24, the other end of which is arranged in the flange 13. In this way, the optical signal from the measurement area 23 is conveyed via the optical fiber 36 to the optical sensor 24. The acquisition optical fiber 36 is of the same length as the illumination optical fiber 34.

[0039] Advantageously, means for holding the end(s) of the illumination optical fiber 34 and the acquisition optical fiber 36 are arranged within the flange 13, for example a cable gland 38.

[0040] Advantageously, the device 14 further comprises an observation window 40 arranged in the flange 13 tangentially to the fluid flow path 16. The observation window 40 may in particular have the shape of a porthole, as can be seen in Figure 2. In the example of Figure 2, the measurement area 23 corresponds to the part of the fluid flow path 16 that is visible through the observation window 40.

[0041] The device 14 advantageously includes a retaining element 42 that is attached to the flange 13 and holds the observation window 40 in place and ensures a seal of the fluid flow path 16. The retaining element 42 may, for example, include a sealing gasket that surrounds the observation window 40.

[0042] The device 14 further comprises a computing unit 50, which is connected to the light source 22 and the light sensor 24 and which is formed, for example, by a memory unit and a processor associated with the memory unit, not shown. The computing unit 50 is advantageously arranged inside the processing box 26.

[0043] The computing unit 50, as can be seen in Fig. 1, includes an estimation module 52. In the example of Fig. 1, the estimation module 52 is implemented as software or software bricks executable by a processor, and a memory unit of the computing unit 50 can store the estimation software, and the processor can execute the estimation software.

[0044] The estimation module 52 is configured to receive the parameters emitted by the light sensor 24 and then determine, via a pre-trained artificial intelligence model, a variable representing the hue of the fluid, the variable representing the hue of the fluid having, as an input variable, each parameter representing the light intensity of the light signal for a given color channel, and an output variable representing the hue of the fluid. The variable representing the hue of the fluid represents the color of the fluid, or alternatively or additionally, represents the opacity of the fluid. In particular, if the light emitted by the light source 22 is monochromatic, the variable representing the hue of the fluid represents the opacity of the fluid. In particular, if the light emitted by the light source 22 is polychromatic, the variable representing the hue of the fluid represents the opacity and / or hue of the fluid.

[0045] The variable representing the color of the fluid is, for example, a purity index of the fluid circulating within the fluid distribution system 10. According to this example, the purity index is a decimal number, typically between 0 and 1, or is binary, typically equal to 0 or 1. If the purity index is binary, the purity index is advantageously a class indicator, with a first class when the fluid 10 is clean and a second class when the fluid 10 is dirty.

[0046] According to one example, when the purity index is a decimal number, it is directly proportional to the opacity of the fluid.

[0047] Alternatively, the variable representing the hue of the fluid is a class indicator of a number of classes, each class corresponding to a given hue of the fluid.

[0048] In a variant 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 Specific Integrated Circuit).

[0049] When the computing unit 50 is implemented as one or more software programs, i.e., as a computer program, also called a computer program product, it can be stored on a computer-readable medium (not shown). The computer-readable medium is, for example, a medium capable of storing electronic instructions and coupled to a bus of a computer system. For example, the readable medium can be an optical disk, a magneto-optical disk, a ROM-type memory, a RAM-type memory, any type of non-volatile memory (e.g., FLASH or NVRAM), or a magnetic card. The computer program, including the software instructions, is then stored on the readable medium.

[0050] The device 14 advantageously includes a human-machine interface 54 connected to the computing unit 50 for displaying information for the user's attention. Alternatively or additionally, the human-machine interface 54 is used by a user to control the device 14. The human-machine interface 54 typically includes a display screen and, optionally, input devices, not shown, such as a keyboard and mouse.

[0051] The artificial intelligence model is pre-trained, advantageously having been pre-trained, for example by machine learning, via supervised learning as known per se.

[0052] Training of an artificial intelligence model is typically performed from training data sets, each of which includes input training data corresponding to the model's input variables, i.e., parameters representing light intensity for a given color channel, and output training data corresponding to expected fluid hue representative variables, i.e., targets, for the input training data. The model is then trained, for example, via backpropagation of an error gradient, where the error is calculated from the difference between the output training data and the output variables estimated by the model from the input training data, i.e., the difference between the target representative variables and the estimated representative variables for the input training data.

[0053] The artificial intelligence model is for example a random forest, a support vector machine or SVM model, or a neural network, or an ANN (artificial neural network).

[0054] In the case of a neural network, it comprises an ordered series of neuron layers, each of which receives its input from the output of the previous layer.

[0055] More precisely, each layer contains neurons that receive their input from the output of neurons in the previous layer, or, for the first layer, from input variables.

[0056] Alternatively, more complex neural network structures can be thought of as having layers that can be connected to layers further away than the previous layer.

[0057] Each neuron is also associated with an operation, ie, the type of processing to be performed by the neuron in the corresponding processing layer.

[0058] Each layer is connected to other layers by multiple synapses. A synaptic weight is associated with each synapse, which forms a connection between two neurons. It is often a real number that can be positive or negative. In some cases, the synaptic weight is a complex number.

[0059] Each neuron can perform a weighted sum of the values ​​received from the neurons of the previous layer, where each value is multiplied by the respective synaptic weight or connection of each synapse between it and the neuron of the previous layer, then apply an activation function, typically a nonlinear function, to the weighted sum, and at the output of the neuron, deliver the value resulting from the application of the activation function, in particular to the neuron of the next layer to which it is connected. The activation function makes it possible to introduce nonlinearity into the processing performed by each neuron. The sigmoid function, the hyperbolic tangent function, and the Heaviside function are examples of activation functions.

[0060] Optionally, each neuron may also apply a multiplication factor, also called a bias, to the output of the activation function, in which case the value delivered at the neuron's output is the product of the bias value and the value from the activation function.

[0061] A method for determining the color of a fluid circulating in flow circuit 12 will now be described. This method is implemented by device 14.

[0062] A fluid, such as a paint or aerosol, circulates within the dispensing system 10, and in particular within the fluid flow path 16 of the device 14. The fluid may be colored, opaque, or in some cases translucent or even transparent.

[0063] During the light emitting step 100, the light source 22 emits polychromatic or monochromatic light towards the fluid and into the measurement area 23. In particular, the light is emitted through the illumination optical fibers 34 towards their ends located within the flange 13. In this way, the fluid circulating in the measurement area 23 is illuminated by the light passing through the observation window 40.

[0064] The light reflected by the fluid forms an optical signal that corresponds to the reflection by the fluid of the light emitted by the light source 22. The optical signal represents the color of the fluid.

[0065] The optical signal circulates from the measurement region 23 through the acquisition optical fiber 36 to the optical sensor 24, which receives the optical signal during a receiving step 102. The optical signal received by the optical sensor 24 is attenuated by the optical fibers 34 and 36, which cause an attenuation that depends on their length and on the wavelengths that make up the light, on the one hand, and the optical signal, on the other hand. More precisely, the light is attenuated when it circulates in the illumination optical fiber 34, and the optical signal is attenuated when it circulates in the acquisition optical fiber 36.

[0066] The opaque flange 13 advantageously makes it possible to strongly limit any light contamination that may arise from the external environment, so that the light signal received by the light sensor 24 during the receiving step 102 is as reliable as possible with minimal external light contamination.

[0067] The light sensor 24 converts the received light signal into parameters, each parameter representing the light intensity of the light signal for a given color channel. In the example, the light sensor 24 converts the light signal into four numbers between 0 and 255, which correspond to the light intensity for each channel in the RVBC system.

[0068] The computational unit 50, and in particular the estimation module 52, receives the parameters during a receiving step 104.

[0069] During determining step 106, the estimation module 52 determines a variable representing the hue of the fluid via an artificial intelligence model. Each parameter 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 a variable representing the hue of the fluid.

[0070] Advantageously, once the variable representative of the fluid hue has been determined by the estimation module 52, during a displaying step 108 a message representative of the variable representative of the fluid hue is displayed in visual and / or possibly audio form on the human machine interface 54. Alternatively, during the displaying step 108 the computing unit 50 transmits the message representative of the variable representative of the fluid hue to a remote terminal, such as a computer or a mobile phone of the user constituting the human machine interface 54.

[0071] Advantageously, the method is implemented by the device 14 to perform any of the following three tasks:

[0072] The first task is to estimate the purity of the fluid circulating within the fluid distribution system 10. Such a task may be performed, for example, by a cleaning fluid during a rinsing process of the system 10. In this case, the variable representing the fluid hue is a fluid purity index, which is in the form of a decimal. For example, the fluid purity index may be a decimal between 0 and 1, or between 0 and 100. The fluid purity index corresponds to the purity of the fluid circulating within the system 10. During the rinsing operation, the system 10 is initially dirty, and the variable representing the fluid hue will be, for example, substantially equal to zero. Then, as the rinsing process progresses, if the hue determination method is repeated over time, the value of the variable representing the fluid hue will increase, for example, to approximately equal 1 or 100 if the fluid is pure, corresponding to a completely clean system 10.

[0073] The second task is a binary classification of the purity of the fluid. In this case, the variable representing the color of the fluid is a fluid purity index, which is in binary form. Preferably, the purity index is a class indicator, with a first class corresponding to when the fluid is clean and a second class corresponding to when the fluid is dirty. This task is performed, for example, during a rinse process of the system 10 to determine whether the system 10 is clean.

[0074] The fluid purity index makes it possible in particular to determine whether the system 10 is clean or dirty, for example during a rinsing operation. Indeed, if the fluid is considered to be dirty, in other words if the purity index determined by the model indicates a class corresponding to a dirty fluid, the system 10 is considered to be dirty, and if the purity index determined by the model indicates a class corresponding to a cleaning fluid, the system 10 is considered to be clean.

[0075] Advantageously, for the first and second tasks, the variable representing the hue of the fluid represents the opacity of the fluid, and if the fluid is opaque it is considered dirty and if it is transparent it is considered clean.

[0076] The third task is to determine the hue of the fluid from a palette of predetermined hues. In this case, the variable representing the hue of the fluid represents the color of the fluid. The variable representing the hue of the fluid is then a class indicator of a number of classes, each class corresponding to a predetermined hue of the fluid. This task may be performed, for example, to verify that the hue of the fluid circulating within system 10 is, in fact, the expected hue.

[0077] Advantageously, if computing unit 50 is configured to perform several of the above three tasks, a separate model trained separately is used for each task. For example, a user indicates which task they want device 14 to perform, and computing unit 50 then uses the associated model.

[0078] Advantageously, if the 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 assess the opacity of the fluid and therefore to reliably determine, for the model, the variable representing the hue of the fluid. The model training is then also performed using monochromatic light.

[0079] Advantageously, the determination method is performed continuously, with successive values ​​of the variable representative of the color of the fluid being stored in the memory of the computing unit 50.

[0080] Optionally, during the detecting step 110, the computing unit 50 detects a fluid change operation when the variable representing the fluid hue changes compared to a previously determined variable representing the fluid hue, in particular when the variation of the variable representing the fluid hue is greater than a predetermined threshold. If the variable representing the fluid hue is a class indicator, the variable representing the fluid hue is considered to have changed when the class it indicates changes. If the variable representing the fluid hue is binary, the variable representing the fluid hue is considered to have changed when the value of the binary value changes, or alternatively, in the case of a decimal, when the value changes by at least 10%. Advantageously, the user selects the variation required for the variable representing the fluid hue to be considered to have changed. The fluid change operation is advantageously a rinsing process or a change in the hue of the fluid circulating in the system 10.

[0081] Advantageously, when a fluid change operation is detected in the detecting step 110 and the continuous variable representing the fluid hue has remained substantially constant for a predetermined period of time, the computing unit 50 determines that the hue change operation or rinsing operation is complete during the determining step 112.

[0082] Where the variable representing the fluid hue is a decimal number, substantially constant means that the variable representing the fluid hue as continuously determined by the artificial intelligence model varies by less than 10% over a predetermined period of time, preferably less than 5% over a predetermined period of time. Where the variable representing the fluid hue is a binary or class indicator, substantially constant means that the variable representing the fluid hue as continuously determined by the artificial intelligence model remains the same, i.e., does not change.

[0083] When the variable representing the fluid hue remains substantially constant over a predetermined period of time, it means that the hue of the fluid circulating within the system 10 has stabilized. In the case of a hue change, this means that the desired hue has been achieved, and in the case of a rinsing process, it means that the system has been fully rinsed. Advantageously, the computing unit 52 commands a specific display on the human-machine interface 54 or the transmission of a specific message to a remote terminal when the fluid change operation is complete.

[0084] The predetermined period used to consider a variable stable is, for example, between 0.5 and 5 seconds.

[0085] Alternatively or additionally, device 14 is configured to transmit a variable representative of the hue of the fluid to a control unit configured to control a valve or set of valves of dispensing system 10 based on the variable representative of the hue of the fluid. In this way, it is possible, for example, to optimize the hue changing operation or to adjust the composition of the fluid to obtain a desired hue.

[0086] In addition to the above explanation regarding the learning or training of the model, the model may be learned, for example, in the form of supervised learning. The model receives as input data including sequences formed from parameters representing the light intensity of an optical signal for a given color channel. The sequences correspond to data provided by an optical sensor belonging to a hue determination device including a length of illumination optical fiber and an acquisition optical fiber. Each sequence is associated with a class or a decimal depending on the task for which the model is trained. The model is then trained to make predictions until satisfactory performance is achieved. Advantageously, a model validation operation is performed.

[0087] This model is then incorporated into the estimation unit of the device 14, whose optical fibers 34 and 36 advantageously have the same length as that used to train the model, thereby allowing the artificial intelligence model to obtain more reliable predictions.

[0088] Alternatively, the artificial intelligence model is trained by reinforcement learning. Thus, while the model is being used to determine the fluid's hue, a user can, for example, indicate 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, allowing the model's predictions to be refined.

[0089] 4 shows the flange 13 of a color determination device 114 according to a second embodiment of the invention. The color determination device 114 is configured to perform color measurements by transmission of an optical signal through a fluid. In this case, the flange 13 of the color measurement device 114 advantageously comprises, on the one hand, two recesses 138 for inserting the ends of the illumination optical fibers 34 and, on the other hand, the acquisition optical fibers 36 and two observation windows 140 arranged tangentially to the fluid flow path 16 across the measurement region 23, each observation window being located at the bottom of one of the recesses 138. Advantageously, the observation windows 140 face each other.

[0090] The method of determining the fluid remains unchanged, except that the light signal corresponds to the transmission by the fluid of light emitted by the light source 22 towards the fluid.

[0091] Any feature described in one embodiment or variant above may be implemented in any other embodiment or variant described above, to the extent technically possible.

Claims

1. A computing unit (50) for a hue determination device (14, 114) of a fluid distribution system (10), said device (14, 114) comprising: at least one light source (22) configured to emit polychromatic or monochromatic light towards the fluid in the measurement area (23); an optical sensor (24) configured to receive optical signals reflected on or transmitted through the fluid, the optical signals corresponding to the polychromatic or monochromatic light emitted towards the fluid reflected or transmitted by the fluid, the optical signals representing a hue of the fluid circulating within the fluid distribution system (10), the optical sensor (24) being further configured to emit parameters, each parameter representing a light intensity of the optical signal for a given color channel; an acquisition optical fiber (36) configured to carry the optical signal to the optical sensor (24); The computing unit (50) comprises an estimation module (52) configured to receive the parameters emitted by the optical sensor (24), the estimation module (52) is further configured to determine a variable representing the hue of the fluid via a pre-trained artificial intelligence model, each parameter representing a light intensity of the light signal for a given color channel being an input variable of the model, and an output variable of the artificial intelligence model being the variable representing the hue of the fluid; The variable representing the hue of the fluid is: the purity index of the fluid circulating in the fluid distribution system (10), or A computing unit (50) characterized by class indicators for a plurality of classes, each class corresponding to a predetermined fluid hue.

2. The computing unit (50) of claim 1, wherein the purity index is a decimal number.

3. 2. The computing unit of claim 1, wherein the purity index is comprised between 0 and 1.

4. The computing unit of claim 1 , wherein the purity index is binary.

5. 2. The computing unit of claim 1, wherein the purity index is a class indicator of a first class corresponding to when the fluid (10) is clean and a second class corresponding to when the fluid (10) is dirty.

6. 2. The computing unit (50) of claim 1, wherein the fluid is a coating or cleaning liquid, or an aerosol consisting of solid colored particles suspended in a gas.

7. A hue determining device (14, 114) for a fluid distribution system (10), the hue determining device (14, 114) comprising: at least one light source (22) configured to emit polychromatic or monochromatic light towards the fluid in the measurement area (23); an optical sensor (24) configured to receive optical signals reflected on or transmitted through the fluid, the optical signals corresponding to the polychromatic or monochromatic light emitted towards the fluid reflected or transmitted by the fluid, the optical signals representing a hue of the fluid circulating within the fluid distribution system (10), the optical sensor (24) being further configured to emit parameters, each parameter representing a light intensity of the optical signal for a given color channel; an acquisition optical fiber (36) configured to carry the optical signal to the optical sensor (24); A hue determination device (14, 114) comprising a calculation unit (50) according to any one of claims 1 to 6.

8. 8. The apparatus (14, 114) of claim 7, wherein the given color channels are a red channel, a green channel, a blue channel, and a clear channel, and the parameters represent light intensities of the optical signal for the red channel, the green channel, the blue channel, and the clear channel, respectively.

9. A fluid distribution system (10) comprising a fluid flow circuit (12) and a hue determination device (14, 114) according to claim 7 connected in series to the fluid flow circuit (12).

10. 10. A method for determining the hue of a fluid circulating in a fluid distribution system (10) implemented by a hue determination device (14, 114) according to claim 7, said method comprising: emitting (100) polychromatic or monochromatic light toward the fluid by the at least one light source (22); receiving (102) the optical signal reflected on or transmitted through the fluid by the optical sensor (24); receiving (104) by the estimation module (52) the parameter representing the light intensity of the light signal for the given color channel; determining (106) the variable representing the hue of the fluid by the estimation module (52).

11. The determination method includes: detecting (110) a fluid change operation when the variable representing the hue of the fluid changes; and determining (112) that a fluid change operation has ended when a fluid change operation is detected and the variable representing the color of the fluid has remained substantially constant for a predetermined period of time; The method of claim 10, wherein the fluid change operation is preferably a rinsing operation of the system or a change in the color of the fluid circulating in the system (10).

12. The method of claim 10 , wherein the fluid is a coating liquid or a cleaning liquid.

Images