Method for determining a condition of calcification in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use

Abstract

A method for determining a condition of calcification and / or accumulation of limescale or other scale in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use is described. The method first comprises a step of detecting at least one value of each of a plurality of parameters related to a thermal response of the thermo-hydraulic system. Such parameters comprise a static gain Kp, a delay time θ between electric heating power supplying and temperature response of the thermo-hydraulic system, and a rise time 휏 of the aforesaid temperature response. The method further includes providing the detected values of each parameter of the aforesaid plurality of parameters as input data to a trained algorithm or artificial intelligence model. The algorithm or model was trained by means of a plurality of input training data. The input training data comprise data representative of values of the aforesaid parameters at each of a plurality of known calcification situations, and data representative of the respective known calcification conditions. The method further comprises the steps of processing the aforesaid input data, by the trained algorithm or model, to obtain output information, representative of the calcification condition; and finally determining the aforesaid condition of calcification and / or accumulation of limescale or other scale based on the aforesaid output information provided by the trained algorithm or model.

Description

[0001] “Method for determining a condition of calcification in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use”

[0002] DESCRIPTION

[0003] TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0004] Field of application.

[0005] The present invention relates to a method for determining a condition of calcification in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use.

[0006] The method employs trained algorithms implemented by electronic processing.

[0007] Description of the prior art.

[0008] In the technical field of beverage preparation machines, both of the professional and household type, the phenomenon of progressive calcification which occurs in the conduit, or hydraulic circuit or thermo-hydraulic system, in which the liquid (water, for example) required for preparing the beverage flows and is heated, is well known.

[0009] Such a phenomenon results in problems related to the worsening of the performance of the beverage preparation machine, which problems are critical and strongly perceived in the technical field considered.

[0010] For this reason, the need to provide methods for estimating the calcification that are quick and simultaneously very accurate and reliable, such as to be capable of being performed even during the machine operation and not require repeated physical measurements of the calcification (which can be performed only under “out of service” conditions of the machine itself) is also felt.

[0011] In this respect, the prior art suggests some solutions, including:

[0012] - calcification estimate based on flow variations measured in the heating tube;

[0013] - calcification estimate based on static comparisons of a temperature detected in the hydraulic circuit in predefined positions with respect to one or more predefined thresholds;

[0014] - calcification estimate based on a plurality of static temperature measurements in different points of the heating conduit of the liquid.

[0015] However, as far as the Applicant is aware, the calcification estimates provided by the known solutions do not provide a sufficiently accurate and reliable prediction of the actual state of calcification.

[0016] Moreover, the processes leading to such estimates are relatively slow.

[0017] In addition to the above, the known solutions suffer from the fact that the calcification occurs also on the sensors, e.g., temperature sensors, that are used for measuring the parameters on which the estimate is based.

[0018] Due to this fact, it is difficult to discriminate the effects due to the limescale deposited in the tube or hydraulic circuit from the effects due to the limescale deposited on the sensor, accordingly making the estimate or determination of the calcification state of the thermo-hydraulic system through which the liquid passes not very accurate, and thus determining further problems of lack of accuracy of the quantity that is to be actually estimated (i.e., the calcification of the thermo-hydraulic system and not the calcification of the sensor).

[0019] Therefore, at the moment, the need is not completely met for an estimate of the calcification state in a thermo-hydraulic system of a beverage preparation machine which, albeit is indirect because it is based on the measurement of physical parameters, is sufficiently accurate, reliable and fast in relation to the requirements of use.

[0020] SUMMARY OF THE INVENTION

[0021] It is the object of the present invention to provide a method for determining a condition of calcification and / or accumulation of limescale or other scale in a thermo- hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use which allows at least partially obviating the above drawbacks with reference to the prior art and meeting the aforesaid needs particularly felt in the technical field considered. Such an object is achieved by a method according to claim 1.

[0022] Some further embodiments of such a method are defined by claims 2-22.

[0023] BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Further features and advantages of the method according to the invention will become apparent from the following description of preferred embodiments, given by way of non-limiting indication, with reference to the accompanying figures, in which:

[0025] - Figure 1 shows a portion of a heating tube of a thermo-hydraulic system of a beverage preparation machine in which a calcification phenomenon may occur, and a temperature sensor arranged therein, i.e., a situation to which the method according to the invention is applicable;

[0026] - Figure 2 shows exemplary time diagrams of a dynamic temperature response which provides information used in the method of the present invention;

[0027] - Figure 3 shows a comparison of a measured temperature response with an estimated temperature response according to a model used in an embodiment of the method according to the invention; - Figure 4 shows a series of diagrams showing the value of thermal parameters considered in the method;

[0028] - Figure 5 shows the trend of a further “pump command” parameter, used in an embodiment of the method;

[0029] - Figure 6 shows a classification of situations corresponding to the presence or absence of a calcification state, determined by applying the method according to an embodiment;

[0030] - Figures 7 and 8 show charts representing the probability of the presence of a calcification state, obtained as the result of an embodiment of the method;

[0031] - Figures 9A and 9B show waveforms of signals that can be used as input for test cycles, provided in an embodiment of the method;

[0032] - Figure 10 shows different possible locations / arrangements of a temperature sensor according to different implementation options of the method;

[0033] - Figure 11 shows a simplified diagram of an example of beverage preparation machine, and related thermo-hydraulic system and heating tube, to which the method of the present invention is applicable.

[0034] DETAILED DESCRIPTION

[0035] A method for determining a condition of calcification and / or accumulation of limescale or other scale in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use is described.

[0036] The method first comprises a step of detecting at least one value of each of a plurality of parameters relating to a thermal response of the thermo-hydraulic system. Such parameters comprise a static gain Kp, a delay time 0 between electric heating power supplying and temperature response of the thermo-hydraulic system, and a rise time T of the aforesaid temperature response.

[0037] The method then comprises providing the detected values of each parameter of the aforesaid plurality of parameters as input data to a trained algorithm or artificial intelligence model, and / or of “machine learning” type. Such an algorithm or model was trained by means of a plurality of input training data. The input training data comprise data representative of values of the aforesaid parameters at each of a plurality of known calcification situations, and further comprise data representative of the respective known calcification conditions.

[0038] The method further comprises the steps of processing the aforesaid input data, by the trained algorithm or model, to obtain output information, representative of the calcification condition; and finally determining the aforesaid condition of calcification and / or accumulation of limescale or other scale based on the aforesaid output information provided by the trained algorithm or model.

[0039] As shown above, the method thus provides the determination of the calcification condition to be performed based on the measurement and / or monitoring and / or estimate of at least three parameters related to the dynamic thermal response of the thermo- hydraulic system (i.e., as shown above, static gain Kp, delay time 0 between electric heating power supplying and thermo-hydraulic system temperature response, and rise time T of the temperature response). As described in greater detail below, such parameters are capable of defining the dynamic thermal response of the thermo-hydraulic system in sufficiently accurate and efficient manner.

[0040] According to an embodiment, the method includes further detecting at least one value of an additional parameter. Such an additional parameter relates to a hydraulic resistance of the aforesaid thermo-hydraulic system.

[0041] In such an embodiment, the method further comprises providing also the aforesaid at least one detected value of the additional parameter, in addition to the detected values of the plurality of parameters relating to the thermal response, as input to the trained algorithm or model; and processing such input data, by the trained algorithm or model, to obtain output information representative of the calcification condition based on both the values of the three aforesaid parameters relating to the thermal response and the detected value of the additional parameter related to hydraulic resistance and / or liquid flow.

[0042] The method finally comprises determining the condition of calcification and / or accumulation of limescale or other scale based on the output information from the trained algorithm or model.

[0043] Therefore, in this embodiment, the determination of the calcification condition depends on four parameters, three of which related to the dynamic thermal response of the thermo-hydraulic system (i.e., as shown above, static gain Kp, delay time 0 between electric heating power supplying and temperature response of the thermo-hydraulic system, and rise time T of the temperature response), and one parameter correlated with the hydraulic resistance experimented by the liquid flow passing through the thermo- hydraulic system under a given condition, and therefore related, ultimately, to such a flow.

[0044] According to an embodiment of the method, the aforesaid thermo-hydraulic system comprises a hydraulic circuit in turn comprising, for example, a heating tube or conduit, intended both to convey and heat the liquid.

[0045] According to an embodiment, the aforesaid additional parameter is a pump command and / or control signal of the thermo-hydraulic system, in which such a pump command and / or control signal is representative of the hydraulic resistance of the thermo- hydraulic system or of the respective hydraulic circuit.

[0046] According to an implementation option, the aforesaid additional parameter is a pump command and / or control signal representative of the liquid flow in the thermo- hydraulic system or in the related hydraulic circuit.

[0047] According to an embodiment of the method, the aforesaid parameters relating to a thermal response of the thermo-hydraulic system are detected by means of one or more temperature sensors operatively connected to the thermo-hydraulic system.

[0048] According to an implementation option of such an embodiment, the aforesaid parameters related to a thermal response of the thermo-hydraulic system are detected by means of a single temperature sensor operatively connected to the thermo-hydraulic system.

[0049] According to an implementation option, the aforesaid temperature sensor is operatively connected to a hydraulic circuit or a hydraulic channel or heating tube of the aforesaid thermo-hydraulic system.

[0050] According to an implementation option, such a thermal sensor is arranged just downstream or at the output of the heating tube.

[0051] Further details of this or other possible arrangements of the temperature sensor are provided in a following part of this description.

[0052] According to an embodiment, the method applies to a professional or household beverage preparation machine, preferably a coffee preparation machine. In this latter case, the aforesaid thermo-hydraulic system is a thermo-hydraulic system of the coffee preparation machine.

[0053] According to an implementation option, the aforesaid heating tube is a heating tube of the beverage preparation machine.

[0054] According to an embodiment of the method, the aforesaid liquid passing through the thermo-hydraulic system (or hydraulic circuit or heating tube) is water.

[0055] According to other embodiments, the aforesaid liquid passing through the thermo-hydraulic system is an infusion or extract, for example, such as tea, coffee, herb tea, milk or other.

[0056] According to an embodiment, the aforesaid thermo-hydraulic system comprises a heating tube. In this case, the aforesaid dynamic temperature response parameters characterize the dynamic temperature response of the heating tube.

[0057] According to an implementation option, the heating tube acts both as heater and as conductor of the liquid.

[0058] According to another implementation option, the aforesaid heating tube acts as conductor of the liquid and is operatively connected to the heater device adapted to heat it.

[0059] According to an embodiment of the method, the aforesaid detection step comprises taking a plurality of measurements of each of said parameters within a predefined time interval.

[0060] In this case, the method comprises the further steps of estimating a statistical value of each of the aforesaid parameters based on the aforesaid plurality of measurements, and providing, as input data, the estimated statistical values for each of the aforesaid parameters to the trained algorithm.

[0061] According to an embodiment of the method, the aforesaid output information, representative of the calcification condition, comprises a level of calcification or accumulation of limescale or other scale.

[0062] According to another embodiment of the method, the aforesaid output information, representative of the calcification condition, comprises a probability of presence or absence of a calcification state with respect to an operating threshold according to a predefined criterion.

[0063] According to another embodiment of the method, the aforesaid output information, representative of the calcification condition, comprise the determination of the presence or absence of a state of calcification or accumulation of limescale or other scale, identified according to a predefined criterion.

[0064] According to an embodiment of the method, the aforesaid step of determining the calcification condition comprises determining a calcification state.

[0065] In this case, the processing step comprises identifying the presence of a calcification state based on a comparison of the detected or estimated values of the aforesaid parameters with respective nominal values of each parameter, or thresholds, representative of a condition of no calcification, or of a condition of a beverage preparation machine in perfect state of cleaning and / or ideal operating condition as designed.

[0066] According to an implementation option, the aforesaid nominal values of the parameters are specific to a specific type of beverage preparation machine (a coffee preparation machine, for example) to which the method applies, and are determined in an initial self-calibration or auto-identification process.

[0067] According to another embodiment of the method, the aforesaid step of determining the calcification condition comprises determining a calcification state, in which the step of processing, by the trained algorithm or model, comprises classifying the calcification condition as “calcification presence state” or “calcification absence state” based on predefined criteria taken into consideration in the step of training the algorithm or model.

[0068] According to an implementation option of the aforesaid embodiment, the aforesaid classification step comprises:

[0069] - partitioning, by the trained algorithm or model, based on the training, a multidimensional vector space having the aforesaid parameters as the coordinates and the number of such parameters as the size, into at least two regions: a first region associated with the presence of the calcification state and a second region associated with the absence of the calcification state;

[0070] - determining the presence or absence of a calcification state based on the fact that an input vector, comprising the input data representative of the detected or estimated values of said parameters, in a specific condition to be assessed, is in the aforesaid first region or in the aforesaid second region, respectively.

[0071] Further details concerning the trained algorithms that can be used in the present method and the function thereof are described in a following part of this description.

[0072] According to an embodiment of the method, the determination of the calcification condition occurs during the operation of the beverage preparation machine, for example when dispensing a beverage, by detecting the aforesaid parameters during the normal operation of the machine.

[0073] According to another embodiment of the method, the determination of the calcification condition occurs during one or more test cycles specifically performed for determining the calcification condition.

[0074] According to possible implementation options, such test cycles are performed while the beverage preparation machine is not under normal operating condition, or while the beverage preparation machine is under normal operating conditions, but in an automated manner not perceivable by the user.

[0075] According to an implementation option, each of the one or more test cycles comprises providing electric power test signals to a heater included in the thermo- hydraulic system, and detecting the aforesaid parameters for each of the test signals.

[0076] According to an implementation option, the aforesaid test signals comprise signals having constant shape and value, for example, preferably, step or square wave signals.

[0077] According to another implementation option, the aforesaid test signals comprise signals characterized by varying values and durations corresponding to pseudo-random numbers.

[0078] According to an embodiment of the method, the aforesaid output information, representative of the calcification condition, comprise a calcification probability.

[0079] According to an implementation option, input data corresponding to detected or estimated values of the aforesaid parameters, corresponding to a plurality of test cycles in which each test cycle is associated with a different and increasing amount of liquid passing through the hydraulic circuit, are provided as input to the trained algorithm or model.

[0080] In this case, the method includes the further step of determining, by the trained algorithm or model, the probability of the presence of a calcification state in an iterative manner based on the determination of the presence or absence of calcification in each test cycle, starting from a state of no calcification.

[0081] Some further details and illustrations of the method according to the present invention are provided below by way of non-limiting example.

[0082] The principle underlying the present invention is the fact that the progressive deposit of limescale in the hydraulic circuit (in the heating tube, for example) of the thermo- hydraulic system (and also, not to be neglected, at the temperature sensor associated therewith, for example, downstream thereof) occurs as a deterioration in the heating performance.

[0083] It has been noted that a phenomenon particularly indicative of the deterioration in heating performance is not so much of a change in the static thermal response (therefore, static variations in the reached temperature with respect to a nominal value) as instead a variation in the dynamic response of the thermo-hydraulic system, for example of the heating tube: the expectation is, and it was demonstrated to be this way, a decrease in the static gain and an increase in the response time given that the limescale layer acts as thermal insulation.

[0084] The variations in the dynamic thermal response have proven to be an effective indicator of the calcification state.

[0085] Additionally, and in combination with the above phenomenon, a further phenomenon can be usefully detected to further improve determining calcification, i.e., the increase in the hydraulic resistance of the thermo-hydraulic system, since the deposit of limescale results in a section reduction of the hydraulic circuit, for example, tube (hereinafter also referred to as an “heater”).

[0086] By analyzing first the dynamic response of the thermo-hydraulic system, it can be noted that the thermo-hydraulic system, from a thermal viewpoint, is well approximated by a first order model with delay time (FOPDT), shown in simplified form in the diagrams reported in Figure 2.

[0087] Such a model can be summarized using the following formula: in which the aforesaid three parameters static gain Kp, delay time 0 and rise time T of the temperature response appear, parameters which, not by chance, are detected in the method according to the invention since they are necessary and sufficient to characterize the dynamic thermal response in a sufficiently accurate manner.

[0088] In an initial process of identification of the beverage preparation machine, the aforesaid generic model (function of Kp, T, 9) is optimized to minimize the mean square error on a data set specifically obtained (heater command as input, output temperature as output).

[0089] According to an implementation option, a series of constant magnitude steps on the heater command are used as input sequence for the optimization. In an application example referring to a beverage preparation machine, the step of generation of data for the identification is introduced at the end of the cleaning step, during the filter support rinsing.

[0090] According to an implementation option, the Nelder-Mead optimization algorithm (which was tested by the Applicant and provided good results) is used to find the optimal set of parameters Kp, T, 9.

[0091] Figure 3 shows an exemplary optimization result, showing that the time response function of the model, which emerged from the identification based on the aforesaid formula with optimized parameters, overlaps the experimental data very well.

[0092] Again with reference to the thermal model used, note that when a step electrical power is applied, the heater of the thermo-hydraulic system has a conventional water temperature dynamic response of the type shown in Figure 2. The fundamental properties of such a response are (as observed above) the static gain Kp, the time delay 9 and the rise time T.

[0093] The features of the response are not constants, rather they depend in a quite pronounced manner on the instantaneous water flow rate, which largely determines the convective heat exchange coefficients. The particular value taken by the response parameters for each flow rate depends on various machine features, including shape and volume of the heater, shape, position and features of the temperature sensor and the housing thereof, thermal insulation.

[0094] For water in the liquid state under stationary conditions, i.e., constant power and at constant input and output temperature, the following energy balance relation is valid with good approximation: where Re is the electric power in Watts, i the yield, m the flow rate in kg / s, cpthe specific heat [J / (kg°C)], Toutand Tinthe input and output water temperature, respectively, into / from the heater, in °C.

[0095] Within the context of the control system, the controlled temperature is preferably measured by means of a sensor placed immediately downstream of the heater.

[0096] The yield in the above relation, less than the unit, establishes how much electric power is actually transferred to water. The remaining part is a loss of energy exchanged with the rest of the machine and / or towards the environment. Moreover, there is a mathematical association between such a yield and the dynamic response features.

[0097] As observed above, the progressive calcification inside the hydraulic tube or conduit of the thermo-hydraulic system is a thermally insulating layer which results in a reduction in the amount of energy that the heater tube can give to water, and therefore in a lowering of the yield value. Under such conditions, the mean temperature of the tube rises and the heat losses to the outside increase.

[0098] It has also been empirically determined that the deposit of limescale tends to accumulate at the output of the heater, just at / on the terminal of temperature sensor. This insulating layer especially affects the dynamic response of the system, which is slowed down. As shown in Figure 2, the expectation from a quality viewpoint is a progressive increase in the delay and rise time accompanied by a decrease in the gain.

[0099] In order to validate the efficacy of using the aforesaid parameters related to the dynamic thermal response, the Applicant has performed different tests, having as object a beverage preparation machine, in which the tests were articulated in a plurality of “tests” for validating the algorithm, starting from the descaled state and observing the parameters described and the output of the model. Such “tests” are also referred to as “runs” hereinafter and in the figures. The diagrams in Figure 4 show, by way of example, the results of four “runs”, i.e., the value of each of the three thermal parameters considered (static gain Kp, delay time 0 and rise time T of the temperature response) as a function of the volume in liters of processed hot water (i.e., water passing through the heater).

[0100] The diagrams in Figure 4 show that the three parameters of the thermal system are a good indication of the calcification state: in all the “runs”, there is observed a tendential increase of T, 9 and a tendential decrease of Kp as the calcification state increases; it emerges that - a fortiori - considering the combination of the three aforesaid parameters provides a reliable and accurate indication concerning the calcification condition.

[0101] Now it is worth noting the further parameter referring to the hydraulic resistance, that is used in combination with the three parameters of the thermal dynamic response, in an embodiment of the method that allows to further improve the results of the determination of the calcification state.

[0102] As the limescale accumulates on the heater, the loss of pressure generated thereby increases. Therefore, the required pump command will be higher, required flow rate and pressure being equal.

[0103] For this reason, in the implementation option herein considered in which it is assumed that the flow rate and pressure of the liquid flow must remain the ones required by the machine, the pump command signal can be considered as an operating parameter that is easily detectable and fairly representative of the hydraulic resistance of the thermo- hydraulic system, i.e., of the hydraulic circuit to which the heating tube and / heater belong.

[0104] Figure 5 shows the increase in the pump command as a function of the liters of processed hot water, according to different “runs” performed.

[0105] As the number of liters of processed water increases (and therefore, as the calcification increases), the pump command, representative of the hydraulic power, takes a clearly increasing tendential trend, as is expected.

[0106] Note that besides being useful for validating the method, the data shown in Figures 4 and 5 can be used, in an embodiment of the method, for the training (included in the method) of the artificial intelligence algorithm (and / or “machine learning-”based algorithm).

[0107] Some further information is provided hereinafter, by way of example, concerning the trained model (i.e., algorithm) that can be used for implementing the invention.

[0108] According to an embodiment, the model employed is a logistic regression, which is a classification algorithm. The training of such a model, according to an implementation option, was performed on data collected as indicated above.

[0109] According to possible implementation options, some techniques, in themselves known in the “data science” field (e.g., data normalization, class balancing, etc.), were also used to improve the results.

[0110] According to other possible embodiments, there are used, as classification algorithms, one or more of the following algorithms, in themselves known:

[0111] - decision tree;

[0112] - Random Forest;

[0113] - Support Vector Machine;

[0114] - neural network.

[0115] As observed above, in an embodiment of the method, there are detected and provided, to a trained algorithm, four input parameters: the aforesaid three identified parameters concerning the thermal response of the thermo-hydraulic system (Kp, 0, T) and the mean pump command during the test routine, representative of the hydraulic resistance.

[0116] Based on these, the trained algorithm estimates the probability that the system is calcified.

[0117] In an implementation option, already mentioned above, of the aforesaid embodiment of the method, the trained algorithm, on the basis of the training received, performs a partition of a four-dimensional vector space having as coordinates the aforesaid four parameters, into at least two regions, a first region associated with the presence of the calcification state and a second region associated with the absence of the calcification state. The presence or absence of a calcification state is determined based on the fact that an input vector, comprising the input data representative of the detected or estimated values of the four parameters, under a specific condition to be assessed, is in the first or second region, respectively.

[0118] In particular, analyzing the four parameters mentioned above through the use of “cluster analysis” techniques, two “clusters” corresponding to the non-calcified and calcified states are identified.

[0119] Figure 6 shows, by way of example, two charts illustrating a partition into two clusters, associated with the presence and absence of the calcification state.

[0120] In this example, the two clusters were identified by applying the “Gaussian Mixture” algorithm.

[0121] In the graphical depiction in Figure 6, the two clusters are identified by different punctiform symbols, as indicated in the legend. Also note that, for graphical reasons and for easier illustration, the chart was reduced from four to two dimensions through the known PCA technique in which the abscissa and ordinate values represent the coordinates of the system with a dimension reduction.

[0122] Based on the above, the trained algorithm is capable of also predicting the probability that the heating tube is under a calcification condition, which depends inter alia on the number of liters of water processed. The results of three examples (“runs”) of such an assessment are shown in Figure 7.

[0123] According to an embodiment, the model or algorithm trained by means of the aforementioned “runs” is used to predict the probability that the system is calcified also in the context of the test cycles (already illustrated above).

[0124] The diagram illustrating this information is shown in Figure 8. In such an example, the trained model / algorithm begins indicating that the system is calcified at about 50 liters: consistently with this, a sharp increase in the aforesaid thermal response parameters occurs at this volume.

[0125] The following are some exemplary details related to the aforementioned embodiment which includes performing the test cycles.

[0126] Concerning the generation of the dedicated test signals, two implementation variants are considered herein: (i) signals having constant shape and value; (ii) signals characterized by random values.

[0127] Figure 9 exemplifies the two approaches.

[0128] In the first case (Fig. 9A), a step power signal, or square wave, is generated in the product development step and coded in the firmware. Such a signal remains unvaried for all the test cycles.

[0129] In the second case (Fig. 9B), a pseudo-random number generator is used to obtain numbers / values defining the duration and magnitude of the steps. The random numbers are generated according to accurate probability distributions to explore the operating field, from a statistical viewpoint, in the best possible. Thus, a different signal capable of increasing the informative level of the data generated is obtained each time.

[0130] The advantage of the dedicated test cycles is mainly the increased accuracy in calculating the parameters and in detecting the limescale.

[0131] On the other hand, the estimate during the dispensing of the beverages, in comparison, is probably less accurate but, given that it is continuously performed, can be highly effective in identifying the progressive calcification state at an early stage.

[0132] With reference to the detection of the temperature and of the temperature trend over time (dynamic response), note that, according to a preferred embodiment (mentioned above), such a detection is performed by means of a single temperature sensor placed preferably at the output or just downstream of the heating tube of the hydraulic circuit of the thermo-hydraulic system.

[0133] Figure 10 illustrates different possible implementation options concerning the positioning of the temperature sensor 3 with respect to a cross section (examples of cross sections indicated in circular shape in Figure 10) and with respect to a longitudinal section of the heating tube 4 and / or junction with the remaining hydraulic circuit (respective examples of longitudinal sections indicated in rectangular shape in Figure 10).

[0134] Note that, in other possible implementation options, the method is applicable to tubes / conduits having the most diverse geometrical junction shapes, narrowings, tapers, and the most diverse section shapes, and accordingly, the temperature sensor can be placed in different positions with respect to any type of tube / conduit considered.

[0135] According to different implementation options, the temperature sensor can be arranged just brushing the tube / hydraulic circuit or arranged so as to touch or penetrate the opposite wall of the tube.

[0136] According to an embodiment, the method is applicable to a hydraulic system or circuit such as the one shown by way of explanation in Figure 1 1 .

[0137] In particular, Figure 1 1 shows a simplified diagram of a beverage preparation machine 100 comprising a thermo-hydraulic system 1 which, in turn, comprises a hydraulic circuit 2.

[0138] Figure 1 1 further shows a temperature sensor T (also indicated by the reference numeral 3), a conduit or tube 4, a heater 5, a pressure sensor P, a pump M, all elements associated with the hydraulic circuit 2.

[0139] There are also depicted a liquid tank 6, a flow meter 7, discharge conduits, or drains, 8, a one-way valve V1 , two further valves V2 and V3.

[0140] The beverage preparation machine 100 further comprises an extraction chamber 9, a pressure control valve V4 and an output U for the beverage.

[0141] As can be noted, the object of the present invention is fully achieved by the method disclosed above by virtue of the respective functional and structural features.

[0142] Indeed, the aforementioned method provides an estimate of state / probability of calcification which, although it is based on a simple temperature detection, is accurate, reliable and quick.

[0143] In particular, the use of parameters related to the temperature dynamic response allows assessing not only the calcification of the tube, but also considers any deposits / insulation of the sensor, differently than an assessment based on a simple static comparison of the temperature with one or more preset thresholds.

[0144] Therefore, due to the aforesaid features, the method according to the invention provides an improved determination of calcification with respect to the prior art, which is capable of considering, and discriminating, the effect of the calcification on the temperature sensor.

[0145] Moreover, the dynamic response does not depend on room temperature, while the static temperature assessment (prior art) would depend on room temperature and would provide results depending thereon (therefore, not absolute and less accurate results).

[0146] In addition, the embodiment of the method comprising the use of a fourth parameter, representative of the hydraulic resistance, provides further advantages in terms of increased flexibility and speed, in light of the fact that the determination depends, case by case, on the quickest among the dynamic phenomena of temperature response and variation in the hydraulic resistance (which aspect can depend, in turn, on the water hardness, for example).

[0147] A further advantage of this embodiment is the increased reliability, by eliminating phenomena of wear and false positives, due to a kind of cross-check between the results suggested by the two groups of variables (thermal variables and variable correlated with hydraulic resistance).

[0148] In order to meet contingent needs, those skilled in the art may make changes and adaptations to the embodiments of the method described above or may replace elements with others which are functionally equivalent, without departing from the scope of the following claims. Each of the features described as belonging to a possible embodiment can be achieved irrespective of the other embodiments described.

Claims

CLAIMS1. A method for determining a condition of calcification and / or accumulation of limescale or other scale in a thermo-hydraulic system adapted to convey and heat liquid in a beverage preparation machine for professional or household use, wherein the method comprises:- detecting at least one value of each of a plurality of parameters relating to a thermal response of the thermo-hydraulic system, said parameters comprising a static gain (Kp), a delay time (0) between electric heating power supplying and temperature response of the thermo-hydraulic system, and a rise time (T) of said temperature response;- providing the detected values of each of said plurality of parameters as input data to a trained and / or machine learning-type algorithm or artificial intelligence model, wherein said algorithm or model was trained by means of a plurality of input training data, comprising data representative of values of said parameters in each of a plurality of known calcification situations and data representative of the respective known calcification conditions;- processing said input data, by said trained algorithm or model, to obtain output information, representative of said calcification condition;- determining said condition of calcification and / or accumulation of limescale or other scale based on the output information provided by said trained algorithm or model.

2. A method according to claim 1 , comprising the further step of:- detecting at least one value of an additional parameter relating to a hydraulic resistance of said thermo-hydraulic system;- providing at least one detected value of the additional parameter, in addition to the detected values of the plurality of parameters relating to the thermal response, as input to said trained algorithm or model;- processing said input data, by said trained algorithm or model, to obtain output information representative of the calcification condition, based on both the values of the parameters relating to the thermal response and the detected value of the additional parameter relating to hydraulic resistance and / or liquid flow;- determining the condition of calcification and / or accumulation of limescale or other scale based on the output information from the trained algorithm or model.

3. A method according to any one of the preceding claims, wherein said thermo-hydraulic system comprises a hydraulic circuit comprising a heating tube or duct.

4. A method according to any one of claims 2 or 3, wherein said additional parameter is a pump command and / or control signal of the thermo-hydraulic system or hydraulic circuit pump, wherein said pump command and / or control signal is representative of said hydraulic resistance of the thermo-hydraulic system and / or of a liquid flow in the hydraulic circuit.

5. A method according to any one of the preceding claims, wherein said parameters relating to a thermal response of the thermo-hydraulic system are detected by a single temperature sensor operatively connected to the thermo-hydraulic system and / or to a hydraulic circuit or hydraulic channel or heating tube of the aforesaid thermo-hydraulic system.

6. A method according to any one of the preceding claims, wherein said beverage preparation machine is a machine for preparing coffee, and said thermo-hydraulic system is a thermo-hydraulic system of the machine for preparing coffee.

7. A method according to any one of the preceding claims, wherein said liquid is water, or wherein said liquid is an infusion or extract, such as tea or coffee or herb tea or milk.

8. A method according to any one of the preceding claims, wherein said thermo- hydraulic system comprises a heating tube, and wherein said dynamic temperature response parameters characterize the dynamic temperature response of the heating tube.

9. A method according to claim 8, wherein said heating tube acts as both a heater and a conductor of the liquid, or wherein said heating tube acts as a conductor of the liquid and is operatively connected to a heater device adapted to heat it.

10. A method according to any one of the preceding claims, wherein said detecting step comprises taking a plurality of measurements of each of said parameters, within a predefined time interval, and wherein the method comprises the further steps of:- estimating a statistical value of each of said parameters based on said plurality of measurements;- providing, as input data, said estimated statistical values for each of said parameters to the trained algorithm.

11. A method according to any one of the preceding claims, wherein said output information, representative of the calcification condition, comprises a level of calcification or accumulation of limescale or other scale and / or a probability or a state of calcification or accumulation of limescale or other scale.

12. A method according to any one of the preceding claims, wherein said step of determining the calcification condition comprises determining a calcification state, and wherein the processing step comprises:- identifying the presence of a calcification state based on a comparison of the detected or estimated values of said parameters with respective nominal values of each parameter or thresholds, representative of a condition of no calcification, or of a condition of beverage preparation machine in perfect state of cleaning and / or ideal operating condition as designed.

13. A method according to claim 12, wherein said nominal values of the parameters are specific to a specific type of beverage preparation machine to which the method applies, and are determined in an initial self-calibration or auto-identification process.

14. A method according to any one of the preceding claims, wherein said step of determining the calcification condition comprises determining a calcification state, and wherein the step of processing, by the trained algorithm or model, comprises:- classifying the calcification condition as a “calcification presence state” or a “calcification absence state” based on the predefined criteria taken into account in the step of training the algorithm or model.

15. A method according to claim 14, wherein said classifying step comprises:- partitioning, by the trained algorithm or model, based on the training, a multidimensional vector space having said parameters as the coordinates and the number of said parameters as the size, into at least two regions, a first region associated with the presence of the calcification state and a second region associated with the absence of the calcification state;- determining the presence or absence of a calcification state based on the factthat an input vector, comprising the input data representative of the detected or estimated values of said parameters, in a specific condition to be assessed, is in said first region or second region, respectively.

16. A method according to any one of claims 1 -15, wherein the determination of the calcification condition occurs during the operation of the beverage preparation machine, for example when dispensing a beverage, by detecting said parameters during the operation of the machine.

17. A method according to any one of claims 1 -15, wherein the determination of the calcification condition occurs during one or more test cycles specifically performed for determining the calcification condition.

18. A method according to claim 17, wherein each of said one or more test cycles comprises providing electric power test signals to a heater included in the thermo- hydraulic system, and detecting said parameters for each of said test signals.

19. A method according to claim 18, wherein said test signals comprise signals having constant shape and value, preferably step or square wave signals.

20. A method according to claim 18, wherein said test signals comprise signals characterized by varying values and durations corresponding to pseudo-random numbers.

21. A method according to claim 1 1 and claim 17, wherein said output information, representative of the calcification condition, comprises a calcification probability.

22. A method according to claim 21 , wherein input data corresponding to detected or estimated values of said parameters, corresponding to a plurality of test cycles, each test cycle being associated with a different and increasing amount of liquid passing through the tube, are provided as input to the trained algorithm or model, and wherein the method further comprises determining, by the trained algorithm or model, the probability of presence of a calcification state in an iterative manner based on the determination of presence or absence of calcification in each test cycle, starting from a state of no calcification.

Images