METHOD FOR OPERATING A CRYOGENIC TUNNEL

Description

[0001] The present invention relates to a method and a device for operating a cryogenic tunnel, a tunnel of the type in which products to be cooled or frozen circulate, equipped with means for injecting a cryogenic fluid as well as means for extracting cold gases resulting from the vaporization of the fluid in the tunnel.

[0002] Document US-5,606,861 illustrates the state of this technical field in the case of so-called “IQF” products (rapid freezing of individual products such as fruits, vegetables, balls, etc.).

[0003] In this industry, users adjust key tunnel operating parameters to maintain optimized production.

[0004] These parameters include the temperature setpoint inside the tunnel, the conveyor speed, and the fan speed.

[0005] These parameters, as we know, strongly influence the quality of the products obtained and the cost of the process through the consumption of cryogen, hence the importance of being able to optimize these parameters.

[0006] However, we know that the operating conditions of such tunnels are not stable, they vary depending on the batches of products processed: in some cases the temperature of the incoming products varies, in other cases the thickness of the products varies from one batch to another, while in other cases the flow rate of products to be processed or the composition of the products changes during the day.

[0007] In this context, optimizing production parameters is a very complex objective, and in practice, users of such tunnels are unable to adjust their tunnel parameters continuously over the course of productions; they actually choose to adopt an "average" setting which is supposed to cover most of their productions more or less correctly.

[0008] For example, we observe that during a given production the cooling power is too high, the temperature of the products leaving the tunnel is then too low, which does not necessarily cause a quality defect depending on the products considered.

[0009] In other cases, the cooling power is too low, the temperature of the products leaving the tunnel is then too high, which this time causes a serious problem of quality and acceptability of the products obtained, and which can lead to the elimination of the products in question, and the need to restart production. This is one of the reasons why producers prefer to adjust their set temperature slightly too low for safety, which of course represents a significant cost.

[0010] One solution to this problem would be to continuously adjust the tunnel operating conditions so that the product is cooled or frozen to just the desired level, for example based on a temperature measurement taken on the products at the tunnel outlet and feedback on the cryogen supply conditions: if the outlet product temperature is lower than the target, the tunnel temperature will be raised, while if the outlet product temperature is too high the tunnel temperature will be rapidly lowered.

[0011] Another solution could be based on the temperature of the products entering the tunnel: when the products arrive at a higher temperature than expected, the tunnel switches to a lower setpoint mode, while when the products arrive at a lower temperature than expected, the tunnel switches to a higher temperature setpoint mode, favoring reduced cryogen consumption.

[0012] The literature in this field mentions in particular the following technical solutions: The CALLIFREEZE ® system from GEA, in the field of mechanical (non-cryogenic) refrigeration, where a controller performs this continuous adjustment of the tunnel parameters by continuously monitoring the crystallized water content in the products, and by adjusting accordingly the residence time of the products in the tunnel, the temperature of the air inside the tunnel, and the speed of the fans, to achieve the desired freezing level, while minimizing energy consumption and maintaining optimum product quality. To do this, a probe is positioned at the tunnel outlet, it measures the "freezing rate" of the products, in other words, when x% of the water in the product is frozen then it is considered that x% of the product is frozen. This rate is calculated from microwave measurements and using comparison tables.When this freezing rate is too low, the controller feeds back on the fan speed to increase them, when this freezing rate is too high then we understand that the tunnel brings too many frigories to the products and consumes too much energy, the controller then feeds back on the fan speed to decrease them. The system described in document WO 2011 / 136900 A1: this system is based on a measurement of the quantity of products entering the tunnel, and on the use of an infrared sensor measuring the temperature of an incoming product, or even of the outgoing products, the controller then recovers all of this data to calculate the quantity of cryogen that should be injected into the tunnel to reduce the temperature of the outgoing products to a desired level.The system described in document EP-3 170 404B1: this system is based on the implementation of a scanning laser, at the entrance to the tunnel, to acquire a sectional image of an incoming product, while an infrared temperature probe is positioned downstream of the laser at the entrance, and a second infrared probe is positioned at the exit of the tunnel. A controller manages all of these sensors and is able to automatically adjust the heat transfer to the products by acting on the arrival of cryogen.

[0013] If the solutions listed above undoubtedly bring an improvement to the process and in particular to the quality of the products obtained, the Applicant considers for its part that improvements are still necessary, and that these can be obtained according to the invention by the following approach: By measuring several key process parameters. By dividing these parameters into two different groups of parameters, characterizing the tunnel: a first group for parameters that can and will be used to anticipate the future freezing power required by the tunnel. This first group includes, for example, one or more of the following parameters: o The temperature of the products entering the tunnel o The volume flow rate of the products entering the tunnel o The mass flow rate of the products entering the tunnel o The color of the products entering the tunnel o The coverage rate of the conveyor belt feeding the tunnel, or o Parameters characterizing the atmosphere surrounding the tunnel in the room: ambient temperature, ambient humidity and atmospheric pressure. We understand, for example, that if the temperature of the products arriving at the entrance of the tunnel increases,then the tunnel will need to provide more frigories. We will then seek to compensate as exactly as possible for the difference in energy of the products. For example, if the products normally have an energy level of 100kcal / kg and they suddenly arrive warmer with an energy of 122kcal / kg, we will then seek to compensate for the 22 kcal / kg more by providing 22 k frigories / kg more inside the tunnel. A second group of parameters that can and will be used to evaluate the final result of the products leaving (including the temperature of the products leaving the tunnel), parameters that will in particular indicate whether the product is correctly frozen. This second group includes for example one or more of the following parameters: o The temperature of the products leaving the tunnel (frozen) o The flow rate of cryogen admitted into the tunnel ∘ The temperature of the gases extracted from the tunnel,typically at the level of two extraction hoods located at the entrance and exit of the tunnel ∘ The temperature prevailing in the room in the vicinity of these two extraction hoods ∘ The hardness of the products leaving the tunnel ∘ The color of the products leaving the tunnel ∘ The percentage of frozen products which are IQF type (i.e. individual, such as fruits, vegetables, balls, etc.). For example, we can count the number of product particles stuck together and make the following calculation: percentage of IQF product = 100 x number of unstuck particles / total number of particles. By particle, we mean here for example a pea, a bean, a raspberry, a chicken nugget, a slice of bacon, a piece of fish, etc. By implementing one of the two or both of the following types of actions based on the values ​​obtained for these key parameters: Anticipation actions,calculated from the values ​​obtained by the parameter(s) of the 1st group (to act upstream on an anticipated / expected deviation in the freezing quality, for example because the temperature of the incoming products is too high) anticipated actions, i.e. actions on the tunnel even before the product comes out too hot or too cold, in other words, there is no point, for example, in waiting to measure temperatures of outgoing frozen products that are too high at the tunnel exit, it is possible to anticipate an action to modify the tunnel parameters; and Feedback (countermeasures) calculated from the values ​​obtained by the parameter(s) of the second group, to rebalance an effective, measured drift in the quality of the outgoing products,for example, because the temperature of the frozen products coming out of the tunnel is too hot. So for these products (too hot) it is too late to correct the process and obtain good freezing, but it is possible to adjust a colder setting of the tunnel so that the following products come out at a satisfactory temperature.

[0014] As will be seen in more detail below, the present invention makes the equipment easier to use: the user sets the temperature of the frozen products, -20°C for example, and the method according to the present invention manages everything else, everything that needs to be managed.

[0015] The user no longer has to ask himself: if I have a fairly hot product entering the tunnel, the flow rate is quite high so I should put the tunnel at -110°C rather than -100°C?, the fans at 90% rather than 50%?, the extraction at X or Y%?.

[0016] And when production conditions change, the user does not have to worry about the tunnel, it will adapt itself to maintain a temperature of frozen products at -20°C.

[0017] It is known that a cryogenic tunnel generally includes the following elements: A conveyor for the products inside the tunnel: changing the speed of the conveyor modifies the residence time of the products and their freezing time. A system for injecting the cryogenic fluid into the space inside the tunnel, a system consisting, for example, of a valve for adjusting the arrival of cryogen in the tunnel, pipes and several nozzles for spraying the cryogen, this being supplemented by a controller capable of modifying the opening rate of the valve to restore the internal temperature in the tunnel to a set point. A ventilation system organizing the transfer of frigories to the products. Means for extracting at variable flow rates all or part of the cold gases resulting from the vaporization of the fluid in the tunnel, traditionally most often with extraction at the entrance and extraction at the exit of the tunnel.A data acquisition and processing unit, capable of receiving data from all of these devices (conveyor speed, fan speed, extractor speed, temperature probes, etc.) and acting on all or part of these devices.

[0018] In addition, one or more of the following devices and the following data may be available: of a given and chosen number of distance sensors, whose role is to measure the thickness of products at different points of the conveyor. When the conveyor is empty of products, the measured distance is zero, while when a product arrives opposite a sensor, this sensor, above the product, measures a distance which is equal to the thickness of this product (by difference). a controller receives all of these measurements and calculates an average thickness of the products present on the conveyor. we therefore do not preferentially use a scanning laser but a group of several distance sensors allowing to adequately cover the dimensions of the tunnel considered, a much more robust solution (fixed sensors having no moving parts). of a measurement of the speed of the conveyor. the controller is then able to carry out the following evaluation: the temperature of the products entering, by a device available on the market for this.the temperature of the products leaving, again by a device available on the market for this purpose. the flow of cryogen injected into the tunnel. the temperature of the gases extracted from the tunnel, typically at the tunnel entrance and exit. the ambient temperature in the room housing the tunnel (and in particular, for example, the temperature prevailing in the vicinity of two extraction hoods at the tunnel entrance and exit), as well as the ambient humidity and atmospheric pressure in this room.as mentioned above, we advantageously have a data acquisition and processing unit capable of receiving all the measured data and of providing feedback for all the necessary actions, in particular for re-establishing the instructions, this unit (computer, automaton or other) can be located in the vicinity of the tunnel in the room, or we can adopt a configuration using such a local unit to send all the data to a remote computer (network), a computer which carries out all the calculations and evaluations before sending the results back to the local unit, a local unit which then carries out the necessary actions to modify the parameters of the tunnel according to the invention.

[0019] Equipped with these two groups of very distinct parameters, we construct a matrix for each group, such as those that we will exemplify below in order to better understand the invention: The first group of measured parameters forms an input of the first matrix, the first group used to calculate and trigger feedforward actions; The second group of measured parameters forms an input of the second matrix, the second group used to calculate and trigger feedback actions;

[0020] As detailed below, the measurements obtained for one or some parameters of these two groups of parameters allow us to calculate the adjustments to be made to a group of action parameters (by anticipation and feedback), a group of actions made up of: ∘ The speed of the conveyor ∘ The speed of the fans inside the tunnel ∘ The temperature of the gases coming from the inlet hood ∘ The temperature of the gases coming from the outlet hood ∘ The temperature setpoint inside the tunnel.

[0021] And then it should be noted that the second entry of the two matrices is made up of the same group of parameters which is the group of action parameters mentioned above.

[0022] As will be demonstrated below, the invention, unlike the prior art, acts not on a single parameter (for example only on the fan speed or only on the heat transfer) but on several parameters governing the operation of the tunnel.

[0023] To better understand the invention, below is an example of a first matrix consisting of the 1st group of parameters and the anticipatory actions exercised, each box of this matrix consisting of a factor establishing a link between a given parameter and a given anticipatory action (anticipatory action located in said group of actions listed above).

[0024] And below we also find an example of a second matrix made up of the 2nd group of parameters and the "feedback" actions carried out, each box of this matrix being made up of a factor establishing a link between a given parameter and a given feedback (feedback located in the said group of actions listed above). Matrix of the 1st group of parameters (so-called “anticipation” parameters)

[0025] Conveyor speed Fan speed Gas hood inlet temperature Gas hood outlet temperature Temperature setting inside the tunnel Incoming product temperature x11 x21 x31 x41 x51 Incoming product flow rate x12 x22 x32 x42 x52 Room temperature x13 x23 x33 x43 x53 Ambient humidity x14 x24 x34 x44 x54 Atmospheric pressure x15 x25 x35 x45 x55 Matrix of the 2nd group of parameters (so-called “feedback” parameters)

[0026] Conveyor speed Fan speed Gas hood inlet temperature Gas hood outlet temperature Temperature setting inside the tunnel Outgoing product temperature y11 y21 y31 y41 y51 Injected cryogenic flow y12 y22 y32 y42 y52 Gas hood inlet temperature y13 y23 y33 y43 y53 Gas hood outlet temperature y14 y24 y34 y44 y54 Temperature near the hood inlet y15 y25 y35 y45 y55 Temperature near the hood outlet y16 y26 y36 y46 y56

[0027] And we will note that if the content of certain boxes is set to zero, in their case we must understand that there is no action or feedback to be taken.

[0028] Let us exemplify in the following the actions that can be taken: o If the box called x11 is at -3, then when the temperature of the incoming product is equal to 5°C, higher than the setpoint, the system will automatically adapt the conveyor speed by 5 x (-3) = -15 units (for example -15%). Consequently, for an incoming product that is warmer than expected, the tunnel will, in anticipation, initiate a reduction in the conveyor speed and a longer freezing time. In other words, the higher temperature of an incoming product will be automatically anticipated and counterbalanced by the system.

[0029] And the same type of action will be defined in the matrix for each pair of parameters.

[0030] And if we decide, in a given pair of parameters (a given box), that we do not want any action, then the link parameter will be set to zero (for example we decide to set X12 to zero and thus variations in product flow will not cause any action on the conveyor speed). This can for example be the case when we want the conveyor to operate at a constant speed. In this case, all the parameters in the matrix having an impact on the conveyor speed will be set to zero. In our previous example, x11 will be adjusted to zero.

[0031] For the second group of parameters (second matrix), the operation will be exemplified below.

[0032] So for example if Y51 = -0.5, the system will then proceed with the following action: if a product leaving the tunnel has a temperature of 6°C, higher than the setpoint, the tunnel will then automatically adjust the temperature inside the tunnel by 6 x (-0.5) = 3°C i.e. by decreasing the setpoint by 3°C per minute, step by step. The temperature of the outgoing products will thus decrease step by step, until it reaches the required setpoint at the outlet. When the required setpoint is obtained, the system stabilizes and the temperature setpoint in the tunnel will also be stabilized at the value that allows optimal cooling of the product.

[0033] Again, if we do not want to take any action on a given pair of parameters, we set the corresponding box in the matrix to zero, for example if box Y22 is set to zero, a variation in the injected cryogen flow rate will not cause any action on the fan speed.

[0034] We can therefore see that in the examples given here, the actions are calculated differently but they act on the same group of action parameters, whether for anticipatory actions or feedback actions.

[0035] Whether we detect that the products arrive too hot (we can then consider reducing the tunnel temperature in anticipation) or that the products come out too cold (we can then consider reducing the tunnel temperature by feedback so that the products that follow are at the right temperature), the action is the same.

[0036] In other words, there can be two different causes / origins for changing a single tunnel parameter.

[0037] Conversely, we can also observe a single parameter which causes the modification of several tunnel parameters.

[0038] As will be clear to those skilled in the art, the matrices presented here can be established for a smaller number of parameters, or for a larger number of parameters, by adding parameters not listed here, for example to process tunnels equipped with more than a single conveyor, or with several temperature zones and therefore several temperature setpoints.

[0039] The following describes an example of experimental determination of the values ​​constituting the boxes of the two matrices described above.

[0040] To do this, to fill a box, that is to say to determine the factor occupying the box in question, a step-by-step process is carried out.

[0041] Let us consider the example of the factor y51 which establishes a relationship between the set temperature of the freezer and the temperature of the products leaving the freezer.

[0042] Step 1: The user starts the device with products. For this adjustment procedure to be successful, it is preferable, even essential, that the process is as stable as possible, that the flow rate and temperature of the products to be frozen are stable, and that the freezer parameters (fan speed, belt speed, extraction speed) are also stable.

[0043] Step 2: When the freezing process is operating stably, the control system of the present invention is put into operation and a target temperature for the frozen products leaving the freezer is set according to the customer's needs. For frozen products, a target temperature of -20°C will often be taken.

[0044] The temperature of the frozen product is then monitored by the user. Preferably, this temperature is recorded and the curve is built live.

[0045] Step 3: Parameter y51 is set to an arbitrarily chosen value, for example 1. All other parameters are set to zero to avoid interference between the control loops.

[0046] Step 4: the user then looks at the behavior of the temperature curve of the frozen product for at least a time equivalent to 4 times the time the product spent in the freezer: o Case A: The temperature varies above and below the setpoint. The amplitude of the variation decreases rapidly over the passages and the temperature ends up stabilizing very close to the setpoint. In this case, the user goes to step 5. o Case B: The temperature curve remains below the setpoint. The user doubles the value of y51. If y51 was equal to 1, it goes to 2. Then the user returns to the beginning of step 4. o Case C: The temperature curve remains above the setpoint. The user doubles the value of y51. If y51 was equal to 1, it goes to 2. Then the user returns to the beginning of step 4...... o Case D: The temperature curve is not stable, it goes above and below the setpoint with a stable or increasing amplitude. The user divides y51 by two. If y51 was equal to 1, it goes to 0.5. Then the user returns to the beginning of step 4.o case E: the temperature curve is not stable, it goes above and below the setpoint with an amplitude that decreases very slowly. The user multiplies y51 by 0.7 (equivalent to -30%). If y51 was equal to 1, it goes to 0.7. Then the user returns to the beginning of step 4.

[0047] Step 5: The user changes the set temperature of the frozen products and then looks at the behavior of the temperature curve of the frozen product for at least a time equivalent to 4 times the time the product spends in the freezer: o case F: the temperature varies above and below the setpoint. The amplitude of the variation decreases rapidly and the temperature ends up stabilizing very close to the setpoint. In this case, the user has precisely determined the value of y51, a parameter allowing a regulation relationship to be established between the pair of parameters described previously. ∘ case G: if the user is not in case F, then he resumes the procedure from the beginning of step 4.

[0048] The same procedure can be applied to all the X or Y parameters for regulating the matrices of the present invention, that is to say to all the pairs of parameters of the process. The user will carry out the same procedure by adjusting each X or Y parameter one by one, taking care to adjust all the other parameters to 0 at the start of the procedure.

[0049] When this entire process is completed, all the X control parameters of the process will have been determined.

[0050] Optimal regulation of the freezer's operation will then be obtained.

[0051] Let us continue in the following to illustrate the invention with a concrete example: as already mentioned, when a cryogenic tunnel operates at a production site, many parameters can change at the same time, and all these parameters have an impact on the freezing achieved.

[0052] If, for example, the incoming product flow rate increases by 20% and goes from 1000 to 1200 kg / h, for example. If, at the same time, the product temperature decreases so that the product needs 5% less cooling (for example, 95 calories / kg instead of 100).

[0053] We can then consider that overall, the tunnel will have to provide 1200 x 95 = 114,000 equivalent power compared to 100,000 before. It must therefore provide 14% more energy.

[0054] Let us summarize below what will then happen depending on the solution chosen by the tunnel user: Basic prior art solution No. 1 without feedback (most widespread system): the tunnel will operate with an unchanged power of 100% and the product will therefore be insufficiently frozen by 14%; Solution No. 2 involving a measurement of the temperature of the incoming product: the system will detect a lower demand for cold because the product arrives colder, it will reduce the tunnel power by 5% and the product will then be 19% cold lacking. We see here that the action on a single parameter has a negative impact and in any case less favorable than basic solution No. 1. Solution No. 3 involving a measurement of the incoming product flow rate: the system will detect an increase in the flow rate of 20% and will trigger an increase in the freezer power of 20%. The product will then be too cold at the outlet and the overconsumption can be estimated at 6%. The product will nevertheless be correct but this solution results in an additional production cost of 6%.solution No. 4 according to the present invention: the system measures the temperature of the incoming products and the flow rate of incoming products (and where applicable other parameters), the system will then note changes in two parameters at the same time, it will make the necessary calculations and find that the result of these two changes is equivalent to an increase in power required of 14%. It will therefore increase the power of the freezer by 14%, exactly what the process needs. The product will therefore come out correctly frozen, at the right temperature, and without overconsumption.

[0055] The invention may also adopt one or more of the following embodiments: all or part of the measurements taken are transferred to remote databases; this data, thus stored remotely, is processed to provide: i) information concerning the effectiveness of the freezing process; j) a dynamic analysis of the freezing tunnel, making it possible to provide, in particular, operating drifts, or maintenance alerts, etc.

[0056] The present invention then relates to a method of operating a cryogenic tunnel in which products to be cooled or frozen circulate, a tunnel which is equipped with means for injecting a cryogenic fluid as well as means for extracting at variable flow rate all or part of the cold gases resulting from the vaporization of said fluid in the tunnel, characterized in that: We have measurements of several parameters qualifying the process. We proceed to the distribution of these parameters in two groups of different parameters, characterizing the tunnel: a first group consisting of measured parameters which can be used to anticipate the future freezing power required by the tunnel, first group which includes one or more parameters among the following parameters: ∘ The temperature of the products entering the tunnel ∘ The volume flow rate of the products entering ∘ The mass flow rate of the products entering the tunnel ∘ The color of the products entering the tunnel ∘ The coverage rate of the conveyor belt feeding the tunnel or ∘ Parameters characterizing the atmosphere surrounding the tunnel in the room: ambient temperature, ambient humidity and atmospheric pressure. A second group of measured parameters which will be used to evaluate the final result of the products leaving,parameters which will notably indicate whether the product is correctly frozen, this second group comprising one or more of the following parameters: ∘ The temperature of the products leaving the tunnel ∘ The cryogen flow rate admitted into the tunnel o The temperature of the gases extracted from the tunnel, typically at the level of two extraction hoods located at the entrance and exit of the tunnel ∘ The temperature prevailing in the room in the vicinity of these two extraction hoods ∘ The hardness of the products leaving the tunnel ∘ The color of the products leaving the tunnel ∘ The percentage of frozen products which are of the IQF type Two types of actions are implemented on these parameters: Anticipation actions calculated according to the values ​​obtained for the parameter(s) of the 1st group, to act upstream on an anticipated / expected deviation in the freezing quality, for example because the temperature of the incoming products is too high compared to a given setpoint,anticipated actions, i.e. actions on the tunnel even before the product comes out, for example too hot or too cold, i.e. actions to modify tunnel parameters; and feedback (countermeasures) calculated based on the values ​​obtained for the parameter(s) of the second group, to rebalance an effective, measured drift in the quality of the products coming out, for example due to the fact that the temperature of the products leaving the tunnel is too high compared to a given target temperature; the actions, anticipation or feedback, being determined by the outputs of the following two matrices governing said actions: Said first group of measured parameters form an input of a first matrix, first group used to trigger actions by anticipation; Said second group of measured parameters form an input of a second matrix, second group used to trigger actions by feedback; the measurements obtained for one or certain parameters of these two groups of parameters making it possible to calculate the adjustments to be made to the group of action parameters (by anticipation and / or feedback) consisting of: ∘ The speed of the conveyor ∘ The speed of the fans internal to the tunnel ∘ The temperature of the gases coming from the inlet hood ∘ The temperature of the gases coming from the outlet hood ∘ The temperature setpoint prevailing inside the tunnel.the second entry of the two matrices being constituted by said group of action parameters, and in that the values ​​constituting the boxes of the two matrices were determined experimentally, each box of these matrices being constituted by a factor establishing a link respectively between a given parameter of the first group and a given anticipation action, and a given parameter of the second group and a given feedback.

Claims

1. A method for operating a cryogenic tunnel in which products to be cooled or frozen circulate, which tunnel is equipped with means for injecting a cryogenic fluid as well as means for extracting all or part of the cold gases resulting from the vaporization of said fluid in the tunnel, characterized in that: Measurements of several parameters qualifying the method are available. These parameters are distributed into two different groups of parameters, characterizing the tunnel: • a first group consisting of measured parameters that can and will be used to anticipate the future freezing power required by the tunnel, this first group comprising one or more parameters from the following parameters: ∘ The temperature of products entering the tunnel ∘ The volumetric flow rate of products entering ∘ The mass flow rate of products entering the tunnel ∘ The color of products entering the tunnel ∘ The coverage rate of the conveyor belt feeding the tunnel or ∘ Parameters characterizing the atmosphere surrounding the tunnel in the room: ambient temperature, ambient humidity and atmospheric pressure. • A second group consisting of measured parameters that will be used to evaluate the final result of outgoing products, parameters that will notably indicate whether the product is properly frozen, this second group comprising one or more parameters from the following parameters: ∘ The temperature of products leaving the tunnel ∘ The cryogenic flow rate admitted into the tunnel ∘ The temperature of gases extracted from the tunnel, typically at the level of two extraction hoods located at the entrance and exit of the tunnel ∘ The temperature prevailing in the room near these two extraction hoods ∘ The hardness of products leaving the tunnel ∘ The color of products leaving the tunnel ∘ The percentage of IQF-type frozen products Two types of actions are implemented on these parameters: • Anticipation actions calculated from the values obtained for the parameter(s) of the 1st group, to act upstream on an anticipated / expected deviation in freezing quality, for example because the temperature of the incoming products is too high compared to a given setpoint; and • Feedback actions (counter-measures) calculated from the values obtained for the parameter(s) of the second group, to rebalance an effective, measured deviation in the quality of the outgoing products, for example due to the fact that the temperature of the products at the tunnel exit is too high compared to a given target temperature; the anticipation or feedback actions being determined by the outputs of the following two matrices governing said actions: • Said first group of measured parameters forms an input of a first matrix, the first group used to trigger anticipation actions; • Said second group of measured parameters forms an input of a second matrix, the second group used to trigger feedback actions; the measurements obtained for one or certain parameters of these two groups of parameters allowing the calculation of adjustments to be made to the group of action parameters (by anticipation and / or feedback) consisting of: ∘ The conveyor speed ∘ The speed of internal tunnel fans ∘ The temperature of gases from the entry hood ∘ The temperature of gases from the exit hood ∘ The temperature setpoint inside the tunnel. the second input of the two matrices being constituted by said group of action parameters, and in that the determination of the values constituting the cells of the two matrices was carried out experimentally, each cell of these matrices being constituted by a factor establishing a link respectively between a given parameter of the first group and a given anticipation action, and a given parameter of the second group and a given feedback action.

2. A method of operation according to claim 1, characterized in that: - said first group of parameters comprises at least two parameters from the following parameters: ∘ The temperature of products entering the tunnel ∘ The volumetric flow rate of products entering ∘ The mass flow rate of products entering the tunnel ∘ The color of products entering the tunnel ∘ The coverage rate of the conveyor belt feeding the tunnel or ∘ Parameters characterizing the atmosphere surrounding the tunnel in the room: ambient temperature, ambient humidity and atmospheric pressure. and in that: - said second group of parameters comprises at least two parameters from the following parameters: ∘ The temperature of products leaving the tunnel ∘ The cryogenic flow rate admitted into the tunnel ∘ The temperature of gases extracted from the tunnel, typically at the level of two extraction hoods located at the entrance and exit of the tunnel ∘ The temperature prevailing in the room near these two extraction hoods ∘ The hardness of products leaving the tunnel ∘ The color of products leaving the tunnel ∘ The percentage of IQF-type frozen products