Method for measuring the dead volume of an anaerobic reactor

Abstract

A method for measuring the dead volume of an anaerobic reactor used to transform an input comprising organic matter into biogas, consists of: - modeling (1) the residence time of the materials in the reactor as a function of: > the maximum degradation potential of the inputs, > the actual degradation of the inputs in the reactor, > a parameter characteristic of the degradation kinetics of the inputs in the predetermined type of reactor, taking into account its operating conditions, - measuring (2): > the maximum degradation potential of the inputs, > an indicator of degradation of the inputs within the reactor, > a parameter characteristic of the degradation kinetics of the inputs in the predetermined type of reactor and its operating conditions, - calculating (3) the estimated residence time of the materials in the reactor from the model and the measurements,- calculate (4) the dead volume from the estimated residence time and the predetermined reactor type. The method allows monitoring of the evolution of the dead volume of a reactor. Figure for abstract: figure 1,

Description

Title of the invention: Method for measuring the dead volume of an anaerobic reactor

[0001] The invention relates to the field of the recovery into biogas, by anaerobic digestion, of organic carbon present in municipal or industrial sludge, organic waste or a mixture thereof. More particularly, the invention relates to a method for measuring the dead volume of a reactor and a method for monitoring a reactor.

[0002] A method for treating organic substrates is already known, for example from document FR3044324. These substrates may be sludge from the treatment of domestic or industrial wastewater. The sludge is subjected to a biological anaerobic digestion treatment, which reduces its volume and produces biogas. The biological degradation process is carried out inside a closed, airtight reactor, also called an anaerobic reactor or anaerobic digester, through which the sludge circulates for a given residence time, undergoing a series of biological transformations. To operate correctly, an anaerobic reactor is temperature-controlled and its contents are stirred.

[0003] Despite all the precautions taken for good sludge transformation, solid matter accumulates in the reactor. This is called silting. These sands occupy a so-called dead volume in the tank which is no longer available for the sludge transformation reaction. In the long term, this dead volume ends up having a significant impact on the productivity of the reactor. The drop in biogas production due solely to silting can be very significant in the long term, sometimes up to 20%.

[0004] It is known to measure the amount of dead volume, to decide whether it is useful to stop the reactor to evacuate the accumulated sand. This operation is obviously costly, not only as such, but also in terms of loss of production. It is therefore advisable to decide on it only after having carefully verified its usefulness.

[0005] Among the known verification techniques, the most reliable to date and also the most widespread is lithium tracing. This conventional method, which is used in various fields in which a fluid flow must be studied, consists of injecting precise doses of lithium at different times, then measuring the lithium concentration at the reactor outlet at different times over a relatively long measurement period. Since lithium is inert with respect to the decomposition of the sludge, it is not affected by the chemical reactions that occur in the reactor. Its concentration at the reactor outlet therefore depends solely on the speed at which it passes through the reactor or, strictly equivalently since its progress in the reactor is not known but only its entry and exit times, on its residence time in the reactor. A residence time distribution curve is thus established. The model described by Levenspiel (see Levenspiel, "Chemical Reaction Engineering", John Wiley and Sons, New York, 1962, pages 242 to 308) is applied to quantitatively distribute the reactor contents between fully mixed active volume, proportion of inactive mixture and proportion of inputs that bypass these two components to short-circuit the reactor outlet. From the residence time distribution curve, the dead volume can be deduced and, if necessary, its magnitude confirmed to justify a reactor shutdown.

[0006] One difficulty with lithium tracing is its very high cost. As a result, many operating sites are reluctant to apply it.

[0007] In addition, the temporality of the application of the tracing method is often unfavorable: lithium tracing being expensive, it is only applied when a problem is proven on the site. In general, the presence of too high a dead volume is strongly suspected due to a significant and persistent drop in production. The tracing then only serves to confirm the hypothesis of silting, but a significant drop in production has already been suffered. It is therefore too late to limit losses. The tracing method did not have the utility of detecting excessive silting. It only served to confirm it. Conversely, sites that want to avoid taking the measurement too late follow a predefined schedule. They then run the risk of carrying out the lithium tracing too early, therefore also in a useless manner. Lithium tracing cannot therefore be considered as a method for monitoring silting.

[0008] Another difficulty with lithium tracing is the potential pollution resulting from its implementation. Although the quantities injected are small and fixed by the standards in force, the lithium used is difficult to recover at the reactor outlet. It will eventually be released, fortunately in small quantities, into the environment, where its great stability will cause it to persist for a long time.

[0009] There is therefore a need for a method for monitoring the dead volume of a reactor that is simple, rapid, inexpensive and non-polluting. The lithium tracing solution is certainly technically available as a measurement method, but the practical conditions of its implementation make it unavailable as a preventive monitoring method.

[0010] The level of silting can also be read a posteriori on production curves of a reactor observed over several years. However, production monitoring is not a reliable method for monitoring silting either. With monthly production reports, the trend is easily masked by the overriding influence of other factors of variation, starting with the composition of inputs, which is not absolutely stable. Furthermore, the accuracy of the measurements is often not sufficient to detect the impact of short-term silting. Finally, the phenomenon is not linear. Silting progresses more quickly at times, depending on various factors.

[0011] The invention therefore aims to provide a real practical solution to the need for preventive monitoring of silting of an anaerobic reactor.

[0012] The subject of the invention is a method for measuring the dead volume of an anaerobic reactor of predetermined type, used to transform inputs comprising organic matter into biogas and digestate, characterized in that it comprises the following steps:

[0013] - model the residence time of the materials in the reactor as a minimum function of the following parameters:

[0014] > maximum potential for degradation of inputs,

[0015] > degradation of inputs effective in the reactor,

[0016] > characteristic parameter of the kinetics of degradation of inputs in the type of predetermined reactor, taking into account its operating conditions,

[0017] - measure the following values:

[0018] > maximum potential for degradation of inputs,

[0019] > indicator of degradation of inputs within the reactor,

[0020] > characteristic parameter of the degradation kinetics of inputs in the type of predetermined reactor and its operating conditions,

[0021] - calculate the estimated residence time of the materials in the reactor from the model and measurements, and the nominal residence time of the materials in the reactor taking into account the predetermined reactor type,

[0022] - calculate the dead volume from the nominal residence time, the residence time estimated and predetermined reactor type.

[0023] The order of the steps is not limiting, since an intermediate result necessary for a given step has been obtained by prior execution of a step for obtaining this intermediate result.

[0024] This process can be implemented regularly and preventively. It then makes it possible to identify deviations before they become critical, all without price or pollution being a barrier to its application.

[0025] In the present application, the composition of the inputs is considered to be reasonably stable because the inputs do not vary from day to day and because the measurements of the maximum degradation potential of the inputs and the degradation kinetics of the inputs can correspond to the degradation indicator of the inputs which is preferably measured on site. Stability is understood to be over a few weeks. In practice, this condition is almost always met.

[0026] The maximum input degradation potential is the maximum capacity of the reactor to degrade the inputs in the absence of oxygen. This parameter, respectively this measurement, corresponds to the maximum theoretical quantity of organic matter that the reactor can treat by the biological processes activated within it.

[0027] An advantage of the above method is first of all to constitute a rapid, inexpensive and non-polluting tool allowing preventive monitoring of the reactor.

[0028] Furthermore, the method provides additional information, such as the maximum degradation potential of the inputs, which is useful for monitoring and optimizing biological treatments.

[0029] According to other optional characteristics of the method, taken alone or in combination:

[0030] - The reactor type can be defined by:

[0031] - its flow type (piston flow or infinitely mixed),

[0032] - its power supply regime (continuous, semi-continuous),

[0033] - its withdrawal regime (continuous, semi-continuous),

[0034] - the number of stages (if reactors in series),

[0035] - the existence of one or more recirculation loops.

[0036] The reactor model represents its fluidics. Depending on the complexity of the latter, the resolution of the model can be explicit, that is to say obtained by inversion of a direct formula, or iterative, by calculation applied to successive time steps.

[0037] - As the maximum potential for degradation of inputs, which corresponds to the maximum capacity of the reactor to degrade inputs in the absence of oxygen, any of the following quantities is used: maximum biochemical methane potential; maximum dry matter reduction potential; maximum volatile matter reduction potential; maximum chemical oxygen demand reduction potential.

[0038] Advantageously, to measure it, a biological method or an infrared measurement is used.

[0039] - As an indicator of effective degradation of inputs in the reactor, we use methane yield, i.e. the amount of methane produced per unit mass of input, or the efficiency of reducing volatile matter, dry matter or chemical oxygen demand.

[0040] Advantageously, it is measured on site.

[0041] - As a characteristic parameter of the kinetics of degradation of inputs in the predetermined type of reactor, taking into account its operating conditions, the hydrolysis rate is used.

[0042] Advantageously, to measure it, a biological method or an infrared measurement is used, or the formula available in the calculation according to the ISO 19388 standard, namely: khyd = 0.045« 1.072T10.

[0043] - In the case of a piston flow reactor, we use as an additional parameter supplementary to the model, respectively as an additional measure, the reactor recycling rate.

[0044] The repeated implementation of the method is made possible by the absence of disruption to production and by the low cost of each repetition. Thanks to these advantages of the measuring method, it is possible to carry out real preventive monitoring of sand encroachment.

[0045] Thus, the invention also relates to a method for monitoring the evolution of the dead volume of an anaerobic reactor used to transform inputs comprising organic matter into biogas and digestate, consisting of repeating, at regular time intervals, the following steps: carrying out the measurement method described above; comparing the results obtained with a predetermined reference; reporting a deviation, if applicable.

[0046] In addition to providing a method for monitoring the silting of a reactor, repeated implementation of the method makes it possible to compensate for a variation in production linked to changes in input quality. Indeed, the proposed method requires measuring again, at each application, the maximum degradation potential of the inputs and the characteristic parameter of the degradation kinetics of the inputs. It is then possible to decouple the part of performance linked to a variation in the inputs and that linked to the dead volume. Thus, even if the performance of the reactor increases due to a supply by inputs producing more methane than in the past, an increase in the dead volume remains detectable. The proposed method then becomes more advantageous than monitoring only the performance of the site. The monitoring method is therefore more independent of the stability of the inputs than any other known method.It is recalled that the stability of the inputs is understood, according to the invention, as satisfied as soon as there is a consistency between the maximum potential for degradation of the inputs, the kinetics of degradation of the inputs and the indicator of degradation of the inputs. In the context of the monitoring method, this consistency results from the fact that the measurements are repeated at each application of the measurement method described above.

[0047] Furthermore, another advantage of this repetition of implementation of the measurement method is that it makes it possible to compensate for the lower precision, if we compare it to that of the lithium tracing method.

[0048] The present invention also relates to computer processing means for monitoring a methanization unit comprising an anaerobic reactor, said computer processing means comprising at least one memory and at least one processor, the memory containing instructions which, when read and executed by the processor, enable the processor to implement the method for measuring the dead volume of the reactor.

[0049] In a particular embodiment, the memory also contains instructions which, when read and executed by the processor, allow the processor to implement the method for monitoring the evolution of the dead volume of the reactor.

[0050] The present invention also relates to a methanization unit from organic materials, comprising an anaerobic reactor equipped with a reactor and computer processing means as described above.

[0051] The invention can also be expressed according to a slightly different approach which achieves the same result, by modeling the production and not the residence time.

[0052] The invention thus also relates to a method for measuring the dead volume of an anaerobic reactor of predetermined type, used to transform inputs comprising organic matter into biogas and digestate, comprising the following steps:

[0053] - model the reactor production in the form of a degradation indicator inputs within the reactor depending on at least the following parameters:

[0054] > residence time of the materials in the reactor,

[0055] > maximum potential for degradation of inputs,

[0056] > characteristic parameter of the kinetics of degradation of inputs in the type of predetermined reactor, taking into account its operating conditions

[0057] - measure the following values:

[0058] > maximum potential for degradation of inputs,

[0059] > characteristic parameter of the degradation kinetics of inputs in the type of predetermined reactor and its operating conditions,

[0060] > indicator of effective degradation of inputs in the reactor.

[0061] - solve the production model by injecting the measured values into it to deduce the residence time of the materials in the reactor,

[0062] - deduce from the residence time the useful volume of the reactor depending on the type of predetermined reactor,

[0063] - obtain the dead volume by subtracting the useful volume from the total volume of the reactor.

[0064] This measurement method according to an alternative approach can also be combined with the secondary characteristics of the measurement method described above, as well as with the other objects of the invention, namely the monitoring method, the computer processing means and the methanization unit. Brief description of the figures

[0065] The invention will be better understood on reading the following description, given solely by way of example and with reference to the appended drawings in which:

[0066] [Fig-1] is a block diagram illustrating the steps of a measurement method according to a example of implementation,

[0067] [Fig.2] is another block diagram illustrating the steps of a reactor monitoring method,

[0068] [Fig.3] is a schematic view of a methanization unit according to a particular embodiment. Detailed description

[0069] The described example of the dead volume measurement method is illustrated by [Fig.l], each block of which corresponds to a step of said method.

[0070] In a first step 1, we start by establishing the model to be used. The performance of the reactor is taken as an input variable to predict the residence time.

[0071] Since the microorganisms involved in the different phases of digestion have very different generation times (Table 1), the selected solids retention time must be greater than the generation time of the slowest growing microbial group. In this case, these are methanogens. Therefore, solids retention times greater than 5 days, generally 15 to 20 days for mesophilic digestion and 8 to 12 days for thermophilic digestion, are selected. [Tables 1] Parameter Hydrolysis / Acidogenesis Methanogenesis Temperature 25 - 35°C Mesophilic: 30 - 42°C Thermophilic: 50 -58°C PH 5.2-6.3 6.7-7.5 Redox potential +400 to -300 mV <-250mV Required C:N:P:S ratio 500 :15:5:3 600 :15:5:3 Generation time 24 - 36 h 5-16 days Trace element - Ni, Co, Mo, Se

[0072] The reactor size chosen aims to maximize energy recovery and sludge stabilization at approximately 80 to 90% of the biochemical methane potential (BMP), while minimizing investment and washing of methanogenic biomass. Indeed, in anaerobic digestion of municipal sewage sludge, once methanogens have developed, hydrolysis is considered the rate-limiting step of the process and follows a first-order rate.

[0073] This phenomenon can be modeled using the relationship between methane production and solids retention time (SRT) in a continuous stirred tank reactor:

[0074] [Math.l] Y CH 4 ( 1" 1+ kltV(tSRT )

[0075] where:

[0076] YCh4, is the methane yield (NLCH4 / kgVS)

[0077] Bo, is the maximum biochemical methane potential -BMP- (NLCH4 / kgVS),

[0078] khyd, is the apparent hydrolysis rate,

[0079] SRT is the retention time of solids in the reactor.

[0080] From the previous formula, the solid residence time model (SRTmodei) can be established according to the following formula:

[0081] [Math.2] = J- ■ (pUy -1)

[0082] During a second step 2, the variables necessary for the application of the model are measured, namely: - the maximum potential for degradation of inputs (Bo), - the indicator of the effective degradation of inputs in the reactor (YCh4) and - kinetics (khyd, which depends minimally on temperature).

[0083] Tests on the biochemical methane potential (BMP) are expensive and long (>30 days). Therefore, to determine the Bo and advantageously the khyd, it is intended to use any method aimed at predicting this parameter. For example, the person skilled in the art knows and can choose from the following methods, which belong to the state of the art: - Correlation tests with aerobic: R. Cossu, R. Raga, Test methods for assessing the biological stability of biodegradable waste, Waste Manag. 28 (2008) 381e388, https: / / doi.Org / 10.1016 / j.wasman.2007.01.014

[0084] S. Pons a, T. Gea, L. Alerm, J. Cerezo, A. Sanchez, Comparison of aerobic and anaerobic stability indices through a MSW biological treatment process, Waste Manag. 28 (2008) 2735e2742, https: / / doi.Org / 10.1016 / j.wasman.2007.12.002 - Regression models using physicochemical characteristics such as input data: V. Dandikas, H. Heuwinkel, F. Lichti, JE Drewes, K. Koch, Predicting methane yield by linear regression models: a validation study for grassland biomass, Bioresour. Technol. 265 (2018) 372e379, https: / / doi.Org / 10.1016 / j.biortech.2018.06.030

[0085] L. Appels, J. Lauwers, G. Gins, J. Degreve, J. Van Impe, R. Dewil, Parameter identification and modeling of the biochemical methane potential of waste activated sludge, Environ. Sci. Technol. 45 (9) (2011) 4173e4178, https: / / doi.org / 10.1021 / esl037113

[0086] A. Mottet, E. François, E. Latrille, J.P. Steyer, S. Deleris, F. Vedrenne, H. Carrere, Estimating anaérobie biodegradability indicators for waste activated sludge, Chem. Eng. J. 160 (2010) 488e496, https: / / doi.Org / 10.1016 / j.cej.2010.03.059

[0087] F. Xu, Z. Wang, Y. Li, Predicting the methane yield of lignocellulosic biomass in mesophilic solid-state anaérobie digestion based on feedstock characteristics and process parameters, Bioresour. Technol. 173 (2014) 168el76, https: / / doi.org / 10.1016 / j .biortech.2014.09.090 - Techniques de spectroscopie, y compris la spectroscopie proche infrarouge : M. Lesteur, E. Latrille, V.B. Maurel, J.M. Roger, C. Gonzalez, G. Junqua, J.P. Steyer, First step towards a fast analytical method for the détermination of Biochemical Methane Potential of solid wastes by near infrared spectroscopy, Bioresour. Technol. 102 (2011) 2280e2288, https: / / doi.Org / 10.1016 / j.biortech.2010.10.044

[0088] JM Triolo, AJ Ward, L. Pedersen, MM Lpkke, H. Qu, SG Sommer, Near Infrared Reflectance Spectroscopy (NIRS) for rapid determination of biochemical methane potential of plant biomass, Appl. Energy 116 (2014) 52e57, https: / / doi.Org / 10.1016 / j.apenergy.2013.ll.006

[0089] All of these methods provide estimates of anaerobic biodegradability within reasonable time frames and margins of error.

[0090] The indicator of the effective degradation of the inputs in the reactor, in the example described the quantity of methane produced per unit mass of YCH4 input, is determined from the site production data.

[0091] The characteristic parameter of the degradation kinetics of the inputs, in the example described the hydrolysis rate khyd, is determined by the above-mentioned rapid methods or using the following formula, available in the calculation according to the ISO 19388 standard:

[0092] [Math.3] ^ = 0.045-1.072™°

[0093] The measurement of khyd is therefore indirect, the direct measurement being that of temperature T.

[0094] Optionally, other parameters, not detailed here but known to the specialist in the field, can be integrated to represent the fluidics of the reactor.

[0095] Note that Bo and khyd (or their equivalents) are measured from input samples, while YCH4 (or equivalent) comes from reactor monitoring measurements.

[0096] In a following step 3, two values of the residence time are calculated.

[0097] On the one hand, the residence time model is used by applying the formula [Math 2] with the measured variables. We thus obtain a value of the estimated residence time, SRTmodei.

[0098] On the other hand, the nominal residence time, or nominal solids retention time (SRTnominai) is calculated taking into account the predetermined reactor type.

[0099] Solids retention time (SRT) is defined as the average time solids spend in a reactor. Wet anaerobic digestion (<10% solids in feedstock) typically uses continuous stirred tank reactors, in which the solids retention time (SRT) and hydraulic retention time (HRT) are the same and calculated as the reactor volume (m3) divided by the daily flow rate (mVday). Thus, in this example, the predetermined reactor type is taken into account.

[0100] The calculation of the nominal retention time of solids SRTnominaire therefore results from the application of the following formula:

[0101] [Math.4] ç dt _ V named Q

[0102] where:

[0103] V is the total volume of the reactor (m3),

[0104] Q is the flow rate of the raw material introduced into the reactor (m3 / day).

[0105] In a final step 4, the dead volume is calculated by taking into account the nominal residence time SRTnominai and the estimated residence time SRTmodei, using the following formula:

[0106] [Math.5] Km = (SRT- SRT,,^ ■ Q

[0107] It can be seen that the order of the steps is not limiting, since a result necessary for a given step has been obtained by prior execution of a step for obtaining this result.

[0108] Equation [Math 1] represents the simplest case, for which a direct formula can be obtained. This equation is theoretically valid only for a continuously stirred tank reactor in steady state. But the method can be extended to other configurations.

[0109] According to another example, the reactor is of the steady-state piston flow type. The following formula must then be substituted for formula [Math 1]:

[0110] [Math.6] 1-R exp(khyâSRT.( 1-R) )-R [YES] where:

[0112] YCh4, is the methane yield (NLCH4 / kgVS)

[0113] Bo, is the maximum biochemical methane potential -BMP- (NLCH4 / kgVS),

[0114] khyd, is the apparent hydrolysis rate,

[0115] SRT is the retention time of solids in the reactor,

[0116] R is the recycle rate of the plug flow reactor.

[0117] Using equation [Math 6], the estimated retention time of solids SRTmodei can be estimated using the following equation:

[0118] [Math.7] Sr— ------- • / / 71 RA——■ I model 1 I

[0119] Alternatively, for non-ideal hydraulic fluidics or if the reactors have not reached a steady state, two-parameter first-order dynamic models (khyd, Bo) can be used. In this case, the model aims to fit a time series of biomethane production instead of a single value of YCH4-

[0120] In [Fig.2], a block diagram of an implementation of a method for monitoring the evolution of the dead volume of the reactor is shown.

[0121] A first step 5 represents a complete execution of the measurement method of [Fig.l]. At each execution, all the measurements are renewed.

[0122] A second step 6 consists of comparing the results obtained during step 5 with a predetermined reference.

[0123] A third step 7 consists of reporting a deviation, if applicable.

[0124] At the end of step 7, the process loops back to step 5, until its forced interruption. (not shown) by an operator.

[0125] In [Fig.3], we see a methanization unit 8 which comprises a reactor 9, computer processing means 10 including a memory 11 and a processor 12. The memory contains instructions 13. When they are read and executed by the processor 12, these instructions 13 allow the processor 12 to implement the method for measuring dead volume of [Fig.1], as well as the method for monitoring the evolution of the dead volume of [Fig.2].

[0126] The invention is not limited to the examples described, which are only intended to facilitate understanding. List of references

[0127] 1: .. .residence time modeling step 2: .. .measurement step 3: .. .step of calculating the nominal residence time 4: .. .dead volume calculation step 5: .. .step of execution of the measurement process 6: .. .results comparison step 7: .. .step of reporting a deviation 8: .. .methanization unit 9: ... reactor 10: ...computer processing means 11: ...memory 12: ...processor 13: ...instructions

Claims

Claims

1. Method for measuring the dead volume of an anaerobic reactor of predetermined type, used to transform inputs comprising organic matter into biogas and digestate, characterized in that it comprises the following steps: - modeling (1) the residence time of the materials in the reactor as a minimum function of the following parameters: > maximum potential for degradation of the inputs, > degradation of the inputs effective in the reactor, > characteristic parameter of the kinetics of degradation of the inputs in the predetermined type of reactor, taking into account its operating conditions, - measuring (2) the following values: > maximum potential for degradation of the inputs, > indicator of degradation of the inputs within the reactor, > characteristic parameter of the kinetics of degradation of the inputs in the predetermined type of reactor and its operating conditions,- calculate (3) the estimated residence time of the materials in the reactor from the model and the measurements, and the nominal residence time of the materials in the reactor taking into account the predetermined reactor type, - calculate (4) the dead volume from the nominal residence time, the estimated residence time and the predetermined reactor type.,

2. A method according to claim 1, wherein, as maximum input degradation potential, any one of the following quantities is used: maximum biochemical methane potential (BMP); maximum dry matter reduction potential; maximum volatile matter reduction potential; maximum chemical oxygen demand reduction potential.

3. Method according to the preceding claim, in which the maximum degradation potential of the inputs is measured by biological method or by infrared measurement.

4. A method according to any one of the preceding claims, wherein, as an indicator of effective input degradation in the reactor, the methane yield (YCH4) is used, i.e. the quantity of methane produced per unit mass of input, or the volatile matter, dry matter or chemical oxygen demand reduction efficiency

5. Method according to any one of the preceding claims, in which, as a characteristic parameter of the kinetics of degradation of the inputs in the predetermined type of reactor, taking into account its operating conditions, the hydrolysis rate (khyd) is used.

6. Method according to the preceding claim, to measure the characteristic parameter of the degradation kinetics of the inputs in the predetermined type of reactor, a biological method or an infrared measurement is used, or the formula available in the calculation according to the ISO 19388 standard, namely: khyd = 0.045«1.072T 10.

7. Method according to any one of the preceding claims, adapted to a plug flow reactor, in which the reactor recycling rate (R) is used as an additional parameter of the model, respectively as an additional measure.

8. Method for monitoring the evolution of the dead volume of an anaerobic reactor used to transform inputs comprising organic matter into biogas and digestate, characterized in that it consists of repeating, at regular time intervals, the following steps: carrying out (5) the measurement method according to any one of the preceding claims; comparing (6) the results obtained with a predetermined reference; signaling (7) a deviation, if applicable.

9. Computer processing means for monitoring a methanization unit (8) comprising an anaerobic reactor (9), said computer processing means (10) comprising at least one memory (11) and at least one processor (12), characterized in that the memory (11) contains at least instructions (13) which, when read and executed by the processor (12), allow the processor (12) to implement the method for measuring the dead volume of the reactor according to any one of claims 1 to 7 and possibly instructions (13) which, when read and executed by the processor (12), allow the processor (12) to implement the method for monitoring the evolution of the dead volume of the reactor according to claim 8.

10. Methanization unit (8) from organic materials, comprising an anaerobic reactor (9) and computer processing means according to claim 9.

Images