Method for estimating the operating status of methane fermentation equipment, method for predicting gas generation rate, control method and device, and computer program

Abstract

To provide a methane fermentation facility capable of estimating an operating state with a high degree of accuracy and realizing the operating state desired by an operator.SOLUTION: A method for estimating an operating state of a methane fermentation facility includes steps of: expressing a methane fermentation model with a state space model using at least some variables of the methane fermentation model as state variables; obtaining observed values in a methane fermentation facility (100) including a fermenter 140; and sequentially estimating the state variables by applying iterative estimation filters to the state space model using the observed values.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for estimating the operational state of a methane fermentation facility, a method for predicting the amount of gas generated, a control method and device, and a computer program, and in particular to a method for estimating the operational state of a methane fermentation facility that is suitable for use in a methane fermentation plant and is capable of estimating the operational state with high accuracy, a method for predicting the amount of gas generated using the results of the operational state estimation method, a control method and device, and a computer program for causing a computer to execute or realize the method and device. [Background technology]

[0002] Methane fermentation is a process in which organic waste and other materials are fermented under anaerobic conditions using methanogens to break them down into biogas containing methane and water. This process has the advantage of significantly reducing the amount of organic matter input, and also of recovering the methane gas produced as energy. However, the methanogens that synthesize methane gas have limited activity under certain conditions, such as temperature and pH, depending on the type. High ammonia and hydrogen concentrations reduce activity, so conditions must be managed appropriately during operation. Several technologies for this purpose have been proposed (Patent Documents 1 and 2).

[0003] Regarding prediction of methane gas generation amount, Patent Document 3 describes a technique for predicting the methane gas generation amount in a methane fermentation process (anaerobic digestion) using a methane fermentation model. Also, Non-Patent Document 1 shows an example of a methane fermentation model. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 4218486 [Patent Document 2] Patent No. 6697201 [Patent Document 3] Japanese Patent Application Laid-Open No. 2005-111338 [Non-patent literature]

[0005] [Non-Patent Document 1] Anaerobic Digestion Model No. 1, Scientific and Technical Report 15. IWA Publishing, 2002 Summary of the Invention [Problem to be solved by the invention]

[0006] However, even when using a methane fermentation model, it was difficult to predict the amount of methane gas produced with sufficient accuracy, because even when using a methane fermentation model, it was not possible to accurately grasp the actual conditions inside the methane fermentation tank.

[0007] In particular, there was a problem in that it was difficult to accurately predict the amount or concentration of biodegradable organic matter (proteins, carbohydrates, lipids, etc.), organic acids, acid-producing bacteria, and methanogens in the methane fermentation tank.

[0008] The present invention has been made to solve the above-mentioned conventional problems, and its objective is to accurately estimate the operating status of a methane fermentation facility, including the concentrations of biodegradable organic matter (proteins, carbohydrates, lipids, etc.), organic acids, acid-producing bacteria, and methanogens in the fermentation tank, and further to enable highly accurate prediction of the amount of methane gas generated, thereby enabling appropriate control. [Means for solving the problem]

[0009] The present invention includes a step of expressing a methane fermentation model as a state space model using at least some variables of the methane fermentation model as state variables, a step of obtaining observed values ​​in a methane fermentation facility including a fermenter, and a step of calculating the observed values. and the estimated value of the observation vector based on the observation equation in the state space model. Applying a recursive estimation filter (recursive Bayes filter) to the state space model Therefore, The state variables are sequentially estimated. so that the estimated value of the state variable approaches the true value.The above-mentioned problems are solved by a method for estimating the operational state of a methane fermentation facility, which comprises the steps of:

[0010] Here, the state variables can be some or all of the concentrations of at least the biodegradable organic matter in the fermenter, the organic acids in the fermenter, the acid-producing bacteria in the fermenter, and the methanogens in the fermenter, and the observed values ​​can be some or all of the amount of methane gas generated from the fermenter and the organic acid concentrations in the fermenter.

[0011] Furthermore, the state variables may be some or all of the concentrations of at least the biodegradable organic matter for each of the multiple raw material types in the fermenter, the organic acids in the fermenter, the acid-producing bacteria in the fermenter, and the methanogens in the fermenter, and the observed values ​​may be some or all of the flow rate of the raw material input to the fermenter, the biodegradable organic matter concentrations for each of the multiple raw material types in the input raw material, the organic acid concentrations for each of the multiple raw material types in the input raw material, the amount of methane gas generated from the fermenter, the flow rate of the fermentation liquid discharged from the fermenter, the temperature of the fermenter, and the amount of fermentation liquid in the fermenter.

[0012] The present invention also solves the above-mentioned problem by using the method for estimating the operational state of a methane fermentation facility to obtain estimated values ​​for some or all of the biodegradable organic matter concentration in the fermenter, the organic acid concentration in the fermenter, the acid-producing bacteria concentration in the fermenter, and the methanogen concentration in the fermenter at a certain point in time, reflecting the estimated values ​​in a state space model, and applying to the state space model at least some or all of the flow rate of raw material input to the fermenter, the biodegradable organic matter concentration for each of a plurality of raw material types in the input raw material, and the organic acid concentration for each of a plurality of raw material types in the input raw material, thereby predicting the amount of methane gas generated from a certain point in time onwards.

[0013] The present invention also solves the above problem by controlling the methane fermentation equipment so that the future methane gas generation amount becomes a target value, using the estimated results obtained by the method for estimating the operational state of the methane fermentation equipment or the predicted results obtained by the method for predicting the gas generation amount of the methane fermentation equipment.

[0014] The present invention also provides a computer program for causing a computer to execute any of the above methods.

[0015] The present invention also provides an apparatus for predicting the operational state of a methane fermentation facility using a methane fermentation model, comprising: a means for expressing the methane fermentation model as a state space model using at least some variables of the methane fermentation model as state variables; a means for obtaining observed values ​​in a methane fermentation facility including a fermenter; and a means for expressing the observed values. and the estimated value of the observation vector based on the observation equation in the state space model. applying a recursive estimation filter to the state space model; Therefore, The state variables are sequentially estimated. so that the estimated value of the state variable approaches the true value. and a means for estimating the operational state of a methane fermentation facility.

[0016] Here, the estimation may be performed by applying a nonlinear Kalman filter.

[0017] The present invention also provides a gas generation amount prediction device for a methane fermentation facility, which is characterized by having a means for predicting the gas generation amount from a certain point in time onwards using the estimation results from the operational state estimation device for the methane fermentation facility.

[0018] The present invention also provides a control device for a methane fermentation facility, characterized by comprising means for controlling the methane fermentation facility so that the future gas generation amount becomes a target value, using the prediction results from the operation state estimation device for the methane fermentation facility or the prediction results from the gas generation amount prediction device for the methane fermentation facility.

[0019] The present invention also provides a computer program for causing a computer to implement any of the above-described devices. [Effects of the Invention]

[0020] According to the present invention, in addition to utilizing a methane fermentation model, a state space model is constructed in which at least some or all of the concentrations of biodegradable organic matter, organic acids, acid-producing bacteria, and methanogens in the fermenter are used as state variables, and these are sequentially estimated using observed values ​​observed in the plant to accurately grasp the state in the methane fermenter, making it possible to predict the amount of methane gas generated with high accuracy. Therefore, it becomes possible to perform appropriate control even in methane fermentation plants where fermentation control is difficult due to the wide variety of input raw materials. Furthermore, it becomes possible to predict future methane gas generation rates according to the raw material input plan, making it possible to accurately control the amount of methane gas generation in a planned manner. [Brief explanation of the drawings]

[0021] [Figure 1] 1 is a flowchart showing the processing procedure of the present invention; [Figure 2] Overall configuration diagram showing an embodiment of the present invention applied to a methane fermentation plant [Figure 3] An example of an outline of a methane fermentation model [Figure 4] A diagram showing an example of the equations that make up the methane fermentation model [Figure 5] A diagram showing the concept of sequential state estimation using a Kalman filter [Figure 6] FIG. 10 is a diagram showing an example of prediction of methane gas generation amount according to the embodiment. [Figure 7] This figure also shows an example of plant operation based on predicted methane gas flow rates. DETAILED DESCRIPTION OF THE INVENTION

[0022] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the contents described in the following embodiments and examples. Furthermore, the constituent elements in the embodiments and examples described below include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the so-called equivalent range. Furthermore, the constituent elements disclosed in the embodiments and examples described below may be appropriately combined or appropriately selected for use.

[0023] The basic processing procedure of the present invention is shown in FIG.

[0024] First, in step 1000, some of the variables of the methane fermentation model (some or all of the concentrations of biodegradable organic matter in the fermenter, organic acids in the fermenter, acid-producing bacteria in the fermenter, and methanogens in the fermenter) are used as state variables to represent the methane fermentation model as a state space model.

[0025] Next, in step 1100, the state variables are sequentially estimated by applying a sequential estimation filter (for example, a Kalman filter) to the state space model using observed values ​​obtained in the methane fermentation facility.

[0026] Next, in step 1200, a state space model reflecting the state variables estimated successively is used to predict the amount of methane gas generated from the methane fermentation tank, and the operation is controlled in a planned manner.

[0027] <Overall structure> FIG. 2 shows the overall configuration of an embodiment of the present invention applied to a methane fermentation plant.

[0028] This methane fermentation plant (also referred to as an actual plant or real plant) 100 is equipped with a crusher 120 for crushing the charged (e.g., thrown) raw materials and removing unsuitable materials, a mixing tank 130 for mixing the raw materials with water to prepare them, a fermenter 140 for anaerobic fermentation of the raw materials sent from the mixing tank 130 to generate biogas containing methane, an observation data storage unit 200 for inputting and storing operation data and analysis values ​​of the methane fermentation plant 100, an operation planning unit 210 for inputting operation conditions and targets, and a calculation / prediction / control unit 220 for performing calculation / prediction / control using a state space model 230 in accordance with information input from the observation data storage unit 200 and the operation planning unit 210.

[0029] In the fermenter 140, the input raw material is converted into organic acids by acid-producing bacteria, and the organic acids are converted into methane gas by methanogens.

[0030] A wide variety of raw materials are fed into the fermentation tank. Because the conversion rate in the fermentation tank varies depending on the type of raw material, high accuracy can be achieved by using a model that accepts multiple types of raw materials. For example, raw materials can be divided into protein-based food waste (abbreviated as food waste), lipid-based food waste, and carbohydrate-based food waste, or they can be further divided into various types depending on their characteristics.

[0031] <Model of a methane fermentation plant> FIG. 3 shows an outline of a methane fermentation model used in the above embodiment, in which a plurality of raw materials (five in this embodiment) are used.

[0032] This methane fermentation model calculates the methane fermentation rate as a function of the feedstock flowing from the mixing tank 130 to the fermenter 140, as follows: Fermenter influent flow rate Ffeed[L / d], Influent biodegradable organic matter concentration Sbvsin1~5 [g / L] for each raw material 1~5 The total biodegradable organic matter concentration in the fermentation tank, Sbvsin [g / L], Inflow organic acid concentration of each raw material 1 to 5 Svfain1~5 [g / L] The total organic acid concentration in the fermentation tank is Svfain [g / L]. As a function of fermenter 140, Fermenter liquid volume V[L], Fermenter temperature Treac [℃], Fermenter biodegradable organic matter concentration Sbvs1~5 [g / L] for each raw material 1~5 The total biodegradable organic matter concentration in the fermenter, Sbvs [g / L], Fermenter organic acid concentration Svfa [g / L], Acid-producing bacteria concentration Xacid [g / L], The methane bacteria concentration is Xmeth [g / L], as a function of the effluent from the fermenter 140 Methane gas flow rate Fmeth [L / d], which represents the amount of methane gas generated. The fermenter effluent flow rate is Fout [L / d].

[0033] A specific example of the equations constituting the methane fermentation model is shown in FIG.

[0034] The symbols in FIG. 4 are as follows: μ1-5: specific growth rate of acid-producing bacteria corresponding to each raw material 1-5 [1 / d] μm: Maximum specific growth rate of acid-producing bacteria [1 / d] Kd: Acid-producing bacteria death rate [1 / d] μc: specific growth rate of methanogens [1 / d] μmc: maximum specific growth rate of methanogens [1 / d] Kdc: Death rate of methanogens [1 / d] Ks: Acid-producing bacteria half-saturation constant [g BVS / L] Ksc: Methanogen half-saturation constant [g VFA / L] k1: BVS yield of acid-producing bacteria [g BVS / (g acidogens / L)] k2: VFA yield of acid-producing bacteria [g VFA / (g acidogens / L)] k3: VFA yield of methanogens [g VFA / (g methanegens / L)] k5: Methane gas yield of methanogens [L / (g methanegens / L)] α: Reaction rate ratio of raw materials [-] b: SRT / HRT ratio [d / d] Here, BVS stands for biodegradable organic matter, VFA stands for organic acid, SRT stands for sludge retention time, and HRT stands for hydraulic retention time.

[0035] An example of a methane fermentation model has been described above, but any methane fermentation model that includes the input of raw materials into the fermenter 140, the organic matter concentration in the fermenter 140, the amount of methane gas generated from the fermenter 140, etc. can be applied to the present invention. For example, the Anaerobic Digestion Model (ADM1) of the International Water Association (IWA) can also be applied to the present invention.

[0036] <Construction of a state space model> Since the methane fermentation model is nonlinear, in constructing the state space model, it is expressed as a discrete-time nonlinear system expressed by the following equations (10) and (11).

number

[0037] However, x k is the state vector, y k is the observation vector, f(x k ) and h(x k ) are nonlinear functions that constitute the state equation (10) and the observation equation (11), respectively. k and v k are the system noise and the observation noise, respectively.

[0038] State vector x k and the observation vector y k Regarding State vector x k is the biodegradable organic matter concentration in the fermenter, Sbvs_ 1~5 , the organic acid concentration in the fermenter Svfa, the acid-producing bacteria concentration in the fermenter Xacid, and the methane bacteria concentration in the fermenter Xmeth are used. Observation vector y k is expressed as the following equations (12) and (13) using the methane gas generation rate Fmeth and the organic acid concentration Svfa in the fermenter. x k = (Sbvs_1_ k , Sbvs_2_ k , Sbvs_3_ k , Sbvs_4_ k , Sbvs_5_ k , Svfa_ k , Xacid_ k , Xmeth_ k ) …(12) y k = (Fmeth_ k , Svfa_ k ) …(13)

[0039] Nonlinear function f(x) in methane fermentation model k) describes the time evolution of the biodegradable organic matter concentration in the fermenter, the organic acid concentration in the fermenter, the acid-producing bacteria concentration in the fermenter, and the methanogen concentration in the fermenter, and is expressed as the following equation (14) using each mass balance equation.

number

[0040] In addition, the nonlinear function h(x k+1 ) is the concentration of methane bacteria in the fermenter at time k+1 calculated by the state equation (10), Xmeth_ k+1 , and the organic acid concentration in the fermenter Svfa_ k+1 Using this, it is expressed as the following equation (15).

number

[0041] The state space model has been explained above. The state vector is the biodegradable organic matter concentration Sbvs_ 1~5 , the organic acid concentration in the fermenter Svfa, the acid-producing bacteria concentration in the fermenter Xacid, and the methanogen concentration in the fermenter Xmeth can be a function of some or all of the methane gas production rate Fmeth and the organic acid concentration in the fermenter Svfa.

[0042] <State estimation using Kalman filter> In the state space model constructed as described above, the present embodiment applies an unscented Kalman filter, which is one of the nonlinear Kalman filter techniques, as a method for sequentially estimating the state variables of the model.

[0043] Specifically, the value of each state variable is estimated using a Kalman filter to correct the error between the estimated value yk of the methane gas generation amount and the organic acid concentration in the fermenter for the state variable xk (biodegradable organic matter concentration in the fermenter, organic acid concentration in the fermenter, acid-producing bacteria concentration in the fermenter, and methane bacteria concentration in the fermenter) at time k and the observed value of the methane gas generation amount and the organic acid concentration in the fermenter.

[0044] An image of the sequential state estimation using the Kalman filter is shown in Figure 5. Through observation, the estimated values ​​of the state variables gradually approach the true values.

[0045] The state estimation using this Kalman filter improves in accuracy as the estimation is repeated over a period of time from the past to a certain point in time. The estimation period is preferably one month or more, and more preferably two months or more.

[0046] As a result, the operating status of the methane fermentation facility can be estimated with high accuracy.

[0047] Although the state variables Sbvs_i, Svfa, and Xmeth in the above state space model are estimated using an unscented Kalman filter, various methods commonly used as a recursive estimation filter (recursive Bayes filter) for nonlinear models may also be used.

[0048] Furthermore, the state variable xk may be some or all of the biodegradable organic matter concentration in the fermenter, the organic acid concentration in the fermenter, the acid-producing bacteria concentration in the fermenter, and the methane bacteria concentration in the fermenter; The error between the estimated value yk and the observed value can be corrected by part or all of the methane gas production rate and the organic acid concentration in the fermenter.

[0049] <Prediction of methane gas generation> As described above, the operating state of the methane fermentation facility at a certain point in time estimated by the Kalman filter is reflected in the state space model, and the raw material input conditions from a certain point in time onwards are set, thereby making it possible to predict the amount of methane gas generated from a certain point in time onwards with high accuracy.

[0050] Specifically, a raw material input plan from a certain point onward is created as an operating condition, and based on this, the next value of the material flowing from the mixing tank 130 into the fermentation tank 140 is referenced. Sbvsin1~5, Svfain1~5, Ffeed, Sbvsin, Svfain

[0051] Then, the methane gas flow rate (i.e., the amount generated) from a certain point in time onwards is predicted using the amounts of substances (Sbvsin1-5, Svfain1-5, Ffeed, Sbvsin, Svfain) flowing into the fermentation tank 140 from a certain point in time onwards and a state space model 230 at a certain point in time in which the operating state of the methane fermentation equipment is estimated using a Kalman filter.

[0052] If a certain point in time is the present time, the future methane gas flow rate from the present time onwards is predicted.

[0053] FIG. 6 shows an example of a predicted amount of methane gas generated according to this embodiment.

[0054] When controlling the amount of methane gas generated, a prediction technique for the methane gas flow rate from a certain time onward is used to control the amount of raw material input and the type of input raw material so as to achieve the target methane gas flow rate. Specifically, the amount of raw material input and the type of input raw material from a certain time onward are determined so as to achieve the target methane gas flow rate. If a certain time is the current time, the amount of raw material input and the type of input raw material can be determined so as to achieve the target methane gas flow rate in the future.

[0055] The period for controlling the amount of methane gas generated can be, for example, about 3 to 14 days, and the amount of raw material input can be determined about once every day to once every 7 days.

[0056] Hereinafter, an example of prediction of methane gas generation amount and control of a methane fermentation plant will be described.

[0057] [Example 1] <Prediction of methane gas generation> In the commercial-scale methane fermentation plant shown in Figure 2, the methane fermentation model shown in Figures 3 and 4 was applied to calculate the biodegradable organic matter concentration Sbvs_ 1~5 A state space model was constructed with the organic acid concentration in the fermenter Svfa, the acid-producing bacteria concentration in the fermenter Xacid, and the methanogen concentration in the fermenter Xmeth as state variables.

[0058] Here, the state vector is the biodegradable organic matter concentration in the fermenter, Sbvs_ 1~5 The organic acid concentration in the fermenter was Svfa, the acid-producing bacteria concentration in the fermenter was Xacid, and the methanogen concentration in the fermenter was Xmeth, and the methane gas production rate Fmeth was used as the observation vector.

[0059] The state variables are sequentially estimated every four hours using a Kalman filter for two months from two months ago to the present time, and the current state of the methane fermentation tank (the biodegradable organic matter concentration in the fermentation tank Sbvs_ 1~5 The organic acid concentration in the fermenter (Svfa), the acid-producing bacteria concentration in the fermenter (Xacid), and the methanogen concentration in the fermenter (Xmeth) were derived.

[0060] As a result, the current state of the fermenter 140 was calculated as shown in Table 1.

[0061] [Table 1]

[0062] The other conditions of the fermenter 140 were as shown in Table 2.

[0063] [Table 2]

[0064] Next, the methane gas flow rate was predicted for one week from the day after the current time. The types and amounts of raw materials input during the prediction period are shown in Table 3. Raw material 1 is food waste, raw material 2 is protein-based food waste, raw material 3 is lipid-based food waste, raw material 4 is carbohydrate-based food waste, and raw material 5 is water-based food waste.

[0065] [Table 3]

[0066] Table 4 shows a comparison of the methane gas flow rate predicted at the current time and the actual methane gas flow rate observed one week after the current time.

[0067] [Table 4]

[0068] The maximum error between the predicted methane gas flow rate and the measured methane gas flow rate was 7%, with an average error of 3%, demonstrating extremely good agreement. However, the error was calculated using the following formula (16). (Error) = |(Predicted methane gas flow rate - Measured methane gas flow rate) / (Measured methane gas flow rate)| …(16)

[0069] [Example 2] <Prediction of methane gas generation> In the commercial-scale methane fermentation plant shown in Figure 2, the methane fermentation model shown in Figures 3 and 4 was applied to calculate the biodegradable organic matter concentration Sbvs_ 1~5 A state space model was constructed with the organic acid concentration in the fermenter Svfa, the acid-producing bacteria concentration in the fermenter Xacid, and the methanogen concentration in the fermenter Xmeth as state variables.

[0070] Here, the state vector is the biodegradable organic matter concentration in the fermenter, Sbvs_ 1~5 The organic acid concentration in the fermenter was Svfa, the acid-producing bacteria concentration in the fermenter was Xacid, and the methanogen concentration in the fermenter was Xmeth, and the methane gas production rate Fmeth was used as the observation vector.

[0071] The state variables are sequentially estimated every four hours using a Kalman filter for two months from two months ago to the present time, and the current state of the methane fermentation tank (the biodegradable organic matter concentration in the fermentation tank Sbvs_ 1~5 The organic acid concentration in the fermenter (Svfa), the acid-producing bacteria concentration in the fermenter (Xacid), and the methane bacteria concentration in the fermenter (Xmeth) were calculated. As a result, the current state of the fermenter 140 was calculated as shown in Table 5.

[0072] [Table 5]

[0073] The other conditions of the fermenter 140 were as shown in Table 6.

[0074] [Table 6]

[0075] Next, the methane gas flow rate was predicted for one week from the day after the current time. The types and amounts of raw materials fed during the prediction period are shown in Table 7. The types of raw materials 1 to 5 were the same as in Example 1.

[0076] [Table 7]

[0077] Table 8 shows a comparison of the methane gas flow rate predicted at the current time and the actual methane gas flow rate observed one week after the current time.

[0078] [Table 8]

[0079] The maximum error between the predicted methane gas flow rate and the measured methane gas flow rate was 7%, and the average error was 4%, demonstrating extremely good agreement.

[0080] [Example 3] <Controlling methane gas generation> Using technology to predict future methane gas flow rates, the amount of raw material input was controlled to achieve the target methane gas flow rate.

[0081] Figure 7 shows the flow rate of methane gas at 8000 Nm 3 This is an example of a plant operated so that the production volume is / d.

[0082] From day 1 to day 149, operation was carried out without using methane gas flow rate prediction technology.

[0083] From the 150th to the 210th day, methane gas flow rate prediction technology was used to estimate the flow rate of 8000 Nm 3 The amount and type of raw material input was determined so that the methane gas generation rate would be 8,000 Nm / d. Specifically, the methane gas generation rate was predicted once every three days for the next seven days, and the rate was set at 8,000 Nm / d for all seven days. 3 The amount and type of raw materials to be added were determined so that / d was achieved.

[0084] As a result, the target value (8000Nm 3 The average error from the target value (8000 Nm / d) was 9%, and the average error from the target value (8000 Nm 3 / d), the average error was 3%. It was confirmed that using methane gas flow rate prediction technology to control the amount of raw material input and the type of raw material to achieve the target methane gas flow rate is extremely effective.

[0085] In this embodiment, a methane fermentation model such as that shown in Figures 3 and 4 is used, and an unscented Kalman filter is used as a sequential estimation filter (sequential Bayes filter) to estimate the state variables of the state space model, but the methane fermentation model and the state estimation method are not limited to these. For example, an anaerobic digestion model (ADM) can be used as the methane fermentation model, and an extended Kalman filter, ensemble Kalman filter, particle filter, etc. can be used as the state estimation method. [Explanation of symbols]

[0086] 100...Methane fermentation plant (actual plant or actual plant) 120...Crusher 130…Mixing tank 140...Fermentation tank 200...Observation data storage unit 210...Operations Planning Department 220...Calculation, prediction, control unit 230...State Space Model

Claims

1. expressing the methane fermentation model as a state space model using at least some of the variables of the methane fermentation model as state variables; Obtaining observations in a methane fermentation facility including a fermenter; and applying a recursive estimation filter to the state space model so as to correct an error between the observed value and an estimated value of an observation vector based on an observation equation in the state space model, thereby recursively estimating the state variables so that the estimated value of the state variables approaches a true value. A method for estimating the operational status of a methane fermentation facility.

2. the state variables are at least some or all of the concentrations of biodegradable organic matter in the fermenter, organic acids in the fermenter, acid-producing bacteria in the fermenter, and methanogens in the fermenter; The observed values ​​are a part or all of the amount of methane gas generated from the fermenter and the organic acid concentration in the fermenter. The method for estimating the operational state of a methane fermentation facility according to claim 1 .

3. the state variables are at least some or all of the concentrations of biodegradable organic matter for each of the plurality of raw material types in the fermenter, organic acids in the fermenter, acid-producing bacteria in the fermenter, and methanogens in the fermenter; The observed values ​​are some or all of the following: the flow rate of the raw material input to the fermenter; the biodegradable organic matter concentration for each of the plurality of raw material types in the input raw material; the organic acid concentration for each of the plurality of raw material types in the input raw material; the amount of methane gas generated from the fermenter; the flow rate of the fermentation liquid discharged from the fermenter; the temperature of the fermenter; and the amount of fermentation liquid in the fermenter. The method for estimating the operational state of a methane fermentation facility according to claim 1 .

4. The method for estimating the operational state of a methane fermentation facility according to any one of claims 1 to 3 is used to obtain estimated values ​​of some or all of the biodegradable organic matter concentration in the fermenter, the organic acid concentration in the fermenter, the acid-producing bacteria concentration in the fermenter, and the methanogen concentration in the fermenter at a certain point in time, Reflecting the estimated value in a state space model; A method for predicting the amount of methane gas generated from a methane fermentation facility, characterized by predicting the amount of methane gas generated from a certain point in time onwards by applying to the state space model some or all of the following: the flow rate of raw material input to the fermentation tank from at least a certain point in time onwards, the biodegradable organic matter concentrations for each of a plurality of raw material types in the input raw material, and the organic acid concentrations for each of a plurality of raw material types in the input raw material.

5. A method for controlling a methane fermentation facility, characterized by controlling the methane fermentation facility so that the future methane gas generation amount becomes a target value, using the estimated results obtained by the method for estimating the operational state of a methane fermentation facility described in any one of claims 1 to 3 or the predicted results obtained by the method for predicting the gas generation amount of a methane fermentation facility described in claim 4.

6. A computer program for causing a computer to execute the method according to any one of claims 1 to 5.

7. An apparatus for predicting the operating state of a methane fermentation facility using a methane fermentation model, a means for expressing the methane fermentation model as a state space model using at least some of the variables of the methane fermentation model as state variables; a means for obtaining observations in a methane fermentation facility including a fermenter; means for sequentially estimating the state variables by applying a recursive estimation filter to the state space model so as to correct an error between the observed value and an estimated value of an observation vector based on an observation equation in the state space model, thereby causing the estimated value of the state variable to approach a true value.

8. 8. The device for estimating the operational state of a methane fermentation facility according to claim 7, further comprising means for performing the estimation by applying a nonlinear Kalman filter.

9. 9. A gas generation amount prediction device for a methane fermentation facility, comprising means for predicting the gas generation amount from a certain point in time onward, using the estimation results obtained by the operational state estimation device for a methane fermentation facility according to claim 7 or 8.

10. A control device for a methane fermentation facility, characterized in that it comprises means for controlling the methane fermentation facility so that the future gas generation amount becomes a target value, using the prediction result by the operation state estimation device for a methane fermentation facility described in claim 7 or 8 or the prediction result by the gas generation amount prediction device for a methane fermentation facility described in claim 9.

11. A computer program for causing a computer to implement the device according to any one of claims 7 to 10.

Images