Control system for biogas power generation equipment and control method for biogas power generation equipment

Abstract

To provide a control system for biogas power generation facilities and a control method for biogas power generation facilities that enable efficient operation of generators attached to the biogas generation facilities.SOLUTION: A control system for biogas power generation facilities has: an acquisition unit 11 that acquires operational information of a biogas generation facility 2; a gas generation amount prediction unit 12 that predicts an amount of biogas or methane gas generated from the biogas generation facility 2 based on the operational information; a gas consumption calculation unit 13 that calculates a gas consumption target value to be consumed by generators 4a, 4b, ... 4x attached to the biogas generation facility 2 from the gas generation amount prediction value predicted by the gas generation amount prediction unit 12; a number determining unit 14 that determines the number of operating generators 4a, 4b,...4x based on the gas consumption target value; and a power generation control unit 15 that controls starting or stopping of the generators 4a, 4b, ... 4x based on a determination result of the number determining unit 14.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a control system for a biogas power generation facility and a control method for a biogas power generation facility, and more particularly, to a control system for a biogas power generation facility and a control method for a biogas power generation facility suitable for treating biogas generated in methane fermentation of sludge.

Background Art

[0002] Biogas generated from a methane fermentation facility is mainly converted into electric power by a power generation facility, sold, and monetized. Efficient operation of a biogas power generation facility and effective utilization of biogas lead to increased profits. However, the amount of biogas generated during methane fermentation depends greatly on the properties of the input raw materials and the active state of microorganisms, making it difficult to predict. In addition, the management method of biogas generation amount and the operation method of the power generation facility also depend largely on the knowledge and experience of engineers. Therefore, research is underway on management and operation methods that do not depend on the proficiency of engineers.

[0003] Japanese Patent Application Laid-Open No. 2020-6291 (Patent Document 1) describes using a function that includes flora data related to two or more types of bacteria and data on the amount of biogas generated as variables as prediction data for predicting the amount of biogas generated.

[0004] Japanese Patent Application Laid-Open No. 2019-141756 (Patent Document 2) describes an example of a waste treatment system that analyzes a photographed image of waste, identifies the type of waste based on the color, density, and unevenness of the waste, and predicts the amount of methane gas or biogas generated using the identified type and the amount of waste input into the methane fermentation tank.

[0005] Japanese Patent Publication No. 2009-33906 (Patent Document 3) describes an example of a control device for a gas power generation facility equipped with multiple generators, which operates the master generator at rated capacity when the amount of stored gas exceeds a standard amount of gas, and operates the slave generators at rated capacity when the amount of stored gas exceeds a set amount of gas. Non-Patent Document 1 describes a technology that uses AI to support the electricity sales business of a biomass power plant. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2020-6291 [Patent Document 2] Japanese Patent Publication No. 2019-141756 [Patent Document 3] Japanese Patent Publication No. 2009-33906 [Non-patent literature]

[0007] [Non-Patent Document 1] Yusuke Yamashina et al., "AI Supports Biomass Power Plant Electricity Sales Operations: From Prediction to Anomaly Detection and Extraction of Influencing Factors," Mitsubishi Heavy Industries Technical Report, Vol. 55, No. 4 (2018), pp. 1-5. [Overview of the project] [Problems that the invention aims to solve]

[0008] However, Patent Document 1 predicts gas generation based on microbial biomass data from the fermentation tank, which necessitates analytical equipment to acquire the biomass data, resulting in time-consuming and costly methods. Furthermore, Patent Document 1 is unsuitable for near real-time prediction due to the time required to acquire the biomass data. Patent Document 2 also requires a waste imaging device, necessitating a separate device for predicting methane gas generation.

[0009] Patent Document 3 describes controlling the operation of multiple generators based on the amount of gas stored in a gas tank, but energy loss occurs when the generators start and stop. Therefore, if the amount of stored gas fluctuates greatly, the frequency of generator starts and stops increases, which may conversely decrease the power generation efficiency. Non-Patent Document 1 describes predicting the amount of gas generated 1 to 3 days in advance using machine learning, but the algorithm is ensemble learning and no time-series analysis is performed. Furthermore, there is no specific consideration of how to operate the generators based on the predicted amount of gas generated.

[0010] In view of the above issues, the present invention provides a control system for a biogas power generation facility and a control method for a biogas power generation facility that can efficiently operate a generator attached to a biogas generation facility. [Means for solving the problem]

[0011] In order to solve the above problems, the inventors of this invention conducted diligent research and found that it is useful to predict the amount of biogas generated from a biogas generation facility, calculate a target gas consumption value for the generator based on the predicted gas generation amount, and further control the number of generators in operation based on the target gas consumption value for the generator.

[0012] Based on the above findings, the present invention is, in one aspect, a control system for a biogas power generation facility comprising: an acquisition unit for acquiring operating information of a biogas generation facility; a gas generation amount prediction unit for predicting the amount of biogas or methane gas generated from the biogas generation facility based on the operating information; a gas consumption amount calculation unit for calculating a target gas consumption amount to be consumed by a generator attached to the biogas generation facility from the predicted gas generation amount value predicted by the gas generation amount prediction unit; a unit number determination unit for determining the number of generators to be operated based on the target gas consumption amount; and a power generation control unit for starting or stopping the generators based on the determination result of the unit number determination unit.

[0013] In one embodiment of the biogas power generation facility control system according to the present invention, the gas generation amount prediction unit predicts a gas generation amount based on a trained model obtained by machine learning using information including at least one of the amount of biomass input to the biogas generation facility, the input biomass concentration, and the processing temperature as training data.

[0014] In another embodiment, the control system for a biogas power generation facility according to an embodiment of the present invention further includes historical measured gas production values ​​from the biogas generating facility as input data for a trained model.

[0015] In yet another embodiment, the control system for a biogas power generation facility according to an embodiment of the present invention includes a gas consumption calculation unit that calculates the average value of the predicted gas generation amount over a predetermined future period predicted by the gas generation amount prediction unit as the target gas consumption value.

[0016] In yet another embodiment, the control system for a biogas power generation facility according to an embodiment of the present invention includes a power generation control unit that controls the starting or stopping of generators based on the result of determining the number of operating generators, so as to minimize the number of times the generators are started or stopped.

[0017] In yet another embodiment, the control system for a biogas power generation facility according to an embodiment of the present invention further comprises a gas tank for storing biogas and a level meter for measuring the level in the gas tank, and further includes a unit number determination unit that determines the number of power generators to be operated based on the level measurement results.

[0018] In yet another embodiment, the control system for a biogas power generation facility according to an embodiment of the present invention further includes a correction unit that corrects the number of operating generators determined by the number determination unit.

[0019] In another aspect, the present invention uses the operation information of a biogas generation facility to predict the future gas generation amount of biogas generated from the biogas generation facility, calculates the target gas consumption amount to be consumed by a generator attached to the biogas generation facility from the predicted gas generation amount of the biogas, determines the number of operating generators based on the target gas consumption amount, and starts or stops the generators based on the determination result of the number of operating generators. This is a control method for biogas power generation equipment.

Effects of the Invention

[0020] According to the present invention, it is possible to provide a control system for biogas power generation equipment and a control method for biogas power generation equipment that can efficiently operate a generator attached to a biogas generation facility.

Brief Description of the Drawings

[0021] [Figure 1] It is a schematic diagram showing an example of biogas power generation equipment according to an embodiment of the present invention. [Figure 2] It is a flowchart showing an example of a control method for biogas power generation equipment according to an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0022] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments shown below illustrate devices and methods for embodying the technical idea of this invention, and the technical idea of this invention does not specify the structure, arrangement, etc. of the components as follows.

[0023] (Biogas Power Generation Equipment) As shown in FIG. 1, the biogas power generation equipment according to an embodiment of the present invention includes a biogas generation facility 2 that generates biogas, a gas tank 3 that stores the biogas generated by the biogas generation facility 2, a plurality of generators 4a, 4b,... 4x that generate electricity using the biogas, and a control device 10 that controls the start and stop of the generators 4a, 4b,... 4x.

[0024] A pretreatment facility 1 may be provided upstream of the biogas generation facility 2 to pretreatment the raw water flowing into the biogas generation facility 2. Furthermore, a posttreatment facility (not shown) may be provided downstream of the biogas generation facility 2 to posttreatment the treated sludge obtained from the biogas generation facility 2.

[0025] Suitable raw water for treatment includes organic wastewater or organic waste containing organic substances. For example, organic sludge such as sewage sludge, human waste, septic tank sludge, and drainage sludge, or organic waste such as kitchen waste and food scraps discharged from various factories are suitable.

[0026] The specific equipment configuration of pretreatment facility 1 is not particularly limited. For example, it is preferable to use a treatment device as pretreatment facility 1 that performs pretreatment on raw water, such as solubilization, coagulation and sedimentation, aerobic treatment, anaerobic treatment, and solid-liquid separation (concentration, dewatering), to obtain sludge or biomass with properties suitable for fermentation treatment in biogas generation facility 2. For example, treatment devices that perform gravity concentration using gravity sedimentation, or mechanical concentration using screens, filter cloths, or centrifugation, etc., are typically available.

[0027] The biogas generation facility 2 is a device that ferments the sludge or biomass flowing into the biogas generation facility 2 to obtain treated sludge and generates biogas. For example, the biogas generation facility 2 can be a digester that anaerobically treats the input sludge with anaerobic microorganisms. The treated sludge processed in the biogas generation facility 2 is typically post-treated using a dewaterer, dryer, etc., before being transported outside. The biogas generated in the biogas generation facility 2 contains hydrogen sulfide in addition to methane gas, carbon dioxide, etc., which are components of biogas, so it is purified by desulfurization treatment, etc., before being stored in the gas tank 3.

[0028] Gas tank 3 contains the biogas generated at the biogas generation facility 2. Gas tank 3 is not limited to a single unit as illustrated in Figure 1, but may be provided one or more times adjacent to each of the multiple generators 4a, 4b, ... 4x. Gas tank 3 is equipped with a level gauge 31 capable of measuring the level (remaining amount) of biogas contained within gas tank 3.

[0029] The measurement results from the level gauge 31 are configured to be output to the control device 10 connected to the level gauge 31. Generators 4a, 4b, ... 4x are each connected to the control device 10, and the starting and stopping of each generator 4a, 4b, ... 4x can be controlled via the control device 10.

[0030] The control device 10 is composed of a general-purpose computer or the like. The control device 10 includes an acquisition unit 11, a gas generation amount prediction unit 12, a gas consumption amount calculation unit 13, a unit number determination unit 14, a power generation control unit 15, and a correction unit 16. In addition, a storage unit 20 for storing various information necessary for processing by the control device 10 and a learning unit 17 for storing various information necessary for predictive analysis using machine learning may be connected to the control device 10.

[0031] The control device 10 may be connected to other biogas power generation facilities 30 via a network 50, and may be configured to mutually exchange control information of generators 4a, 4b, ... 4x with the other biogas power generation facilities 30. The control device 10 may also be connected to an operation management facility 40, etc., which is capable of centrally managing multiple facilities, including the biogas power generation facility 30 according to the embodiment of the present invention, and may be configured to mutually exchange control information of generators 4a, 4b, ... 4x, trained models, past performance data, etc.

[0032] The acquisition unit 11 acquires operational information of the biogas generation facility 2 that generates biogas. The operational information includes, for example, the amount of biomass input to the biogas generation facility 2, the biomass concentration (TS, VS), COD, pH, treatment temperature, volumetric load, hydraulic residence time, the effective capacity of the treatment tank of the biogas generation facility 2, the stirring speed, the biogas or methane gas generation rate (sludge decomposition rate), the TS, VS, pH of the treated sludge processed in the biogas generation facility 2, the amount of sludge extracted, etc. The acquisition unit 11 may further acquire pretreatment information of the pretreatment facility 1. The pretreatment information includes the treatment conditions of the pretreatment facility 1, the sludge concentration (TS, VS) of the raw water, the coagulant injection rate, the concentration of biomass obtained by pretreatment (TS, VS), the suspended solids concentration (SS), the amount of biomass extracted and transferred, etc.

[0033] The gas generation prediction unit 12 predicts the amount of biogas or methane gas contained in biogas based on operational information. For example, the gas generation prediction unit 12 performs statistical analysis and simulations based on operational information, which includes a database storing at least one of the following: the amount of input biomass, the input biomass concentration, the processing temperature, and the amount of biogas or methane gas generated or the conversion rate of input biomass to methane gas depending on the type of input biomass, to calculate a predicted value for the amount of biogas or methane gas generated in the present or future for a predetermined period. The predicted value for biogas gas generation may be obtained by referring to past databases and calculating the average value of actual gas generation values ​​for a predetermined period in the past, or by performing simulations using analysis software for analyzing operational information.

[0034] The accuracy of predicting biogas or methane gas generation can be improved by using machine learning. For example, the gas generation prediction unit 12 predicts the amount of biogas or methane gas generated based on a trained model obtained by machine learning using operational information, which includes at least one of the following as training data: the amount of biomass input to the biogas generation facility 2, the biomass concentration, and the processing temperature. By using machine learning, the operational information of the biogas generation facility 2 can be analyzed over time, allowing for a more accurate prediction of biogas or methane gas generation over a predetermined period in the future. Note that the trained model may be created on a computer separate from the control device 10.

[0035] The explanatory variables of the trained model used in the predictive analysis by the gas generation amount prediction unit 12 include operating information of the biogas generation facility 2 (e.g., fermentation treatment conditions, input raw materials, information on treated sludge, etc.) and pretreatment information of the pretreatment facility 1. For example, if a sewage sludge digester is provided as the biogas generation facility 2, explanatory variables can be appropriately selected from among, for example, the treatment temperature of the digester, the amount of sludge input to the digester, the input sludge concentration, etc.

[0036] As for machine learning algorithms, various known analytical tools such as random forests and neural networks (ANN, RNN) can be appropriately selected and used. In particular, for predicting the amount of biogas generated according to this embodiment, LSTM or GRU are preferred among RNNs, and LSTM is even more preferred, but the invention is not limited to these.

[0037] In the case of RNN, the time step, which is a hyperparameter, is preferably set to 90 days or less, and more preferably to within the fermentation period (hydrological residence time) of the biogas generation facility 2, from the viewpoint of improving accuracy. If there are multiple biogas generation facilities 2, the gas generation amount prediction unit 12 may predict the gas generation amount for each of the biogas generation facilities 2, or it may predict the total amount generated.

[0038] When using machine learning to predict biogas or methane gas production, it is preferable to further include historical measured gas production data from biogas production facility 2 as input data for the trained model. By including historical measured gas production data as input data other than the explanatory variables mentioned above, predictions can be made that take historical measured gas production data into account, thereby improving the accuracy of the prediction.

[0039] If the period for predicting biogas generation is set too far into the future, the difference between the current gas generation amount and the predicted future gas generation amount will become large. As a result, the biogas stored in gas tank 3 may reach a predetermined level more quickly. It is preferable to determine the period for predicting biogas generation amount by considering the variability of the predicted gas generation amount and the effective capacity of gas tank 3.

[0040] For example, regarding the period for predicting biogas generation, it is preferable to determine, for example, how many days into the future to predict, by conducting a preliminary desktop verification based on actual gas generation measurements from the past month, so that the number of operating generators 4a, 4b, ... 4x and the number of starts and stops are optimized. In one embodiment, although not limited to the following, the biogas generation prediction period is preferably 10 days or less, more preferably 1 to 7 days, and even more preferably 1 to 3 days. For example, if the biogas generation prediction period is 3 days, the gas generation prediction unit 12 predicts the biogas generation amount after 1 day, 2 days, and 3 days.

[0041] The gas consumption calculation unit 13 calculates the target gas consumption value to be consumed by the generators 4a, 4b, ... 4x attached to the biogas generation facility 2 from the predicted gas generation amount predicted by the gas generation amount prediction unit 12. For example, the gas consumption calculation unit 13 calculates the target gas consumption value based on the predicted gas generation amount predicted by the gas generation amount prediction unit 12, a predetermined rated power generation efficiency, and the actual power generation efficiency, so that power is generated appropriately within the range in which the power generation efficiency of the generators 4a, 4b, ... 4x achieves a predetermined target efficiency.

[0042] Furthermore, it is preferable that the gas consumption calculation unit 13 calculates a target gas consumption value such that the number of starts and stops of generators 4a, 4b, ... 4x is minimized. For example, if the gas consumption calculation unit 13 predicts that there will be a rapid fluctuation in the predicted gas generation amount in a very short period of time in the future, even if the gas generation amount tends to stabilize within a certain range thereafter, it smooths the predicted gas generation amount for the very short period of time in which the rapid fluctuation in gas generation amount occurs, and calculates a target gas consumption value. This prevents the generators 4a, 4b, ... 4x from being started and stopped due to rapid fluctuations in gas generation, thereby reducing the number of starts and stops of generators 4a, 4b, ... 4x.

[0043] Since a large amount of energy is consumed when starting or stopping generators 4a, 4b, ..., 4x, increasing the number of starts and stops will increase the amount of energy consumed. According to this embodiment, the target value of gas consumption is calculated so that the number of starts and stops of generators 4a, 4b, ..., 4x is minimized, so that power generation can be performed with energy loss of generators 4a, 4b, ..., 4x reduced as much as possible. "Number of starts and stops" basically refers to the number of times generators 4a, 4b, ..., 4x are started or stopped. For example, when generator 4a goes from a stopped state to an started state, it is counted as one time, and when it goes from an started state to a stopped state, it is counted as one time. For a simpler definition of the number of starts and stops, it is also acceptable to count the number of changes in the number of operating generators as one time. In this case, for example, when the number of operating generators is changed from 8 to 10, it is counted as one time, and when the number is changed from 10 to 13, it is counted as one time.

[0044] The biogas generated from the biogas generation facility 2 contains methane gas, carbon dioxide, hydrogen sulfide, etc., and its composition varies depending on the properties of the sludge fed into the biogas generation facility 2 and the treatment conditions. Therefore, it is preferable for the gas consumption calculation unit 13 to calculate the average value of the predicted gas generation amounts of biogas or methane gas over a predetermined future period predicted by the gas generation amount prediction unit 12, and to determine this as the gas consumption target value. In this way, by calculating the gas consumption target value considering the time series of biogas generation over a certain period from the present to the future, the impact of very short-term fluctuations in biogas generation on the start or stop operations of generators 4a, 4b, ... 4x can be reduced. As a result, the number of starts and stops of generators 4a, 4b, ... 4x can be reduced, and more efficient power generation can be achieved.

[0045] Furthermore, if the predicted gas generation amount remains stable over a long period and there is no need to consider fluctuations in gas generation amount, the gas consumption calculation unit 13 may determine the predicted gas generation amount for any specific time in the future as the target gas consumption amount.

[0046] The unit number determination unit 14 determines the number of operating generators 4a, 4b, ..., 4x based on the gas consumption target value calculated by the gas consumption calculation unit 13. For example, the unit number determination unit 14 uses the following formula: (Number of operating units) = (Target gas consumption value) ÷ (Gas consumption per generator) ... (1) Based on this, the number of operating generators 4a, 4b, ..., 4x can be determined.

[0047] Furthermore, it is even more preferable for the unit number determination unit 14 to determine the number of operating generators 4a, 4b, ..., 4x based on the measurement results of the gas tank 3 level by the level gauge 31. By determining the number of operating generators based on the gas tank 3 level, processing can be carried out according to the level of biogas in the gas tank 3, thus enabling efficient generator operation.

[0048] For example, if the measurement result of the gas tank 3 level exceeds the upper limit, the number of operating generators 4a, 4b, ... 4x can be determined to be greater than the value obtained by dividing the target gas consumption by the gas consumption per generator. On the other hand, if the measurement result of the gas tank 3 level falls below the lower limit, the number of operating generators can be set to be less than the value obtained by dividing the target gas consumption by the gas consumption per generator.

[0049] For example, in equation (1) above, if the level of gas tank 3 exceeds the upper limit, the unit 14 determines the number of operating units so that the decimal part of the value obtained in equation (1) is rounded up to an integer value. If the level of gas tank 3 exceeds the lower limit, the unit determines the number of operating units so that the decimal part of the value obtained in equation (1) is truncated to an integer value.

[0050] The unit number determination unit 14 may also correct the calculation formula in equation (1) by providing predetermined parameters. For example, when the unit number determination unit 14 wants to raise the level of the gas tank 3, it may decide to subtract a correction parameter a (integer) from the calculation result of equation (1). In the above example using equation (1), the case where the gas consumption of generators 4a, 4b, ... 4x is the same is explained, but the unit number determination unit 14 is not limited to this example. For example, it is obvious that the unit number determination unit 14 can appropriately optimize the number of operating units according to the relative gas consumption of generators 4a, 4b, ... 4x.

[0051] The timing of measuring the level in gas tank 3 is not particularly limited. For example, after the unit number determination unit 14 determines the number of operating units based on the gas consumption target value, the level in gas tank 3 is measured before the power generation control unit 15 reflects the determined number of operating units in the operation control of the generators. If the level in gas tank 3 exceeds the upper or lower limit, the unit number determination unit 14 may recalculate the number of operating units based on equation (1). Alternatively, if the unit number determination unit 14 predicts that the level in gas tank 3 will exceed the upper or lower limit based on the prediction result of the gas generation amount prediction unit 12, it may determine the number of operating units using equation (1) before the level reaches the upper or lower limit.

[0052] The control device 10 may further include a correction unit 16. The correction unit 16 corrects the number of operating generators 4a, 4b, ... 4x determined by the unit number determination unit 14. For example, the correction unit 16 may correct the number of operating generators if, as a result of measuring the level of the gas tank 3 by the level gauge 31, the trend of the level rising or falling does not change even after a certain period of time has elapsed. For example, if the level of the gas tank 3 has reached its upper limit and it is desired to lower the level of the gas tank 3, but the level continues to rise even after the number of operating generators has been set, the correction unit 16 can correct the number of operating generators to be one more than the result determined by the unit number determination unit 14, thereby enabling the gas tank 3 to switch to a downward trend earlier.

[0053] Even if the level in gas tank 3 exceeds the upper or lower limit, depending on the gas generation prediction, the correction unit 16 may not need to correct the number of operating generators 4a, 4b, ... 4x. Specifically, this is the case, for example, when the predicted gas generation value temporarily decreases, but is expected to increase soon after. In such cases, if the number of operating generators is determined based on the temporary predicted gas generation value, the number of starts and stops of generators 4a, 4b, ... 4x may increase, resulting in energy loss.

[0054] The correction unit 16, for example, when the gas tank 3 level reaches its upper limit, will not perform a correction to the number of operating units if the target gas consumption value at that time is smaller than the actual gas consumption ((gas consumption per generator) × (number of operating units)). This allows for more efficient power generation while minimizing energy loss by reducing the number of starts and stops of generators 4a, 4b, ... 4x.

[0055] The power generation control unit 15 controls the activation or deactivation of generators 4a, 4b, ..., 4x based on the determination result of the number of generators determination unit 14, so as to minimize the number of times each generator is started or stopped. For example, the power generation control unit 15 can control the activation or deactivation of generators 4a, 4b, ..., 4x only when it becomes necessary to newly activate or deactivate generators 4a, 4b, ..., 4x based on the determination result of the number of generators determination unit 14 and the operating status of generators 4a, 4b, ..., 4x.

[0056] According to the control system for a biogas power generation facility in an embodiment of the present invention, the gas consumption calculation unit 13 calculates a target gas consumption value based on the predicted gas generation value of the gas generation prediction unit 12, and based on this target gas consumption value, the number of generators 4a, 4b, ... 4x to be operated is determined so as to minimize the number of times each generator is started and stopped. This provides a control system for a biogas power generation facility that can efficiently operate the generators attached to the biogas generation facility.

[0057] (Control method and power generation method for biogas power generation equipment) As shown in Figure 2, the control method and power generation method for a biogas power generation facility according to an embodiment of the present invention includes: step S1 of acquiring operating information of the biogas generating facility 2; step S2 of predicting the future amount of biogas generated from the biogas generating facility 2 using the operating information; step S3 of calculating a target gas consumption value to be consumed by the generator attached to the biogas generating facility from the predicted biogas gas generation value; step S4 of determining the number of operating generators 4a, 4b, ... 4x based on the target gas consumption value; step S5 of determining whether or not to correct the determined number of operating generators 4a, 4b, ... 4x; step S6 of correcting the number of operating generators 4a, 4b, ... 4x; and step S7 of starting or stopping the generators 4a, 4b, ... 4x based on the result of determining the number of operating generators.

[0058] In step S1 of Figure 2, the acquisition unit 11 in Figure 1 acquires various information necessary for predicting the amount of biogas generated from the biogas generation facility 2. For example, the acquisition unit 11 acquires operating information of the biogas generation facility 2 and, if necessary, pre-treatment information of the pre-treatment facility 1. The acquisition unit 11 can improve the accuracy of the prediction by acquiring past measured gas generation amounts from the biogas generation facility 2 via the network 50 from the storage unit 20 or the operation management facility 40 and using this as input data.

[0059] In step S2 of Figure 2, the gas generation prediction unit 12 of Figure 1 predicts the current and future biogas generation amounts from the biogas generation facility 2 over a predetermined period, based on the operating information acquired by the acquisition unit 11. The biogas generation amount prediction may be calculated based on actual past biogas generation amounts stored in the storage unit 20, or, as described above, it may be predicted using a trained model obtained by performing machine learning.

[0060] In step S3 of Figure 2, the gas consumption calculation unit 13 in Figure 1 calculates a target gas consumption value based on the predicted biogas gas generation amount for a predetermined period predicted in step S2, so that the power generation efficiency of generators 4a, 4b, ... 4x is within an appropriate range. In step S4 of Figure 2, the unit number determination unit 14 in Figure 1 determines the number of generators 4a, 4b, ... 4x to operate based on the target gas consumption value, so that the power generation efficiency of generators 4a, 4b, ... 4x is operated within an appropriate range.

[0061] In step S5 of Figure 2, the correction unit 16 in Figure 1 determines whether it is necessary to correct the number of operating generators 4a, 4b, ..., 4x determined by the number determination unit 14. For example, if the correction unit 16 determines, based on the measurement results of the level gauge 31 provided in the gas tank 3 in Figure 1, that the level of the gas tank 3 will exceed the upper limit in the future and that it is necessary to correct the number of operating generators, then in step S6, it corrects the number of operating generators to increase. After the correction, the process proceeds to step S7. If, in step S5, the correction unit 16 determines that it is not necessary to correct the number of operating generators 4a, 4b, ..., 4x determined by the number determination unit 14, the process proceeds to step S7. In step S7, the power generation control unit 15 controls the activation and deactivation of generators 4a, 4b, ..., 4x based on the determination result of the number of operating generators 4a, 4b, ..., 4x.

[0062] Thus, according to the control system for a biogas power generation facility in the embodiment of the present invention, the amount of gas generated by the biogas generating facility can be predicted, and based on that, the number of operating generators 4a, 4b, ... 4x can be determined and corrected. Therefore, even without the experience of skilled technicians, it is possible to continuously operate generators 4a, 4b, ... 4x in a stable and efficient manner. [Examples]

[0063] Examples of the present invention are shown below along with comparative examples. These examples are provided to help you better understand the present invention and its advantages, and are not intended to limit the invention.

[0064] The controlled system consisted of a biogas power generation facility comprising a digester for anaerobic digestion of sewage sludge, a gas tank for storing the digester gas generated from the digester, and multiple generators that use the digester gas stored in the gas tank to generate electricity. A dataset including the amount of digester gas generated, the amount of sludge input to the digester, the concentration of sludge in the digester, and the concentration of concentrated sludge treated by gravity concentration or mechanical concentration during pretreatment before input to the digester was used as training data. An LSTM machine learning algorithm was used to construct a model for predicting the amount of digester gas generated. Three digesters were used, and a model was constructed for each. The sum of the predicted values ​​for each tank was used as the predicted gas generation value for the gas generation facility.

[0065] The number of generators to be operated was to be set when the gas tank level reached the upper or lower limit line. When the gas tank level reached the upper limit line, the number of operating generators was set to an integer value obtained by rounding up the decimal point of the value calculated using formula (1). When the gas tank level reached the lower limit line, the number of operating generators was set to an integer value obtained by rounding down the decimal point of the value calculated using formula (1).

[0066] By performing machine learning using LSTM, the average of the predicted gas generation amounts for future periods of 3, 5, and 7 days was set as the target gas consumption value. Based on this target gas consumption value, the number of generators to operate was determined, and the starting and stopping of the generators was controlled. The number of starts and stops was counted as one time when the number of operating generators changed. As a result, the number of starts and stops was lowest when the average was set to 3 days. Therefore, in this embodiment, for the prediction of gas generation, predicted gas generation amounts for 1 day, 2 days, and 3 days later were obtained, and the average of these predicted values ​​was set as the target gas consumption value.

[0067] Under the above conditions, a 20-day verification was conducted, resulting in 6 generator starts and stops in this embodiment. In the same biogas power generation facility as in this embodiment, when set by the operator, the number of starts and stops over 20 days was 14. In other words, it was found that this embodiment can reduce the number of starts and stops. [Explanation of Symbols]

[0068] 1…Pre-treatment facility 2… Biogas generation facilities 3…Gas tank 4a, 4b, ...4x... Generators 10...Control device 11…Acquisition part 12...Gas generation amount prediction unit 13...Gas consumption calculation unit 14…Unit number determination section 15…Power generation control unit 16...Correction section 17…Learning Department 20...Storage section 30… Biogas power generation facilities 31... Level gauge 40…Operation Management Facilities 50…Network

Claims

1. An acquisition unit that acquires operating information of a biogas generation facility, A gas generation amount prediction unit predicts the amount of biogas or methane gas generated from the biogas generation facility based on the aforementioned operating information, A gas consumption calculation unit calculates a target gas consumption value to be consumed by the generator attached to the biogas generation facility from the predicted gas generation value predicted by the gas generation prediction unit, A unit for determining the number of generators to be operated based on the aforementioned gas consumption target value, Based on the determination result of the unit number determination unit, a power generation control unit controls the starting or stopping of the generator. Equipped with, The gas generation amount prediction unit, A control system for a biogas power generation facility, which includes predicting the predicted gas generation amount based on a trained model obtained by machine learning using as training data information the amount of biogas or methane gas generated from the biogas generation facility, the amount of biomass input to the biogas generation facility, and information including at least one of (A) to (C) below, and information including at least one of (1) to (9) below. (A) Biomass input concentration (B) COD of input biomass (C) pH of input biomass (1) Processing temperature (2) Volume load (3) Hydraulic residence time (4) Effective capacity of the treatment tank (5) Stirring speed of the treatment tank (6) Gas generation rate of biogas or methane gas or sludge decomposition rate (7) Concentration of treated sludge (8) pH of treated sludge (9) Amount of sludge removed from treated sludge

2. The acquisition unit acquires at least one of the following driving information (10) to (14): The control system for a biogas power generation facility according to claim 1, wherein the gas generation amount prediction unit predicts the gas generation amount prediction value based on a trained model obtained by machine learning using information including at least one of the following (10) to (14) as training data as the operation information. (10) Sludge concentration of raw water supplied to the pretreatment facility of the biogas generation facility (11) Coagulant injection rate of the pretreatment facility (12) Concentration of biomass obtained in the pretreatment facility (13) Suspended solids concentration of biomass obtained at the pretreatment facility (14) Amount of biomass extracted and transported

3. The control system for a biogas power generation facility according to claim 1 or 2, further comprising past measured gas production amounts of the biogas generating facility as input data for the trained model.

4. A control system for a biogas power generation facility according to any one of claims 1 to 3, wherein the gas consumption calculation unit calculates the average value of the predicted gas generation amount over a predetermined future period predicted by the gas generation amount prediction unit as the gas consumption target value.

5. A control system for a biogas power generation facility according to any one of claims 1 to 4, comprising the power generation control unit controlling the starting or stopping of the generators based on the result of determining the number of operating generators, so as to minimize the number of times the generators are started or stopped.

6. A gas tank for storing the biogas, A level meter is used to measure the level of the gas tank. Furthermore, A control system for a biogas power generation facility according to any one of claims 1 to 5, further comprising the unit number determination unit determining the number of operating generators based on the measurement results of the level.

7. A control system for a biogas power generation facility according to any one of claims 1 to 6, further comprising a correction unit for correcting the number of operating generators determined by the number determination unit.

8. Using the operating information of the biogas generation facility, the future amount of biogas generated from the biogas generation facility is predicted. From the predicted amount of gas generated by the biogas, a target value for the amount of gas to be consumed by the generator attached to the biogas generation facility is calculated. Based on the aforementioned gas consumption target value, the number of operating generators is determined. Based on the determination of the number of operating units, the generator is started or stopped. It has, A method for controlling a biogas power generation facility, comprising predicting the amount of gas generated by a gas generation prediction unit, based on a trained model obtained by machine learning using as training data information the amount of biogas or methane gas generated from the biogas generation facility, the amount of biomass input to the biogas generation facility, information including at least one of (A) to (C) below, and information including at least one of (1) to (9) below, as operating information. (A) Biomass input concentration (B) COD of input biomass (C) pH of input biomass (1) Processing temperature (2) Volume load (3) Hydraulic residence time (4) Effective capacity of the treatment tank (5) Stirring speed of the treatment tank (6) Gas generation rate of biogas or methane gas or sludge decomposition rate (7) Concentration of treated sludge (8) pH of treated sludge (9) Amount of sludge removed from treated sludge

9. The method for controlling a biogas power generation facility according to claim 8, wherein the gas generation amount prediction unit predicts the gas generation amount prediction value based on a trained model obtained by machine learning using information including at least one of the following (10) to (14) as training data as the operation information. (10) Sludge concentration of raw water supplied to the pretreatment facility of the biogas generation facility (11) Coagulant injection rate of the pretreatment facility (12) Concentration of biomass obtained in the pretreatment facility (13) Suspended solids concentration of biomass obtained at the pretreatment facility (14) Amount of biomass extracted and transported

Images