Abnormality diagnosing device and gasification system

Abstract

The gasification system includes an evaluation unit that evaluates whether the gasification system is in an abnormality tendency based on a correlation between a specific process value in the gasification system and a specific component concentration in a gasification gas generated by the gasification system, and a notification unit that notifies an evaluation result of the evaluation unit to the outside.

Description

Technical Field

[0001] The present invention relates to an abnormality diagnosis device and a gasification system. This application claims the benefit of priority based on Japanese Patent Application No. 2020-136252, filed on August 12, 2020, the contents of which are incorporated herein by reference. Background Art

[0002] Patent Document 1 below discloses an abnormality diagnosis method and an abnormality diagnosis system. The abnormality diagnosis method in Patent Document 1 includes: a model creation step for creating a simulation model of a monitored object; an operation start step for starting operation of the monitored object; a measurement step for measuring internal state quantities of the monitored object under operating conditions and extracting measured values; a prediction step for inputting control input values ​​identical to the operating conditions of the monitored object into the simulation model to calculate predicted values ​​of the internal state quantities of the monitored object; a Mahalanobis distance calculation step for calculating the Mahalanobis distance based on the difference between the measured values ​​and the predicted values; and an abnormality diagnosis step for diagnosing whether the operating conditions of the monitored object are abnormal based on the Mahalanobis distance.

[0003] Patent Document 2 below discloses a simulation device and a simulation method. The simulation device described in Patent Document 2 is a device for predicting the operating state of an actual process and comprises: a parameter storage unit that associates and stores adjustment parameters of a simulation model set in synchronization with the operating state of a simulator device with information on operating conditions obtained from an actual plant; a process simulator unit that uses the simulation model to predict the operating state based on the operating conditions of the prediction target; and an adjustment parameter setting unit that dynamically sets the adjustment parameters of the simulation model. The adjustment parameter setting unit obtains adjustment parameters associated with information on operating conditions similar to the prediction target from the parameter storage unit and sets the adjustment parameters of the simulation model based on the adjustment parameters.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Laid-Open No. 2016-151909

[0007] Patent Document 2: Japanese Patent Application Publication No. 2019-021032 Summary of the Invention

[0008] Problems to be solved by the invention

[0009] One of the aforementioned monitored objects is a gasification system. As is well known, such a gasification system gasifies solid fuels (feedstocks) such as biomass through thermal decomposition and reduction reactions, thereby producing gaseous fuels (gasified gas) such as methane as a product. The operational stability of such a gasification system is known to be affected by factors such as the moisture content of the feedstock and the gasification temperature. In order to stably generate gasified gas in a gasification system, it is extremely important to optimize the operating parameters corresponding to the feedstock.

[0010] However, conventional gasification systems use the well-known PID control (proportional / integral / differential control) to control various operating variables in the gasification system. When operating parameters deviate from pre-set limits (upper and lower limits), the system's operating state is determined to be abnormal, and an alarm is output. Consequently, conventional gasification systems suffer from the inability to predict operational anomalies in the gasification system.

[0011] The present invention has been made in view of the above circumstances, and an object of the present invention is to predict operational abnormalities in a gasification system in advance.

[0012] Means for solving problems

[0013] An abnormality diagnosis device according to one embodiment of the present invention includes: an evaluation unit that evaluates whether a gasification system is prone to abnormality based on a correlation between a specific process value in the gasification system and a specific component concentration in a gasification gas generated by the gasification system; and a notification unit that notifies an external user of the evaluation result of the evaluation unit.

[0014] The evaluation unit may also include: a temperature prediction unit that calculates a predicted value of the process value; and a concentration prediction unit that calculates a predicted value of the component concentration. The evaluation unit may evaluate whether the gasification system is in an abnormal tendency by comparing a distance evaluation index with the distance evaluation index obtained when the gasification system is normal. The distance evaluation index is given based on the difference between the predicted value and the measured value of the process value and the difference between the predicted value and the measured value of the component concentration.

[0015] The temperature prediction unit may also use a physical model of the gasification system to obtain a predicted value of the process value.

[0016] The concentration prediction unit may obtain the predicted value of the component concentration using a learned model of the gasification system.

[0017] The evaluation unit may evaluate whether the gasification system is in an abnormality tendency when the predicted value of the component concentration differs from the actually measured value of the component concentration by a predetermined amount or more.

[0018] The process value may be the gasification temperature, and the component concentration may be the methane concentration.

[0019] Another aspect of the present invention is a gasification system including the above-mentioned abnormality diagnosis device.

[0020] Effects of the Invention

[0021] According to the present invention, operational abnormality of a gasification system can be predicted in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram showing the outline of a twin-tower gasifier T (gasification system) in one embodiment of the present invention.

[0023] Figure 2 This is a block diagram showing the functional configuration of an abnormality diagnosis device D according to one embodiment of the present invention.

[0024] Figure 3 This is a flowchart showing the operation of the abnormality diagnosis device D according to one embodiment of the present invention.

[0025] Figure 4 : is a characteristic diagram showing the temporal change of the Mahalanobis distance in one embodiment of the present invention. DETAILED DESCRIPTION

[0026] Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

[0027] First, refer to Figure 1 The twin-tower gasifier T of this embodiment will be described. This twin-tower gasifier T includes a gasifier 1, a combustion furnace 2, a first cyclone 3, and a second cyclone 4 as its main components. It produces gaseous fuel (gasification gas) from a predetermined feedstock N supplied from the outside. This twin-tower gasifier T corresponds to the gasification system of the present invention.

[0028] The raw material N in this embodiment is a solid fuel such as coal or biomass. The coal mentioned above can be general coal or low-quality coal such as lignite or subbituminous coal. Biomass can be, for example, woody biomass such as thinnings or resource crop biomass such as sugarcane. The raw material N is fed into the twin-tower gasifier T with its particle size adjusted to a predetermined size.

[0029] Among the components of the twin-tower gasifier T, gasifier 1 is a thermal decomposition furnace that generates gasified gas by thermally decomposing and reducing the raw material N. This gasifier 1 forms a gasification chamber Rg encompassing an internal space of predetermined volume. As shown in the figure, gasifier 1 includes a raw material inlet 1a, a heat medium receiving port 1b, a bubble fluidized bed 1c, a first discharge port 1d, and a second discharge port 1e.

[0030] The raw material inlet 1a is an opening for feeding the raw material N into the gasification chamber Rg. As shown in the figure, it is provided on the side of the gasifier 1. Specifically, the raw material inlet 1a is positioned so that the raw material N can be fed onto the bubbling fluidized bed 1c formed at the bottom of the gasification chamber Rg, allowing the raw material N fed from the outside to fall onto the bubbling fluidized bed 1c.

[0031] The heat medium receiving port 1b is an opening for receiving the high-temperature granular heat medium P (heat medium) flowing from the first cyclone 3 into the gasification chamber Rg. As shown in the figure, it is located at one horizontal end of the upper portion of the gasifier 1. Specifically, the heat medium receiving port 1b allows the high-temperature granular heat medium P flowing from the first cyclone 3 to flow into the gasification chamber Rg from above and from one end. The granular heat medium P is a granular (solid) heat medium having a predetermined particle size, such as sand.

[0032] As shown in the figure, a bubble fluidized layer 1c is formed across the entire bottom of the vaporization chamber Rg. The raw material N and granular heat medium P are fluidized by steam J injected upward from below. For example, a bellows is provided at the bottom of the vaporization chamber Rg to restrict the passage of the raw material N and granular heat medium P, while allowing the steam J to pass through. The bubble fluidized layer 1c uses the steam J to fluidize the raw material N and granular heat medium P, etc., located above the bellows.

[0033] In the gasification chamber Rg, which is provided with such a bubble fluidized bed 1c, the raw material N is thermally decomposed using a high-temperature granular heat medium P as a heat source, thereby generating a pyrolysis gas. This pyrolysis gas is then reduced using steam J to generate gaseous components such as a gasified gas G. Furthermore, in the gasification chamber Rg, solid components such as char (pyrolysis char) and tar are generated by the thermal decomposition of the raw material N. The pyrolysis char is a substance primarily composed of carbon (C) and can function as a fuel.

[0034] As shown in the figure, a first discharge port 1d is provided on the side of the gasifier 1 and is an opening for primarily discharging the solid components and granular heat medium P into the combustion furnace 2. Specifically, the first discharge port 1d is provided on the side of the gasifier 1, thereby selectively discharging the solid components and granular heat medium P, which have a greater specific gravity than the gas components, into the combustion furnace 2.

[0035] As shown in the figure, in contrast to the first outlet 1d, the second outlet 1e is an opening provided in the upper portion of the gasifier 1, primarily discharging gas components such as the gasification gas G toward the second cyclone 4. Specifically, the second outlet 1e is provided in the upper portion of the gasifier 1, thereby selectively discharging gas components such as the gasification gas G, which have a lower specific gravity than the solid components, toward the second cyclone 4. Furthermore, these gas components contain a small amount of combustion ash (solids) having a lower specific gravity.

[0036] As shown in the figure, the raw material inlet 1a and the heat medium receiving port 1b are relatively close together. Furthermore, as shown in the figure, the first discharge port 1d and the second discharge port 1e are relatively close together, but sufficiently separated from the raw material inlet 1a and the heat medium receiving port 1b. This separation of the raw material inlet 1a and the heat medium receiving port 1b from the first discharge port 1d and the second discharge port 1e is intended to ensure sufficient reaction distance within the vaporization chamber Rg.

[0037] In such a gasifier 1, the raw material N and granular heat medium P introduced into the gasification chamber Rg are transported from one end to the other while being fluidized by the steam J. Furthermore, during this transport, the raw material N undergoes a thermal decomposition reaction and a reduction reaction of the thermally decomposed gas proceeds in the gasifier 1, generating gasification gas G.

[0038] The combustion furnace 2 heats the granular heat medium P by burning the pyrolyzed char flowing in from the gasifier 1. Its internal space forms a combustion chamber Rb. The combustion of the pyrolyzed char produces high-temperature combustion gas and ash in the combustion furnace 2. This combustion gas serves as a heat source to heat the granular heat medium P flowing in from the gasifier 1. As shown in the figure, the combustion furnace 2 has a receiving port 2a, a circulating fluidized bed 2b, and an outlet 2c.

[0039] Receiving port 2a is an opening for receiving the solid components from gasifier 1. As shown in the figure, it is provided on the side of combustion furnace 2. Specifically, receiving port 2a is positioned so that granular heat medium P can be dropped onto circulating fluidized bed 2b formed at the bottom of combustion chamber Rb. This allows the solid components received from gasifier 1, namely granular heat medium P and pyrolyzed char, to fall onto circulating fluidized bed 2b.

[0040] As shown in the figure, a circulating fluidized layer 2b is formed across the entire bottom of the combustion chamber Rb. Air A injected upward from below fluidizes the granular heat medium P and pyrolytic char. For example, the bottom of the combustion chamber Rb is provided with multiple air diffusion pipes that blow air A upward. This circulating fluidized layer 2b fluidizes the granular heat medium P and pyrolytic char located above the air diffusion pipes and allows the air A to act as an oxidant on the pyrolytic char, thereby causing the pyrolytic char to combust.

[0041] In the combustion chamber Rb equipped with such a circulating fluidized layer 2b, pyrolyzed char combusts in the presence of air A, generating high-temperature combustion gas. Furthermore, in the combustion chamber Rb, the high-temperature combustion gas exchanges heat with the granular heat medium P, thereby heating the granular heat medium P. In the combustion chamber Rb, the combustion gas and combustion ash flow upward from the bottom in an ascending flow, and this ascending flow blows the granular heat medium P upward from the bottom.

[0042] The discharge port 2c is an opening for discharging the combustion gas, combustion ash, and granular heat medium P toward the first cyclone 3. As shown in the figure, it is provided in the upper portion of the combustion furnace 2. Specifically, the discharge port 2c is provided in the upper portion of the combustion furnace 2, thereby discharging the granular heat medium P, heated to a sufficiently high temperature, along with the combustion gas and combustion ash, toward the first cyclone 3.

[0043] The first cyclone 3 is a solid-gas separation device that separates the combustion gas and granular heat medium P flowing from the combustion furnace 2. The first cyclone 3 draws in the combustion gas and granular heat medium P from the side, creating a downward swirling flow. This separation is achieved by utilizing the difference in specific gravity between the combustion gas, combustion ash, and granular heat medium P. The first cyclone 3 has a first discharge port 3a and a second discharge port 3b.

[0044] The first discharge port 3a is an opening for discharging combustion ash and granular heat medium P, which are solid components, toward the gasifier 1. As shown in the figure, it is located at the bottom (lower portion) of the first cyclone 3. Since the first discharge port 3a is located at the bottom of the first cyclone 3, the combustion ash and granular heat medium P, which have a higher specific gravity among the combustion gas and granular heat medium P, are discharged.

[0045] The second discharge port 3b is an opening for discharging combustion gas to the outside. As shown in the figure, it is provided in the horizontal center of the first cyclone 3. Since the second discharge port 3b is provided in the horizontal center of the first cyclone 3, the combustion gas (gas) with a lower specific gravity is discharged to the outside.

[0046] The second cyclone 4 is a solid-gas separation device that separates gas components, such as the gasification gas G, and combustion ash flowing in from the second outlet 1e of the gasifier 1. This second cyclone 4 draws in gas components and combustion ash from the side, creating a downward swirling flow. This separation utilizes the difference in specific gravity between the gas components and combustion ash. This second cyclone 4 has a discharge valve 4a and a discharge outlet 4b.

[0047] The discharge valve 4a is an on-off valve that discharges the solid combustion ash to the outside. As shown in the figure, it is located at the bottom (lower portion) of the second cyclone 4. The discharge valve 4a is located at the bottom of the second cyclone 4, so it selectively discharges the combustion ash, which has a higher specific gravity than the gas component.

[0048] The discharge port 4b is an opening for discharging gas components such as the vaporized gas G to the outside. As shown in the figure, it is provided in the horizontal center of the second cyclone 4. Since the discharge port 4b is provided in the horizontal center of the second cyclone 4, gas components with a lower specific gravity are discharged to the outside.

[0049] In the twin-tower gasifier T configured in this manner, while the granular heat medium P circulates in the order of gasifier 1, combustion furnace 2, first cyclone 3, and finally gasifier 1, a gas component including gasification gas G is generated from the feedstock N in the gasifier 1 (gasification chamber Rg). This gas component then flows from the gasifier 1 into the second cyclone 4, where combustion ash, a solid impurity, is selectively removed. The gasification gas G, which contains a small amount of gaseous impurities, is then supplied to the outside as a product of the twin-tower gasifier T.

[0050] In this twin-tower gasifier T, various operating variables are regulated by a system control device (not shown). Specifically, the twin-tower gasifier T is equipped with various sensors (not shown) for detecting various process variables, such as the supply rate (steam flow rate), flow rate (steam flow rate), and layer temperature (gasification temperature) of water vapor J in the bubble fluidized layer 1c, as well as the supply rate (air flow rate) of air A in the circulating fluidized layer 2b. The system control device performs PID control of operating variables such as the steam flow rate, steam flow rate, gasification temperature, and air flow rate based on the detection values ​​(controlled variables) of these various sensors, thereby achieving stable operation of the twin-tower gasifier T.

[0051] Next, refer to Figure 2 The abnormality diagnosis device D of this embodiment will be described. This abnormality diagnosis device D diagnoses the aforementioned twin-tower gasifier T. As shown in the figure, it includes a storage device 5, an operating device 6, and a computing device 7. This abnormality diagnosis device D is attached to the twin-tower gasifier T as an auxiliary device supporting the system control device.

[0052] The storage device 5 is a nonvolatile storage device such as a hard disk, and stores at least the learned model 5a and the physical model 5b. Furthermore, the storage device 5 stores an abnormality diagnosis program as one of the application programs executed by the operating device 6. The storage device 5 provides the learned model 5a, the physical model 5b, and the abnormality diagnosis program to the operating device 6 in response to a read command input from the operating device 6.

[0053] The learned model 5a includes a database (teacher database) containing a large amount of teacher data representing the relationship between specific process values ​​and specific component concentrations in the gasification gas G during normal operation of the twin-tower gasifier T. The learned model 5a is a mathematical model (mathematical model) obtained by machine learning of the teacher data. When the actual measured values ​​of the process values ​​are input (specified), the learned model 5a outputs predicted values ​​of the specific component concentrations corresponding to the actual measured values.

[0054] When such a learned model 5a receives inputs such as measured values ​​of steam flow rate, steam flow velocity, vaporization temperature, air flow rate, and the composition of the raw material N (raw material composition), it outputs a predicted value of the concentration of methane gas (CH4), the main component of the vaporized gas G (methane concentration), as a characteristic component concentration corresponding to these measured values. This methane concentration corresponds to the specific component concentration in the present invention.

[0055] The physical model 5b is a mathematical model that simulates the behavior of the twin-tower gasifier T under normal conditions. When input (specified) with actual measured values ​​of specific process values, the physical model 5b outputs predicted values ​​for the twin-tower gasifier T under normal conditions, which differ from the specific process values. For example, when input with the composition of the raw material N (raw material composition), steam flow rate, steam flow velocity, and air flow rate, the physical model 5b outputs a corresponding vaporization temperature. This vaporization temperature corresponds to the specific process value in the present invention.

[0056] The vaporization temperature, or the layer temperature of the bubble fluidized layer 1c, serving as a specific process value, is a physical quantity that exhibits a relatively good correlation with the methane concentration of the gasified gas G, serving as a specific component concentration. The abnormality diagnosis device D of this embodiment focuses on this correlation between the vaporization temperature and the methane concentration. Based on this correlation, it evaluates the coordination of the twin-tower gasifiers T in a spatially spaced manner, thereby identifying abnormality trends.

[0057] The operating device 6 is an input device operated by the user of the abnormality diagnosis device D, and is a pointing device such as a keyboard and / or a mouse. The user inputs various operation instructions to the operating device 6 to cause the abnormality diagnosis device D to start abnormality diagnosis processing based on the abnormality diagnosis program.

[0058] The computing device 7 is an information processing unit primarily comprised of a CPU (Central Processing Unit). The CPU executes an abnormality diagnosis program read from the storage device 5 to diagnose abnormalities in the twin-tower gasifier T (gasification system). Furthermore, the computing device 7 notifies the external user of the diagnostic results, namely, whether the twin-tower gasifier T is exhibiting abnormality trends.

[0059] In addition to the abnormality diagnosis program, the computing device 7 also performs abnormality diagnosis of the twin-tower gasifier T (gasification system) based on the learned model 5a and physical model 5b read from the storage device 5, as well as specific process values ​​obtained from the system control device, namely, raw material composition, steam flow rate, steam flow velocity, gasification temperature, and air flow rate. Details will be described later.

[0060] The computing device 7 also updates the learned model 5a stored in the storage device 5. Specifically, when the computing device 7 determines that the twin-tower gasifier T is operating normally based on the various process values ​​described above, the computing device 7 updates the learned model 5a by importing data representing the relationship between the measured values ​​of the raw material composition, steam flow rate, steam flow velocity, gasification temperature, and air flow rate and the calculated value of the methane concentration as new teaching data into the learned model 5a.

[0061] The storage device 5, operating device 6, and computing device 7 constitute evaluation means for evaluating whether the twin-tower gasifier T is exhibiting abnormal trends based on the correlation between the gasification temperature (a specific process value) in the twin-tower gasifier T and the methane concentration (a specific component concentration) of the gasification gas G output from the twin-tower gasifier T. The computing device 7 also functions as a notification means for externally notifying the evaluation results of the evaluation means.

[0062] Next, refer to Figure 3 5 , the operation of the abnormality diagnosis device D according to the present embodiment will be described in detail.

[0063] When a start instruction is input from the operating device 6, the arithmetic device 7 starts abnormality diagnosis processing of the twin-tower gasifier T according to the abnormality diagnosis program. During this abnormality diagnosis processing, the arithmetic device 7 first obtains measured values ​​of the raw material composition, steam flow rate, steam flow velocity, vaporization temperature, and air flow rate from the system control device (step S1).

[0064] The computing device 7 then analyzes the methane concentration based on the measured values ​​of these process values ​​and acquires the analyzed value as the measured methane concentration value Cr (step S2). The measured methane concentration value Cr (analyzed value) is acquired, for example, based on historical data showing the relationship between specific process values ​​(raw material composition, steam flow rate, steam flow velocity, vaporization temperature, and air flow rate) and the methane concentration.

[0065] Next, the computing device 7 inputs the raw material composition, steam flow rate, steam flow velocity, vaporization temperature, and air flow rate obtained in step S1 into the learned model 5a retrieved from the storage device 5, thereby obtaining a predicted methane concentration value Ce after a predetermined time (step S3). The computing device 7 and the learned model 5a in the storage device 5 constitute a concentration prediction unit that calculates the predicted methane concentration value Ce.

[0066] Next, the computing device 7 determines whether the predicted value Ce has fluctuated significantly relative to past normal values ​​for the methane concentration in the twin-tower gasifier T during normal conditions (step S4). Specifically, if the predicted value Ce of the methane concentration exceeds a concentration threshold value Cth, the computing device 7 determines that the predicted value Ce has fluctuated significantly relative to past normal values. If the predicted value Ce of the methane concentration does not exceed the concentration threshold value Cth, the computing device 7 determines that the predicted value Ce has not fluctuated significantly. Specifically, the concentration threshold value Cth is the limit of the range of past normal values.

[0067] Then, if the determination in step S4 is "yes," that is, if the predicted value Ce indicates a change exceeding the concentration threshold value Cth, the computing device 7 obtains the predicted value Te of the vaporization temperature from the physical model 5b (step S5). The computing device 7 and the physical model 5b in the storage device 5 constitute a temperature prediction unit for obtaining the predicted value Te of the vaporization temperature (process value).

[0068] That is, the computing device 7 inputs the raw material composition, steam flow rate, steam flow rate, vaporization temperature, and air flow rate, excluding the vaporization temperature, among the measured values ​​of the raw material composition, steam flow rate, steam flow rate, vaporization temperature, and air flow rate obtained in step S1, into the physical model 5b, thereby obtaining the vaporization temperature corresponding to the raw material composition, steam flow rate, steam flow rate, and air flow rate from the physical model 5b as the predicted value Te of the vaporization temperature.

[0069] Next, the computing device 7 calculates a difference vector Vc related to the methane concentration using the measured value Cr of the methane concentration obtained in step S2 and the predicted value Ce of the methane concentration calculated in step S3 (step S6). Specifically, the computing device 7 calculates the difference vector Vc by calculating the difference between the measured value Cr and the predicted value Ce.

[0070] Next, the computing device 7 calculates a difference vector Vt related to the vaporization temperature using the actual vaporization temperature value Tr obtained in step S1 and the predicted vaporization temperature value Te calculated in step S5 (step S6). Specifically, the computing device 7 calculates the difference vector Vt by calculating the difference between the actual vaporization temperature value Tr and the predicted vaporization temperature Te.

[0071] Next, the computing device 7 calculates the Mahalanobis distance M using the methane concentration difference vector Vc calculated in step S6 and the vaporization temperature difference vector Vt calculated in step S7 (step S8). The Mahalanobis distance M is a quantity given by the difference between the measured value Tr and the predicted value Te of the vaporization temperature, and the difference between the measured value Cr and the predicted value Ce of the methane concentration. This Mahalanobis distance M corresponds to the distance evaluation index of the present invention.

[0072] The Mahalanobis distance M is a well-known distance concept based on statistics of correlation between multiple variables. In addition, the Mahalanobis distance, which is the same concept as the Mahalanobis distance M, is described in detail in the aforementioned Patent Document 1. Due to this, a detailed description of the calculation method of the Mahalanobis distance M is omitted here, but the Mahalanobis distance M is calculated by using the covariance matrix associated with the teacher data used for learning the learning completion model 5a.

[0073] Next, the computing device 7 compares the currently calculated Mahalanobis distance M with a predetermined distance threshold Mth to determine whether the twin-tower gasifier T has reached an abnormal state after a predetermined period of time (step S9). The distance threshold Mth is a limit value of the Mahalanobis distance M set based on the past Mahalanobis distance of the twin-tower gasifier T when the gasifier T was operating normally.

[0074] Then, when the determination in step S9 is "Yes," that is, when the Mahalanobis distance M exceeds the distance threshold Mth, the calculation device 7 outputs an alarm (step S10). The alarm is output to an output device such as a separately provided display device and / or the system control device.

[0075] According to this embodiment, operational abnormalities in the twin-tower gasifier T can be predicted in advance. Therefore, the user of the abnormality diagnosis device D and / or the operator of the twin-tower gasifier T can recognize the tendency of the twin-tower gasifier T to become abnormal based on the alarm output by the abnormality diagnosis device D before the twin-tower gasifier T actually becomes abnormal.

[0076] In addition, if the determination result of step S4 is "No", the operation device 7 repeatedly performs the abnormality diagnosis process of steps S1 to S10. That is, the operation device 7 repeatedly performs the abnormality diagnosis process at predetermined time intervals according to the pre-set time schedule. Figure 4 As shown, the Mahalanobis distance M is sequentially acquired as time series data. The calculation device 7 sequentially compares the Mahalanobis distance M calculated at each cycle time with the distance threshold Mth to monitor whether the twin-tower gasifier T reaches an abnormal state during a plurality of cycle times.

[0077] If the determination in step S9 above is "No," the computing device 7 performs an update process for the teacher database in the learned model 5a (step S11). Specifically, in this case, the twin-tower gasifier T is in a normal state. Therefore, the various process values ​​acquired in step S1 and the measured methane concentration value Cr acquired in step S2 are state quantities indicating the normal state of the twin-tower gasifier T. Therefore, these process values ​​and the measured methane concentration value Cr are sufficiently qualified as new teacher data.

[0078] In response to this situation, the computing device 7 of this embodiment registers the measured values ​​of the various process values ​​and the measured value Cr of the methane concentration as new training data in the training database of the storage device 5. This new registration of training data updates the training database. This updating of the training database further enriches the machine learning in the learning completion model 5a of the storage device 5.

[0079] Furthermore, if the determination in step S4 is "No," the arithmetic device 7 in this embodiment does not proceed to steps S5 through S11 and begins the next cycle of abnormality diagnosis processing. Therefore, the abnormality diagnosis device D in this embodiment can omit steps S5 through S11, thereby reducing the computational load on the arithmetic device 7.

[0080] In addition, the present invention is not limited to the above-described embodiment, and for example, the following modified examples are conceivable.

[0081] (1) In the above embodiment, the layer temperature (gasification temperature) of the bubble fluidized layer 1c is set to a specific process value, and the methane concentration of the gasification gas G is set to a specific component concentration. However, the present invention is not limited to this. As long as the correlation between various process quantities as input values ​​related to the twin-tower gasifier T (gasification system) and various component concentrations of the gasification gas G as output values ​​changes according to the behavior of the twin-tower gasifier T, process values ​​other than the gasification temperature may be set to a specific process value, and component concentrations other than the methane concentration may be set to a specific component concentration.

[0082] For example, instead of the vaporization temperature, the steam flow rate, i.e., the flow rate of water vapor J supplied to the bubble fluidized bed 1c, may be set to a specific process value corresponding to the methane concentration (specific component concentration). This steam flow rate is a process quantity that has a correlation with the methane concentration (specific component concentration) similar to the vaporization temperature.

[0083] More specifically, the gasification temperature may be replaced with the layer temperature of the bubble fluidized layer 1c by using the representative temperature of the gasifier 1, the upper temperature of the combustion furnace 2, the middle temperature of the combustion furnace 2, the exhaust gas temperature at the outlet of the combustion furnace 2, or the methane gas temperature at the outlet of the gasifier 1. Furthermore, the component concentrations may be replaced with the methane concentration by using the hydrogen concentration, the carbon dioxide concentration, or the carbon monoxide concentration.

[0084] (2) In the above embodiment, a twin-tower gasifier T is used as the gasification system, but the present invention is not limited thereto. That is, a gasifier other than the twin-tower gasifier T may be used as the gasification system. For example, fluidized bed gasification systems and fixed bed gasification systems are known as gasification systems, and the present invention can be applied to either fluidized bed gasification system or fixed bed gasification system.

[0085] (3) In the above embodiment, the abnormality diagnosis device D is provided as a device attached to the twin-tower gasifier T. However, the present invention is not limited to this. For example, the abnormality diagnosis device D may be disposed at a location separate from the twin-tower gasifier T, i.e., the system control device, and the abnormality diagnosis device D may be connected to the system control device via a communication line to perform abnormality diagnosis of the twin-tower gasifier T.

[0086] In this case, a single abnormality diagnosis device D can be connected to the system control devices of multiple twin-tower gasifiers T via a communication line, thereby performing abnormality diagnosis on multiple twin-tower gasifiers T simultaneously and in parallel (time-sharing). This method of operating the abnormality diagnosis device D can improve the operating efficiency of the abnormality diagnosis device D, thereby reducing the cost associated with abnormality diagnosis.

[0087] (4) In the above embodiment, the Mahalanobis distance M is used as the distance evaluation index, but the present invention is not limited thereto. That is, a distance evaluation index other than the Mahalanobis distance M may be used to evaluate the abnormality tendency of the twin-tower gasifier T.

[0088] Explanation of symbols

[0089] A. Air,

[0090] D Abnormal diagnosis device,

[0091] J Water vapor,

[0092] G gasification gas,

[0093] N raw materials,

[0094] P granular heat carrier,

[0095] Rb combustion chamber,

[0096] Rg vaporization chamber,

[0097] T Twin tower gasifier (gasification system),

[0098] 1 gasifier,

[0099] 1a Raw material input port,

[0100] 1b Heating medium receiving port,

[0101] 1c Bubble flow layer,

[0102] 1d First row exit,

[0103] 1e Second row exit,

[0104] 2 combustion furnaces,

[0105] 2a Receiving port,

[0106] 2b circulating fluidized bed,

[0107] 2c discharge outlet,

[0108] 3. First cyclone,

[0109] 3a First row exit,

[0110] 3b Second row exit,

[0111] 4 Second cyclone,

[0112] 4a discharge valve,

[0113] 4b discharge outlet,

[0114] 5 storage devices,

[0115] 5a Learning completion model,

[0116] 5b Physical Model,

[0117] 6 operating devices,

[0118] 7. Computing device.

Claims

1. An abnormality diagnosis device, characterized in that: have: an evaluation unit for evaluating whether the gasification system is in an abnormal trend based on a correlation between one of a plurality of process values ​​in the gasification system and one of a plurality of component concentrations in a gasification gas generated by the gasification system; as well as a notification unit that notifies the evaluation result of the evaluation unit to the outside, The evaluation unit includes: a process value prediction unit that obtains a predicted value of the one process value; and a concentration prediction unit that obtains a predicted value of the one component concentration. The evaluation unit evaluates whether the gasification system is in a tendency to become abnormal by comparing a distance evaluation index with the distance evaluation index obtained when the gasification system is normal. The distance evaluation index is given based on the difference between the predicted value and the measured value of the one process value and the difference between the predicted value and the measured value of the one component concentration.

2. The abnormality diagnosis device according to claim 1, characterized in that: The process value prediction unit calculates a predicted value of the one process value using a physical model of the gasification system.

3. The abnormality diagnosis device according to claim 1 or 2, characterized in that: The concentration prediction unit obtains a predicted value of the concentration of the one component using a learned model of the gasification system.

4. The abnormality diagnosis device according to claim 1 or 2, characterized in that: The evaluation unit evaluates whether the gasification system is in an abnormality trend when the predicted value of the concentration of the one component exceeds a predetermined concentration threshold.

5. The abnormality diagnosis device according to claim 1 or 2, characterized in that: The one process value is the gasification temperature, and the one component concentration is the methane concentration.

6. A gasification system, characterized in that: have: The abnormality diagnosis device according to any one of claims 1 to 5.

Images