Information processing device, information processing method, and program

Abstract

Accurately predict future steam generation. [Solution] The control unit of the information processing device outputs steam generated using the thermal energy of the input exhaust gas, acquires plant data related to the amount of steam generated in a boiler configured to clean the heat transfer surface of the heat exchanger, acquires process data related to the cleaning of the heat transfer surface of the heat exchanger, derives the fouling coefficient at the first time based on at least one of the plant data and process data at the first time, and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, predicts the fouling coefficient at the second time, which is a time later than the first time, based on the derived fouling coefficient, the first state-space model, and process planning data related to the cleaning of the heat transfer surface, and predicts the amount of heat absorbed by the heat exchanger at the second time based on plant planning data related to the amount of steam generated at the second time, the predicted fouling coefficient, and the first prediction model.

Description

[Technical Field]

[0001] This invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] A well-known example of a thermal power plant equipped with a boiler and a steam turbine is a waste incineration power plant. In such a plant, the boiler outputs steam generated by utilizing the thermal energy of exhaust gases input from the incineration chamber. The output steam is supplied to a steam turbine and used for power generation.

[0003] A boiler generally consists of an evaporator and a superheater. The evaporator is a water-cooled wall located in a radiant heat transfer chamber downstream of the combustion chamber for fuel such as waste. The superheater is located in a convection heat transfer chamber downstream of the radiant heat transfer chamber. The evaporator heats water to produce steam. The superheater heats the steam to produce superheated steam, which is supplied to a steam turbine. During this process, the exhaust gas exchanges heat with the evaporator at the heat transfer surface and with the superheater at the heat transfer surface.

[0004] A technique for estimating the amount of steam generated in a boiler is disclosed, for example, in Patent Document 1.

[0005] Here, if dirt such as ash adheres to the heat transfer surface, the heat transfer performance of the heat transfer surface decreases, which reduces the efficiency of heat exchange, and consequently, the amount of steam generated in the boiler and the amount of power generated by the steam turbine also decreases. Patent Document 1 does not take into account the effect of dirt in the boiler, so there is room for improvement in the method of estimating the amount of steam generated.

[0006] Techniques for estimating or considering contamination on heat transfer surfaces are disclosed, for example, in Patent Documents 2 and 3. Cleaning devices for removing contamination adhering to heat transfer surfaces are also known. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Patent No. 7216566 [Patent Document 2] Japanese Patent Application Publication No. 5-280703 [Patent Document 3] Japanese Patent Application Publication No. 10-227404 [Overview of the project] [Problems that the invention aims to solve]

[0008] Fouling on the heat transfer surfaces of a boiler significantly impacts plant performance, including steam generation and, consequently, power generation. Therefore, there has been a need for technology that can more accurately predict future steam generation in boilers, taking into account heat transfer surface fouling.

[0009] The present invention has been made in view of the above, and its object is to provide an information processing device, an information processing method, and a program that can more accurately predict the future amount of steam generated in a boiler. [Means for solving the problem]

[0010] To solve the above-mentioned problems and achieve the objective, one aspect of the present invention is an information processing device comprising a control unit having hardware, the control unit acquires plant data related to the amount of steam generated in a boiler having a heat exchanger, which outputs steam generated using the thermal energy of input exhaust gas, and is configured to be able to clean the heat transfer surface of the heat exchanger, acquires process data related to cleaning the heat transfer surface of the heat exchanger, derives a fouling coefficient at a first time based on at least one of the plant data and the process data at a first time and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, predicts a fouling coefficient at a second time which is a time later than the first time based on the derived fouling coefficient, the first state-space model and process planning data related to cleaning the heat transfer surface, and predicts the amount of heat absorbed by the heat exchanger at the second time based on plant planning data related to the amount of steam generated at the second time, the predicted fouling coefficient and the first prediction model.

[0011] The control unit may correct the derived contamination coefficient by data assimilation processing and predict the contamination coefficient at the second time based on the corrected contamination coefficient.

[0012] The control unit may derive the degree of difficulty in removing dirt at the first time based on at least one of the plant data and the process data at the first time, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable. It may then predict the degree of difficulty in removing dirt at the second time based on the derived degree of difficulty in removing dirt, the second state-space model, and the process plan data, and further predict the amount of heat absorbed by the heat exchanger at the second time based on the predicted degree of difficulty in removing dirt.

[0013] The control unit may correct the derived difficulty level of stain removal by data assimilation processing and predict the difficulty level of stain removal at the second time based on the corrected difficulty level of stain removal.

[0014] The control unit may predict the amount of steam generated in the boiler at the second time.

[0015] The boiler may have at least one of an evaporator and a superheater as the heat exchanger.

[0016] One aspect of the present invention is an information processing method executed by an information processing apparatus including a control unit having hardware. The control unit has a heat exchanger, outputs steam generated by using the thermal energy of the input exhaust gas, acquires plant data related to the amount of steam generated in a boiler configured to be able to clean the heat transfer surface of the heat exchanger, acquires process data related to the cleaning of the heat transfer surface of the heat exchanger, and based on at least one of the plant data and the process data at a first time and a first state space model including a fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, derives the fouling coefficient at the first time, and based on the derived fouling coefficient, the first state space model, and process plan data related to the cleaning of the heat transfer surface, predicts the fouling coefficient at a second time, which is a time after the first time, and based on plant plan data related to the amount of steam generated at the second time, data of the predicted fouling coefficient, and a first prediction model, predicts the heat absorption amount of the heat exchanger at the second time.

[0017] The control unit may correct the derived fouling coefficient by data assimilation processing and predict the fouling coefficient at the second time based on the corrected fouling coefficient.

[0018] The control unit may derive the degree of difficulty in removing dirt at the first time based on at least one of the plant data and the process data at the first time, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable. It may then predict the degree of difficulty in removing dirt at the second time based on the derived degree of difficulty in removing dirt, the second state-space model, and the process plan data, and further predict the amount of heat absorbed by the heat exchanger at the second time based on the predicted degree of difficulty in removing dirt.

[0019] The control unit may correct the derived difficulty level of stain removal by data assimilation processing and predict the difficulty level of stain removal at the second time based on the corrected difficulty level of stain removal.

[0020] The control unit may predict the future amount of steam generated in the boiler at the second time point.

[0021] The boiler may have at least one of an evaporator and a superheater as the heat exchanger.

[0022] One aspect of the present invention is a program that causes a hardware-based control unit to acquire plant data related to the amount of steam generated in a boiler having a heat exchanger, which outputs steam generated using the thermal energy of input exhaust gas, and is configured to be able to clean the heat transfer surface of the heat exchanger; acquire process data related to cleaning the heat transfer surface of the heat exchanger; derive the fouling coefficient at the first time based on at least one of the plant data and the process data at the first time and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable; predict the fouling coefficient at a second time, which is a time later than the first time, based on the derived fouling coefficient, the first state-space model, and process planning data related to cleaning the heat transfer surface; and predict the amount of heat absorbed by the heat exchanger at the second time based on plant planning data related to the amount of steam generated at the second time, the predicted fouling coefficient data, and the first prediction model.

[0023] The control unit may be instructed to correct the derived contamination coefficient by data assimilation processing and to predict the contamination coefficient at the second time point based on the corrected contamination coefficient.

[0024] The control unit may be instructed to derive the degree of difficulty in removing dirt at the first time based on at least one of the plant data and the process data at the first time, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable; predict the degree of difficulty in removing dirt at the second time based on the derived degree of difficulty in removing dirt, the second state-space model, and the process plan data; and further predict the amount of heat absorbed by the heat exchanger at the second time based on the predicted degree of difficulty in removing dirt.

[0025] The control unit may be instructed to correct the derived difficulty level of stain removal by data assimilation processing, and to predict the difficulty level of stain removal at the second time based on the corrected difficulty level of stain removal.

[0026] The control unit may be instructed to predict the amount of steam generated in the boiler at the second time point.

[0027] The boiler may have at least one of an evaporator and a superheater as the heat exchanger. [Effects of the Invention]

[0028] According to the information processing device, information processing method, and program of the present invention, it is possible to more accurately predict the future amount of steam generated in a boiler. [Brief explanation of the drawing]

[0029] [Figure 1] Figure 1 is a schematic overall diagram showing a waste incineration power plant to which the prediction device according to the embodiment is applied. [Figure 2] Figure 2 is a block diagram showing the configuration of the prediction device. [Figure 3] Figure 3 is a graph showing an example of the changes over time in the outlet temperature of the radiant heat transfer chamber, the fouling coefficient in the evaporator, and the difficulty of fouling removal. [Modes for carrying out the invention]

[0030] Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In all the drawings of the following embodiment, the same or corresponding parts will be denoted by the same reference numerals. Furthermore, the present invention is not limited to the embodiment described below.

[0031] (Waste Incineration Power Plant) Figure 1 is a schematic overall diagram showing a waste incineration power plant to which the prediction device according to the embodiment is applied. As shown in Figure 1, the waste incineration power plant 100 comprises an incinerator 1, a boiler 2, a steam turbine 3, a dust removal device 4, a chimney 5, a cleaning control device 6, and a prediction device 7. The cleaning control device 6 and the prediction device 7 are configured to communicate with each other via a network 8.

[0032] Incinerator 1 is a facility that burns waste and discharges high-temperature exhaust gas. Incinerator 1 is, for example, a grate-type waste incinerator.

[0033] Boiler 2 generates steam using the thermal energy of the exhaust gas that enters the boiler inlet from the outlet of incinerator 1, and outputs it to steam turbine 3. Boiler 2 is equipped with a radiant heat transfer chamber 10 on the upstream side and a convection heat transfer chamber 11 on the downstream side in the direction of exhaust gas flow.

[0034] The radiant heat transfer chamber 10 is equipped with an evaporator 12, which is an example of a heat exchanger. This evaporator 12 is also called, for example, a water-cooled wall. Water is supplied to the evaporator 12 from the water supply section 14. The evaporator 12 outputs steam generated by heat exchange between the water flowing inside and the exhaust gas to the steam drum 15. The convection heat transfer chamber 11 is equipped with a superheater 13, which is another example of a heat exchanger. The superheater 13 is supplied with steam from the steam drum 15. The superheater 13 outputs superheated steam generated by heat exchange between the steam flowing inside and the exhaust gas to the steam turbine 3.

[0035] Furthermore, boiler 2 is equipped with a cleaning device 16 for cleaning the heat transfer surface of evaporator 12. In addition, boiler 2 is equipped with a cleaning device 17 for cleaning the heat transfer surface of superheater 13. In other words, boiler 2 is configured to allow cleaning of the heat transfer surfaces of both evaporator 12 and superheater 13. Cleaning device 16 is a cleaning device that cleans the heat transfer surface of evaporator 12 by spraying it with liquid or steam, for example. Cleaning device 17 is a cleaning device that cleans the heat transfer surface of superheater 13 by applying a pressure wave of combustion gas, for example.

[0036] The steam turbine 3 generates electricity by receiving superheated steam from the superheater 13.

[0037] The dust removal device 4 removes soot and other particles from the exhaust gas output from the boiler outlet of the boiler 2. The dust removal device 4 is composed of, for example, a bag filter.

[0038] The chimney 5 releases exhaust gas, from which soot and other particles have been removed by the dust removal device 4, into the atmosphere.

[0039] The equipment described above is equipped with sensor groups 21 to 27. Sensor group 21 is located at the inlet of the radiant heat transfer chamber 10 (radiant heat transfer chamber inlet or boiler inlet). Sensor group 21 includes a temperature sensor for measuring the temperature of the exhaust gas input to the radiant heat transfer chamber 10, a flow rate sensor for measuring the flow rate of the exhaust gas, a concentration sensor for measuring the concentration of components in the exhaust gas, and the like.

[0040] The sensor group 22 is located at the outlet of the radiant heat transfer chamber 10 and the inlet of the convection heat transfer chamber 11 (either the radiant heat transfer chamber inlet or the convection heat transfer chamber inlet). The sensor group 22 includes a temperature sensor that measures the temperature of the exhaust gas output from the radiant heat transfer chamber 10 and input into the convection heat transfer chamber 11, a flow rate sensor that measures the flow rate of the exhaust gas, a concentration sensor that measures the concentration of components in the exhaust gas, and the like.

[0041] The sensor group 23 is installed at the outlet of the convection heat transfer chamber 11 (convection heat transfer chamber outlet or boiler outlet). The sensor group 23 includes a temperature sensor for measuring the temperature of the exhaust gas output from the convection heat transfer chamber 11, a flow rate sensor for measuring the flow rate of the exhaust gas, a concentration sensor for measuring the concentration of components in the exhaust gas, and the like.

[0042] The sensor group 24 is installed in the middle of the water channel connecting the water supply unit 14 and the evaporator 12. The sensor group 24 includes a temperature sensor for measuring the temperature of the water supplied from the water supply unit 14 to the evaporator 12, a flow sensor for measuring the flow rate of the water, and the like.

[0043] Sensor group 25 is installed in the middle of the waterway connecting the evaporator 12 and the steam drum 15. Sensor group 24 includes a temperature sensor for measuring the temperature of the steam flowing out of the evaporator 12 to the steam drum 15, a flow sensor for measuring the steam flow rate, and the like.

[0044] The sensor group 26 is installed in the middle of the steam path connecting the steam drum 15 and the superheater 13. The sensor group 26 includes a temperature sensor for measuring the temperature of the steam supplied from the steam drum 15 to the superheater 13, a flow sensor for measuring the steam flow rate, and the like.

[0045] The sensor group 27 is installed in the steam path connecting the superheater 13 and the steam turbine 3. The sensor group 27 includes a temperature sensor for measuring the temperature of the steam supplied from the superheater 13 to the steam turbine 3, a flow rate sensor for measuring the steam flow rate, and the like.

[0046] The physical quantity data measured by the sensor groups 21-27 described above is an example of plant data related to the amount of steam generated in boiler 2. This data is transmitted to the prediction device 7 and stored chronologically in the memory unit 7b of the prediction device 7, which will be described later. The plant data may also be transmitted to the prediction device 7 via the network 8.

[0047] The cleaning control device 6 is a control device that controls the operation of the cleaning devices 16 and 17. The cleaning control device 6 controls the timing of cleaning by the cleaning devices 16 and 17 (cleaning start time and cleaning end time), cleaning intensity, etc. The cleaning timing and cleaning intensity are examples of process data related to cleaning the heat transfer surface of the evaporator 12 and the heat transfer surface of the superheater 13. This data is stored over time in the memory of the computer that constitutes the cleaning control device 6 and transmitted to the prediction device 7 via the network 8.

[0048] Next, the prediction device 7 will be described. Figure 2 is a block diagram showing the configuration of the prediction device 7. The prediction device 7 comprises a control unit 7a, a storage unit 7b, and a communication unit 7c.

[0049] The control unit 7a includes a processor such as a CPU (Central Processing Unit) with hardware, and a main memory unit such as RAM (Random Access Memory) and ROM (Read Only Memory). The control unit 7a loads the program stored in the storage unit 7b into the working area of ​​the main memory unit and executes it, thereby realizing a function that matches a predetermined purpose through the execution of the program. The storage unit 7b is composed of a storage medium selected from volatile memory such as RAM, non-volatile memory such as ROM, EPROM (Erasable Programmable ROM), hard disk drive (HDD), and removable media. Removable media include, for example, USB (Universal Serial Bus) memory, or disk recording media such as CD (Compact Disc), DVD (Digital Versatile Disc), or BD (Blu-ray® Disc). Alternatively, the storage unit 7b may be configured using a computer-readable recording medium such as an externally insertable memory card.

[0050] The storage unit 7b can store the operating system (OS), various programs, various tables, and various databases necessary for the operation of the prediction device 7. These programs can also be recorded on computer-readable recording media such as hard disks, flash memory, CD-ROMs, DVD-ROMs, and flexible disks and widely distributed. The storage unit 7b may also be located on another server that can communicate via various networks.

[0051] The communication unit 7c is, for example, a LAN (Local Area Network) interface board or a wireless communication circuit for wireless communication. The LAN interface board and wireless communication circuit are connected to the network 8. The communication unit 7c is connected to the network 8 and configured to communicate with the cleaning control device 6 and other devices and servers.

[0052] Network 8 consists of LANs, internet networks, and mobile phone networks. Network 8 may include, for example, public communication networks such as the internet, as well as other communication networks such as WANs (Wide Area Networks), telephone communication networks such as mobile phones, and wireless communication networks such as Wi-Fi. Considering the security of transmitted and received data, the communication line between the cleaning control device 6 and the prediction device 7 can be a dedicated line or a VPN line. The cleaning control device 6 and the prediction device 7 may also be configured as a single unit, and they may be installed in the same facility or in different facilities.

[0053] (Operation of the prediction device) Next, the operation of the prediction device 7 will be described. First, the control unit 7a of the prediction device 7 derives the fouling coefficients of the heat transfer surfaces of the evaporator 12 and the superheater 13 at a given first time, based on at least one of the plant data and process data at a given first time, and a first state-space model that includes a pre-constructed fouling coefficient as a state variable.

[0054] Here, the fouling coefficient is a measure of how much the heat transfer performance of a heat transfer surface is impaired by fouling adhering to that surface. While not limited to this, for example, the fouling coefficient has the following relationship to the overall heat transfer coefficient: 1 / k = 1 / α1 + 1 / α2 + δ / λ + x f

[0055] Note that k is the overall heat transfer coefficient [W / (m 2 ·K) is the heat transfer coefficient [W / (m²) between the high-temperature side fluid (exhaust gas in this embodiment) and the surface of the heat transfer surface on the high-temperature side fluid side. 2 ·K) is the heat transfer coefficient [W / (m²) between the low-temperature fluid (water or steam in this embodiment) and the surface of the heat transfer surface on the low-temperature fluid side. 2 x is (·K), where δ is the thickness of the heat transfer surface [m], λ is the thermal conductivity of the material (e.g., metal) constituting the heat transfer surface [W / (m·K)], and x f The fouling coefficient [(m2 The formula is [·K) / W]. As can be seen from the above formula, the larger the fouling coefficient, the smaller the overall heat transfer coefficient, and the lower the heat transfer performance of the heat transfer surface.

[0056] Through diligent research by the inventors, a state equation can be defined as a mathematical state-space model (first state-space model) in which the fouling coefficient and at least one of plant data and process data are state variables. Therefore, the fouling coefficient can be derived using this state equation, and the change in the fouling coefficient over time can be accurately captured.

[0057] Furthermore, the control unit 7a derives the degree of difficulty in removing fouling when cleaning the heat transfer surfaces of the evaporator 12 and the superheater 13, based on at least one of the plant data and process data, and a pre-constructed second state-space model that includes the degree of difficulty in removing fouling as a state variable.

[0058] Here, the difficulty of removing dirt is a measure of the difficulty of removing dirt when cleaning the heat transfer surface. While not limited to this, for example, the difficulty of removing dirt has the following relationship with the dirt coefficient: (Difficulty of stain removal) = (Cleaning strength) / [(Stain coefficient before cleaning) - (Stain coefficient after cleaning)]

[0059] As can be seen from the above formula, the smaller the difference in the soiling coefficient before and after cleaning, the greater the difficulty in removing the dirt. In other words, even with the same cleaning intensity, the greater the difficulty in removing the dirt, the less likely it is that the soiling coefficient after cleaning will decrease. Furthermore, using the above formula, the soiling coefficient after cleaning can be calculated from the soiling coefficient before cleaning, the cleaning intensity, and the difficulty in removing the dirt.

[0060] The cleaning strength can be determined, for example, as follows: For example, in a cleaning device that cleans by spraying liquid or steam onto a heat transfer surface, the cleaning strength may be determined by the product of the injection flow rate per unit area and the injection time. For example, in a cleaning device that cleans by applying a pressure wave of combustion gas to a heat transfer surface, the cleaning strength may be determined by the product of the amount of fuel used to generate the pressure wave per unit area and the number of repetitions.

[0061] Through diligent research by the inventors, a state equation can be defined as a mathematical state-space model (second state-space model) in which the degree of difficulty in removing fouling and at least one of the plant data and process data are state variables. Therefore, the degree of difficulty in removing fouling can be derived using this state equation, and the change in the degree of difficulty in removing fouling over time can be accurately captured.

[0062] Furthermore, the control unit 7a predicts the fouling coefficient and the difficulty of fouling removal at the second time point based on each state-space model and process plan data related to the cleaning of the heat transfer surface at the second time point. The second time point is a time later than the first time point and is a future time. Here, the process plan data related to the cleaning of the heat transfer surface at the second time point is the plan data corresponding to the process data for cleaning at the second time point.

[0063] In other words, the control unit 7a can predict future soiling coefficients and soiling difficulty levels based on soiling coefficients and soiling difficulty levels derived as past or present values.

[0064] Furthermore, the control unit 7a predicts the amount of heat absorbed by the evaporator 12 and the amount of heat absorbed by the superheater 13 at the second time, based on the plant planning data at the second time, the predicted fouling coefficient and fouling removal difficulty data, and the first prediction model.

[0065] Here, the plant planning data at the second time step refers to the planning data for the exhaust gas to be input to the boiler 2 at the second time step, and includes, for example, at least one of the data for the quantity, temperature, or composition of the exhaust gas. Such data for the quantity, temperature, or composition of the exhaust gas can be derived, for example, based on planning data such as the type and quantity of waste to be burned in the incinerator 1, the air ratio during combustion, and the combustion temperature.

[0066] While not limited to these, for example, the first predictive model for evaporator 12 is constructed based on a heat acquisition model. The heat acquisition model consists of a combination of convective heat transfer, radiative heat transfer, and conductive heat transfer. For example, the amount of heat acquired by evaporator 12 is calculated by a conventional method using the temperature, component concentration, and flow rate of the exhaust gas input to the radiative heat transfer chamber 10, measured by sensor group 21; the temperature, component concentration, and flow rate of the exhaust gas output from the radiative heat transfer chamber 10, measured by sensor group 22; the temperature and flow rate of the water supplied to evaporator 12, measured by sensor group 24; the temperature and flow rate of the steam output from evaporator 12, measured by sensor group 25; and the area, shape, and fouling coefficient of the heat transfer surface of evaporator 12. Therefore, the fouling coefficient of evaporator 12 can be calculated inversely from the amount of heat acquired by evaporator 12 and the above data. The amount of heat absorbed by the evaporator 12 when performing this reverse calculation is derived from the temperature and flow rate of the water supplied to the evaporator 12, and the temperature and flow rate of the steam output from the evaporator 12.

[0067] Furthermore, although not limited thereto, for example, the first prediction model for the superheater 13 is constructed based on the heat acquisition model. The heat acquisition model consists of a combination of convective heat transfer, radiant heat transfer, and conductive heat transfer. For example, the amount of heat acquired by the superheater 13 is calculated by a conventional method using the temperature, component concentration, and flow rate of the exhaust gas input to the convective heat transfer chamber 11 (i.e., output from the radiant heat transfer chamber 10), measured by sensor group 22; the temperature, component concentration, and flow rate of the exhaust gas output from the convective heat transfer chamber 11, measured by sensor group 23; the temperature and flow rate of the steam supplied to the superheater 13, measured by sensor group 26; the temperature and flow rate of the steam output from the superheater 13, measured by sensor group 27; and the area, shape, and fouling coefficient of the heat transfer surface of the superheater 13, which have been measured in advance. Therefore, the fouling coefficient of the superheater 13 can be calculated inversely from the amount of heat acquired by the superheater 13 and the above data. The amount of heat absorbed by the superheater 13 when performing this reverse calculation is derived from the temperature and flow rate of the steam supplied to the superheater 13, and the temperature and flow rate of the steam output from the superheater 13.

[0068] Furthermore, the control unit 7a can calculate and predict the amount of steam generated in the boiler 2 at a second time point based on the predicted values ​​of the heat recovered from the evaporator 12 and the heat recovered from the superheater 13.

[0069] The fouling coefficient and the difficulty of fouling removal will be explained in more detail. Figure 3 is a graph showing an example of the change over time between the exhaust gas temperature at the outlet of the radiant heat transfer chamber 10 (outlet temperature), the fouling coefficient in the evaporator 12, and the difficulty of fouling removal. The horizontal axis represents the number of operating days of boiler 2.

[0070] Even if the temperature and flow rate of the exhaust gas at the inlet of the radiant heat transfer chamber 10 remain approximately constant, the outlet temperature and fouling coefficient increase as the number of operating days progress. This is because as the amount of fouling adhering to the heat transfer surface of the evaporator 12 increases due to the exhaust gas, the heat transfer efficiency of the heat transfer surface decreases, reducing the efficiency of heat exchange between the exhaust gas and water, and causing the exhaust gas to be output at a relatively high temperature. Furthermore, as fouling that cannot be completely washed accumulates on the heat transfer surface and the degree of fouling adhesion increases over time, it becomes more difficult to remove the fouling, thus increasing the difficulty of fouling removal.

[0071] Here, if the heat transfer surface is cleaned at the timing indicated by the arrow in the diagram, the fouling coefficient decreases as the dirt is removed, and the outlet temperature also decreases. However, the difficulty of removing the dirt continues to increase. If cleaning is repeated under the same cleaning conditions, the fouling coefficient decreases each time, the heat transfer performance improves, and the outlet temperature decreases, but since the difficulty of removing the dirt gradually increases, the value after the decrease will rise over time.

[0072] Figure 3 shows the changes over time in the fouling coefficient and the difficulty of removing fouling in the evaporator 12, and the changes over time in the fouling coefficient and the difficulty of removing fouling in the superheater 13 show a similar trend.

[0073] From the above, accurately understanding the fouling coefficient and the difficulty of fouling removal is important for predicting the future steam generation amount in boiler 2 and the power generation amount of steam turbine 3, and is also important for determining the cleaning plan and setting of cleaning conditions in cleaning devices 16 and 17. For example, if the difficulty of fouling removal is known, it is possible to estimate the fouling coefficient after cleaning, and consequently, to predict the amount of heat recovered after cleaning more accurately. In addition, by knowing the difficulty of fouling removal, cleaning operations can be carried out at an appropriate frequency or intensity to obtain the desired cleaning effect, and planned cleaning (setting process plan data) can be carried out.

[0074] Furthermore, in the prediction device 7, the control unit 7a corrects the fouling coefficient and fouling removal difficulty using data assimilation processing with a Kalman filter or the like, and can predict the fouling coefficient and fouling removal difficulty at a second time point based on the corrected fouling coefficient and fouling removal difficulty. Data assimilation processing is a process that corrects state variables using observed variables. Here, the observed variables that are actually measured are the temperature, flow rate, and component concentration of the exhaust gas. As a result, the prediction device 7 can take into account the future operating conditions of the boiler 2 and more accurately predict the future steam generation amount in the boiler 2 and the power generation amount of the steam turbine 3, and can more effectively determine the cleaning plan and cleaning conditions in the cleaning devices 16 and 17. The data for the future operating conditions of the boiler 2 is stored, for example, in the storage unit 7b of the prediction device 7.

[0075] Furthermore, by sequentially performing corrections through such data assimilation processing, the accuracy of predictions can be further improved.

[0076] As described above, in the prediction device 7, based on the theoretical model considering the fouling coefficient and the difficulty of fouling removal, and the state space model constructed based on the plant data and process data acquired over time, the fouling coefficient and the difficulty of fouling removal are derived, and furthermore, these fluctuations can be corrected by data assimilation processing using a Kalman filter or the like, so that the future water vapor generation amount can be predicted more accurately. Furthermore, by performing the prediction of the water vapor generation amount in real time, the combustion in the incinerator 1 and the fouling removal by the cleaning devices 16 and 17 can be controlled.

[0077] (State equation) Hereinafter, a specific example of the state equation with the fouling coefficient and the difficulty of fouling removal as state variables will be described.

[0078] The state variables can be set as in the following equation (1). [Number] Here, x f represents the fouling coefficient, and x d represents the difficulty of fouling removal.

[0079] When the time is k, the state equation can be defined as in the following equation (2). [Number]

[0080] Equation (2) can be expressed as, for example, equations (3-1) and (3-2). Here, equation (3-1) is the state equation corresponding to the first state space model, and equation (3-2) is the state equation corresponding to the second state space model. [Number] Here a is the fouling accumulation rate, and x f (k) represents the degree of influence on x f (k + 1). b is the cleaning effect coefficient, and g(x f(k), x d (k) and u(k) are cleaning effect functions, and the second term of equation (3-1) represents the overall cleaning effect. The cleaning efficiency function can be illustrated as shown in equation (4).

number

[0081] The observation equation can be defined as shown in equation (5) below.

number

[0082] Equation (5) can be expressed as, for example, equations (6-1) and (6-2).

number

[0083] Also, h t (x f (k), xd (k)), h s (x f (k), x d (k)) can be expressed, for example, as in equations (7-1) and (7-2).

number

[0084] In the state-space model constructed as described above, the Unscented Kalman Filter, a type of nonlinear Kalman filter, can be applied as a method for sequentially estimating the state variables of the model.

[0085] Specifically, the values ​​of each state variable are estimated using a Kalman filter to correct the error between the estimated values ​​y(k) of the exhaust gas temperature and heat transfer amount after heat transfer for the state variable x(k) (fouling coefficient, difficulty of fouling removal) at time k, and the observed values ​​of the exhaust gas temperature and heat transfer amount after heat transfer.

[0086] This Kalman filter-based state estimation is performed sequentially from the past to a certain point in time, and the estimation accuracy improves as the estimation is repeated.

[0087] Based on the above, it is possible to estimate the fouling state in a boiler (fouling coefficient and difficulty of fouling removal) with high accuracy.

[0088] Furthermore, in order to sequentially estimate the state variables, various methods commonly used as sequential estimation filters (sequential Bayesian filters) for nonlinear models may be used, not limited to the Unscented Kalman Filter. For example, it is possible to apply the Ensemble Kalman Filter, Particle Filter, etc.

[0089] Furthermore, the state variable x(k) may be either the fouling coefficient of the heat exchanger, the difficulty of fouling removal, or both. The error between the estimated value y(k) and the observed value may be corrected by either the temperature of the exhaust gas after heat transfer, the amount of heat transferred, or both.

[0090] Furthermore, the estimated fouling coefficient and fouling removal difficulty can be used to determine the optimal cleaning timing and intensity. For example, an operational strategy can be developed to perform cleaning only when the fouling coefficient exceeds a certain threshold and the fouling removal difficulty is within an acceptable range. In addition, monitoring the trend of fouling removal difficulty can help understand the long-term performance degradation trend of the boiler and use this information to optimize maintenance plans.

[0091] (Recording medium) In the embodiment described above, a program capable of executing the information processing method by the prediction device 7 can be recorded on a recording medium readable by a computer or other machine or device (hereinafter referred to as "computer, etc."). By having the computer, etc. read and execute the program on the recording medium, the computer functions as the prediction device 7. Here, a recording medium readable by a computer, etc. refers to a non-temporary recording medium that stores information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer, etc. Examples of such recording media that can be removed from a computer, etc. include flexible disks, magneto-optical disks, CD-ROMs, CD-R / Ws, DVDs, BDs, DATs, magnetic tapes, and memory cards such as flash memory. Examples of recording media fixed to a computer, etc. include hard disks and ROMs. Furthermore, SSDs can be used as both a recording medium that can be removed from a computer, etc. and a recording medium that is fixed to a computer, etc.

[0092] Furthermore, the program to be executed by the prediction device 7 may be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network.

[0093] (Other embodiments) Furthermore, in the prediction device 7, the aforementioned "part" can be replaced with "circuit," etc. For example, the communication part can be replaced with a communication circuit.

[0094] Further effects and modifications can be readily derived by those skilled in the art. Broader aspects of the present invention are not limited to the specific details and representative embodiments expressed and described above. Accordingly, various modifications are possible without departing from the spirit or scope of the overall concept of the invention as defined by the appended claims and their equivalents.

[0095] For example, in the above embodiment, the boiler has an evaporator and a superheater as heat exchangers, but it may have only one of them.

[0096] Furthermore, for example, the prediction of the heat exchanger's heat gain at the second time step is not limited to being based on plant planning data, fouling coefficient, fouling removal difficulty, and the first prediction model, but may also be based on plant planning data, fouling coefficient, and the first prediction model.

[0097] Furthermore, the types of numerical values ​​and information mentioned in the above-described embodiment are merely examples, and different types of numerical values ​​and information may be used as needed. The present invention is not limited by the description and drawings that constitute part of the disclosure of the present invention in the above-described embodiment. [Explanation of Symbols]

[0098] 1: Incinerator 2: Boiler 3: Steam turbine 4: Dust removal device 5: Chimney 6: Washing control device 7: Prediction device 7a: Control part 7b: Storage section 7c:Communication Department 8: Network 10: Radiation Heat Transfer Chamber 11: Convection Heat Transfer Chamber 12: Evaporator 13:Superheater 14: Water supply section 15: Steam drum 16,17: Washing equipment 21, 22, 23, 24, 25, 26, 27: Sensor group 100: Waste Incineration Power Plant

Claims

1. It includes a control unit with hardware, The control unit, In a boiler equipped with a heat exchanger, which outputs steam generated using the thermal energy of the input exhaust gas, and configured to allow cleaning of the heat transfer surface of the heat exchanger, plant data related to the amount of steam generated is acquired. Process data related to the cleaning of the heat transfer surface of the heat exchanger is obtained. Based on at least one of the plant data and process data at the first time step, and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, the fouling coefficient at the first time step is derived. Based on the derived fouling coefficient, the first state-space model, and the process plan data related to the cleaning of the heat transfer surface, the fouling coefficient at a second time, which is later than the first time, is predicted. Based on the plant planning data related to the amount of steam generated at the second time point, the predicted fouling coefficient, and the first prediction model, the amount of heat absorbed by the heat exchanger at the second time point is predicted. Information processing device.

2. The control unit corrects the derived contamination coefficient through data assimilation processing and predicts the contamination coefficient at the second time point based on the corrected contamination coefficient. The information processing apparatus according to claim 1.

3. The control unit, Based on at least one of the plant data and process data at the first time step, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable, the degree of difficulty in removing dirt at the first time step is derived. Based on the derived difficulty level for removing the dirt, the second state-space model, and the process planning data, the difficulty level for removing the dirt at the second time step is predicted. Furthermore, based on the predicted difficulty of removing the dirt, the amount of heat absorbed by the heat exchanger at the second time point is predicted. The information processing apparatus according to claim 1.

4. The control unit corrects the derived difficulty level of stain removal by data assimilation processing, and predicts the difficulty level of stain removal at the second time based on the corrected difficulty level of stain removal. The information processing apparatus according to claim 3.

5. The control unit predicts the amount of steam generated in the boiler at the second time point. The information processing apparatus according to claim 1.

6. The boiler has at least one of an evaporator and a superheater as the heat exchanger. The information processing apparatus according to claim 1.

7. An information processing method performed by an information processing device equipped with a control unit having hardware, The control unit, In a boiler equipped with a heat exchanger, which outputs steam generated using the thermal energy of the input exhaust gas, and configured to allow cleaning of the heat transfer surface of the heat exchanger, plant data related to the amount of steam generated is acquired. Process data related to the cleaning of the heat transfer surface of the heat exchanger is obtained. Based on at least one of the plant data and process data at the first time step, and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, the fouling coefficient at the first time step is derived. Based on the derived fouling coefficient, the first state-space model, and the process plan data related to the cleaning of the heat transfer surface, the fouling coefficient at a second time point, which is later than the first time point, is predicted. Based on the plant planning data related to the amount of steam generated at the second time point, the predicted fouling coefficient data, and the first prediction model, the amount of heat absorbed by the heat exchanger at the second time point is predicted. Information processing methods.

8. The control unit corrects the derived contamination coefficient through data assimilation processing and predicts the contamination coefficient at the second time point based on the corrected contamination coefficient. The information processing method according to claim 7.

9. The control unit, Based on at least one of the plant data and process data at the first time step, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable, the degree of difficulty in removing dirt at the first time step is derived. Based on the derived difficulty level for removing the dirt, the second state-space model, and the process planning data, the difficulty level for removing the dirt at the second time step is predicted. Furthermore, based on the predicted difficulty of removing the dirt, the amount of heat absorbed by the heat exchanger at the second time point is predicted. The information processing method according to claim 7.

10. The control unit corrects the derived difficulty level of stain removal by data assimilation processing, and predicts the difficulty level of stain removal at the second time based on the corrected difficulty level of stain removal. The information processing method according to claim 9.

11. The control unit predicts the future amount of steam generated in the boiler at the second time point. The information processing method according to claim 7.

12. The boiler has at least one of an evaporator and a superheater as the heat exchanger. The information processing method according to claim 7.

13. A control unit having hardware, In a boiler equipped with a heat exchanger, which outputs steam generated using the thermal energy of the input exhaust gas, and configured to allow cleaning of the heat transfer surface of the heat exchanger, plant data related to the amount of steam generated is acquired. Process data related to the cleaning of the heat transfer surface of the heat exchanger is obtained. Based on at least one of the plant data and process data at the first time step, and a first state-space model that includes the fouling coefficient of the heat transfer surface of the heat exchanger as a state variable, the fouling coefficient at the first time step is derived. Based on the derived fouling coefficient, the first state-space model, and the process plan data related to the cleaning of the heat transfer surface, the fouling coefficient at a second time, which is later than the first time, is predicted. Based on the plant planning data related to the amount of steam generated at the second time point, the predicted fouling coefficient data, and the first prediction model, the amount of heat absorbed by the heat exchanger at the second time point is predicted. A program that performs an action.

14. The control unit, The derived contamination coefficient is corrected through data assimilation processing, and the contamination coefficient at the second time point is predicted based on the corrected contamination coefficient. The program according to claim 13 that causes the following to be performed.

15. The control unit, Based on at least one of the plant data and process data at the first time step, and a second state-space model that includes the degree of difficulty in removing dirt, which is a measure of the difficulty in removing dirt when cleaning the heat transfer surface of the heat exchanger, as a state variable, the degree of difficulty in removing dirt at the first time step is derived. Based on the derived difficulty level for removing the dirt, the second state-space model, and the process planning data, the difficulty level for removing the dirt at the second time step is predicted. Furthermore, based on the predicted difficulty of removing the dirt, the amount of heat absorbed by the heat exchanger at the second time point is predicted. The program according to claim 13 that causes the following to be performed.

16. The control unit, The derived difficulty level for removing dirt is corrected through data assimilation processing, and the difficulty level for removing dirt at the second time point is predicted based on the corrected difficulty level for removing dirt. The program according to claim 15 that causes the following to be performed.

17. The control unit, Predict the amount of steam generated in the boiler at the second time point. The program according to claim 13 that causes the following to be performed.

18. The boiler has at least one of an evaporator and a superheater as the heat exchanger. The program according to claim 13.

Images