JP5378288B2 - Plant control device and thermal power plant control device - Google Patents

Description

  The present invention relates to a plant control device, and more particularly to a thermal power plant control device that generates power using fossil fuels such as coal.

  The plant control device processes a measurement signal of a state quantity obtained from a plant that is a control target, calculates a control signal to be given to the control target, and transmits the control signal to the control target. An algorithm for calculating a control signal is mounted on the control device of the plant so that the measurement signal of the state quantity of the plant satisfies the target value.

  As a control algorithm used for plant control, there is a PI (proportional / integral) control algorithm. In PI control, a value obtained by multiplying a deviation between a measurement signal of a plant state quantity and its target value by a proportional gain is added to a value obtained by time-integrating the deviation to derive a control signal to be given to the controlled object.

  Since the control algorithm using PI control can describe the input / output relationship with a block diagram or the like, the causal relationship between the input and the output is easy to understand, and has a lot of application results. However, when the plant is operated under conditions that are not assumed in advance, such as a change in the operation state of the plant or a change in the environment, an operation such as changing the control logic may be required.

  On the other hand, control methods that can adapt to changes in plant operating conditions and environments include control algorithms and control methods that use adaptive control that automatically corrects parameter values and learning algorithms.

  As a method of deriving the control signal of the plant controller using the learning algorithm, the plant characteristics are estimated based on the measured data of the plant and the data constructed based on numerical analysis. A general method is to construct a statistical model and autonomously learn optimal control logic for this statistical model.

  As techniques relating to such a learning control method using a statistical model, Patent Documents 1 and 2 describe a technique for constructing a statistical model using plant measurement data and numerical analysis data.

  In the above known technology, when a control system is introduced into a plant, a statistical model is constructed using only numerical analysis data, and the characteristics of the model are modified using measurement data accumulated after the introduction. It becomes possible to learn flexible control logic that follows characteristic changes.

JP 2007-264796 A JP 2009-128972 A

  When the techniques disclosed in Patent Literature 1 and Patent Literature 2 are applied to a plant control device, the statistical model is adaptively modified in accordance with changes in the plant over time and changes in characteristics. A flexible control system can be constructed.

  Here, generally, in plant control, it takes at least a few minutes to a few tens of minutes for the plant characteristics to stabilize after changing the operating conditions by the control. Expected to get the maximum. Therefore, when the known technique is used, it is preferable that the construction of the statistical model and the learning of the control logic by the learning algorithm are completed within this control cycle.

  Further, in the above-mentioned known technique, when a statistical model is corrected using measurement data, it is desirable that the period required for correction is as short as possible. If time is required for the correction, a statistical model different from the actual plant characteristics may be used. Control based on the learned control logic is executed, and there is a possibility of adversely affecting the plant operation such as loss of operation cost.

  On the other hand, in the known technique, the model correction period for the statistical model at the time of introduction of the control system depends on the accuracy of the initial statistical model, that is, the accuracy of the numerical analysis data. That is, when the error between the numerical analysis data and the measurement data is small, the model can be corrected with a small amount of measurement data, so the model correction period is shortened. However, if the error is large, a large amount of measurement data is required for the model correction, and therefore the model correction period may be long. In addition, if the error is very large, there is a possibility that correction of the model that follows the plant characteristics with desired accuracy may not be executed.

  Furthermore, since the calculation time required for building a statistical model (model building time) generally increases in proportion to the number of model data, it is difficult to build and learn a statistical model within the control cycle due to the accumulation of model data described above. There is a possibility.

  The object of the present invention is that, in the initial statistical model at the time of introduction of the control system, even when the error between the numerical analysis data and the actual plant characteristics is large, the model correction by the measurement data can be completed in a short period of time, or by the data accumulation An object of the present invention is to provide a plant control device or a thermal power plant control device having a function capable of avoiding an increase in statistical model construction time.

  In the plant control device of the present invention, the plant control device includes a control device that takes in a measurement signal that is a state quantity of the plant from the plant and calculates a control signal that controls the plant using the measurement signal. The control device includes a measurement signal database that captures and stores a measurement signal that is a state quantity of the plant, measurement model data converted from the measurement data of the plant stored in the measurement signal database, and a physical model that simulates the plant A model construction database that stores model construction data including at least one of analysis model data that is an output of the plant obtained by numerical analysis, and a control signal to the plant using the model construction data stored in the model construction database The value of the measurement signal that is the state quantity of the plant when A statistical model that simulates the control characteristics of the plant to be estimated, and a method of generating a model input corresponding to the control signal that is given to the plant so that a model output corresponding to the measurement signal achieves a target value using the statistical model An operation method learning unit to learn, a learning information database for storing learning information data related to learning constraint conditions and learning results in the operation method learning unit, a measurement signal of the measurement signal database, and learning information data of the learning information database A control signal generation unit that calculates a control signal transmitted to the plant using a model, and a model adjustment unit that executes selection of analysis model data stored in the model construction database or deletion of measurement model data The operation method learning unit is configured such that the statistical model is modeled by the model correction unit. Characterized by being configured to generate a model output using the modification result in over data.

  The control apparatus for a thermal power plant according to the present invention takes in a measurement signal, which is a state quantity of the plant, from a thermal power plant equipped with a boiler, and calculates a control signal for controlling the thermal power plant using the measurement signal In the control device for a thermal power plant including the device, the measurement signal is at least one of the concentrations of nitrogen oxide, carbon monoxide, and hydrogen sulfide contained in the exhaust gas discharged from the boiler of the thermal power plant. The control signal includes an air flow rate supplied to the boiler of the thermal power plant, an opening degree of an air damper that adjusts the air flow rate, a fuel flow rate supplied to the boiler, and exhausted from the boiler. Including a signal representing at least one of the exhaust gas recirculation flow rates for recirculating the exhaust gas to the boiler, and the control device includes the thermal power plant It is obtained by numerical analysis of a measurement signal database that captures and saves measurement signals that are state quantities, measurement model data converted from plant measurement data stored in the measurement signal database, and a physical model that simulates the thermal power plant. A model construction database that stores model construction data including at least one of analysis model data that is an output of the plant, and a control signal is given to the plant using the model construction data stored in the model construction database. A statistical model for simulating plant control characteristics for estimating the value of a measurement signal that is a state quantity of the plant, and the model output corresponding to the measurement signal using the statistical model is given to the plant so as to achieve a target value Operation method for learning how to generate model inputs corresponding to control signals A learning unit, a learning information database that stores learning information data related to learning constraints and learning results in the operation method learning unit, a measurement signal of the measurement signal database, and a learning information data of the learning information database A control signal generation unit that calculates a control signal transmitted to the model construction unit, and a model adjustment unit that executes selection of analysis model data stored in the model construction database or deletion of measurement model data. The method learning unit is configured such that the statistical model generates a model output using a correction result of model construction data by the model correction unit.

  According to the present invention, in the initial statistical model at the time of introduction of the control system, even when the error between the numerical analysis data and the actual plant characteristics is large, the model correction by the measurement data can be completed in a short period of time, or by the data accumulation It is possible to realize a plant control device and a thermal power plant control device having a function capable of avoiding an increase in statistical model construction time.

The block diagram which shows the structure of the control apparatus of the plant which is 1st Example. The flowchart which shows the operation | movement flow at the time of learning of the operating method in the control apparatus of the plant by 1st Example described in FIG. 1, operation of a plant, and correction of model data. The block diagram which shows the structure of the model data correction part in the control apparatus of the plant by 1st Example described in FIG. The figure which shows the aspect of the data preserve | saved in the model construction data in the control apparatus of the plant by 1st Example described in FIG. The flowchart which shows the operation | movement flow of the model correction part in the control apparatus of the plant by 1st Example described in FIG. The flowchart which shows the operation | movement flow of the analysis model data correction in the control apparatus of the plant by 1st Example described in FIG. The schematic diagram explaining the correction mechanism of the analysis model in the control apparatus of the plant by 1st Example described in FIG. The flowchart which shows the operation | movement flow of the model data deletion in the control apparatus of the plant by 1st Example described in FIG. FIG. 3 is an example of a screen displayed on the image display device when setting model input / output in the plant control apparatus according to the first embodiment shown in FIG. 1; FIG. The plant control apparatus by 1st Example described in FIG. 1 WHEREIN: An example of the screen displayed on an image display apparatus, when setting model correction conditions. The plant control apparatus by 1st Example described in FIG. 1 WHEREIN: An example of the screen displayed on an image display apparatus, when confirming a model correction result. The schematic block diagram which shows the structure of the thermal power plant which is a 2nd Example to which the control apparatus of a plant is applied. The schematic structure figure which shows the structure of the air heater with which the thermal power plant of 2nd Example described in FIG. 12 was equipped. FIG. 13 is an example of a screen displayed on the image display device when setting model input / output in the thermal power plant control apparatus according to the second embodiment shown in FIG. 12. FIG. 13 is an example of a screen displayed on the image display device when setting model correction conditions in the thermal power plant control apparatus according to the second embodiment shown in FIG. 12.

  Next, embodiments of a plant control apparatus and a thermal power plant control apparatus according to the present invention will be described with reference to the drawings.

  In the plant control device having a configuration common to both the plant control device and the thermal power plant control device, the model correction unit constituting the control device is used for model construction using information stored in the model construction database. It is desirable to include at least one of an analysis data correction function for correcting model output value information of data (analysis model data) by numerical analysis included in data and a data deletion function for deleting redundant model construction data. .

  In addition, the information stored in the model construction database includes at least one information among an identifier indicating whether the data is measurement data or numerical analysis data, an input condition in the model input space, an output condition, and a creation time thereof. It is desirable that

  The analysis data correction function selects the analysis model data candidate for correcting the model output value based on the distance between the latest measurement data and the analysis model data, and selects based on the threshold and the neighborhood of N It is desirable to provide at least one of the functions to be performed.

  The analysis data correction function calculates the model output value for the latest measurement data, and the model output value information of the selected analysis model data when the error between the estimated value and the actual measurement data value is less than the threshold. It is desirable to provide at least one of functions for correcting the error based on the error.

  In addition, the data deletion function is based on the distance between the latest measurement data and model construction data, and selects the model construction data whose distance is below the threshold. It is desirable to provide at least one of the functions of further selecting only data whose elapsed time is equal to or greater than the threshold time.

  The control device is connected to an image display device, a function of displaying information stored in the model construction database on the image display device, a function of setting a model correction condition used in the model correction unit via the image display device, It is desirable to have at least one of the functions for displaying the correction result of the model construction data by the model correction unit on the image display device.

  By providing a function for inputting model correction condition settings via an image display device, a plant operator can set appropriate model correction conditions according to the control needs of the plant. Furthermore, by providing a function for displaying the transition of the estimation error due to the model correction on the image display device, the plant operator confirms whether or not the desired model estimation accuracy has been acquired by the model correction. Model modifications can be performed.

  In addition, when the control device is applied to a thermal power plant, the control device for the thermal power plant having a control signal generation unit that derives a control signal to be given to the thermal power plant using a measurement signal acquired from the thermal power plant It becomes.

  These measurement signals include signals representing at least one of the concentrations of nitrogen oxides, carbon monoxide, and hydrogen sulfide contained in the gas discharged from the thermal power plant. The control signal includes a signal for determining at least one of the opening degree of the air damper, the air flow rate, the fuel flow rate, and the exhaust gas recirculation flow rate.

  The control device includes a statistical model for estimating a value of a measurement signal when a control signal is given to the thermal power plant, and an opening degree, an air flow rate, and a fuel of an air damper of the thermal power plant used to construct the statistical model. The control signal so that the model output corresponding to the measurement signal achieves a target value by using a model construction database for storing data including at least one of the flow rate and the exhaust gas recirculation flow rate, and the statistical model. An operation method learning unit that learns a generation method of a model input corresponding to the above, a learning information database that stores information on learning constraint conditions and learning results in the operation method learning unit, and information stored in the model construction database A model correction unit for correcting the model construction data included.

  Further, the control device is connected to the image display device, a function of displaying information stored in the model construction database on the image display device, and a function of setting the model correction condition used in the model correction unit via the image display device. It is desirable to provide at least one of the functions for displaying the correction result of the model construction data by the model correction unit on the image display device.

  In the embodiment in which the control device is applied to the control of the thermal power plant, setting information regarding the burner corresponding to the model input in the thermal power plant and the air amount of the after-air port is input via the image display device.

  Next, a plant control apparatus and a thermal power plant control apparatus according to embodiments of the present invention will be described with reference to the drawings.

  First, a plant control apparatus according to a first embodiment will be described with reference to the drawings.

  FIG. 1 is a system configuration diagram of a plant control apparatus according to the first embodiment. As shown in FIG. 1, a plant 100 to be controlled is controlled by a control device 200.

  Since the control device 200 that controls the plant 100 is connected to the maintenance tool 910, an operator of the plant 100 can connect the external input device 900 and the image display device 920 (for example, a CRT display) connected to the maintenance tool 910. Thus, the control device 200 can be controlled.

  The control device 200 includes a measurement signal conversion unit 300, a numerical analysis unit 400, a statistical model 500, a model correction unit 600, a control signal generation unit 700, and an operation method learning unit 800 as arithmetic devices. ing.

  The control device 200 is provided with a measurement signal database 210, a model construction database 220, a learning information database 230, a control logic database 240, and a control signal database 250 as databases (DB).

  The control device 200 is provided with an external input interface 201 and an external output interface 202 as interfaces with the outside.

  In the control device 200, the measurement signal 1 obtained by measuring the various state quantities of the plant from the plant 100 is taken into the measurement signal database 210 of the control device 200 via the external input interface 201. The control signal generator 700 is configured to output, via the external output interface 202, the control signal 15 for controlling the plant 100 to be controlled, for example, as the control signal 16 for controlling the flow rate of air to be supplied. ing.

  In the control device 200, the measurement signal 2 obtained by measuring the state quantity of the plant 100 captured from the plant 100 via the external input interface 201 is stored in the measurement signal database 210.

  In addition, the control signal 15 generated by the control signal generation unit 700 provided in the control device 200 is stored in the control signal database 250 provided in the control device 200, and the control signal for the plant 100 from the external output interface 202. 16 is output.

  The measurement signal conversion unit 300 provided in the control device 200 converts the measurement signal data 3 stored in the measurement signal database 210 into model construction data 4. The model construction data 4 is stored in the model construction database 220. Here, the model construction data derived from the measurement signal data 3 is referred to as measurement model data. In addition, the operation condition obtained as the control result immediately before included in the measurement signal data 3 is input to the control signal generation unit 700 provided in the control device 200.

  The numerical analysis unit 400 provided in the control device 200 predicts the characteristics of the plant 100 using a physical model that simulates the plant 100. The numerical analysis data 5 obtained by being executed by the numerical analysis unit 400 is stored in the model construction database 220. Here, model construction data derived from the numerical analysis data 5 is referred to as analysis model data.

  The model correction unit 600 provided in the control device 200 has a function of updating model output value information included in the model construction data 7 fetched from the model construction database 220 and a function of deleting unnecessary model construction data. The later model construction data 8 is stored in the model construction database 220.

  The operation method learning unit 800 provided in the control device 200 generates learning data 12 and stores it in the learning information database 230.

  The statistical model 500 provided in the control device 200 has a function of simulating the control characteristics of the plant 100. That is, the control signal 16 is given to the plant 100, and a function equivalent to obtaining the measurement signal 1 for the control result is simulated. For this simulation operation, the statistical model 500 uses the model input 9 received from the operation method learning unit 800 and the model construction data 6 stored in the model construction database 220.

  This model input 9 corresponds to the control signal 16. From the model input 9 and the model construction data 6, the statistical model 500 obtains a model output 10 by simulating a characteristic change due to control of the plant 100 by a statistical method represented by a neural network.

  The model output 10 obtained by the statistical model 500 is a predicted value of the measurement signal 1 of the plant 100. Note that the number of model inputs 9 and model outputs 10 is not limited to one, and a plurality of types can be prepared.

  In the control signal generator 700 provided in the control device 200, the measurement signal 1 is set to a desired value using the learning information data 13 output from the learning information database 230 and the control logic data 14 stored in the control logic database 250. The control signal 15 is generated as follows.

  The control logic database 250 stores a control circuit for calculating the control logic data 14 and control parameters. As a control circuit for calculating the control logic data 14, a publicly known PI (proportional / integral) control can be used.

  The operation method learning unit 800 learns the operation method of the model input 9 using the learning information data 11 including the learning constraint condition and the learning parameter setting condition stored in the learning information database 230. The learning data 12 that is the learning result is stored in the learning information database 230.

  As described above, in the operation of the control device 200, the model correction unit 600 has a mechanism for correcting the information included in the model building data 7 stored in the model building database 220, thereby providing statistical model characteristics and actual plant characteristics. Even when there is a large error, since the analytical model data is appropriately corrected based on the measurement model data, the estimation accuracy of the plant characteristics in the statistical model 500 can be improved in a short period of time.

  In addition, since the measurement model data accumulated after the start of operation has a function of deleting data that has been passed for a certain period from the time of data creation, an increase in statistical model calculation time due to accumulation of model data can be avoided.

  Detailed functions of the statistical model 500, the model correction unit 600, and the operation method learning unit 800 installed in the control device 200 will be described later.

  The learning data 12 stored in the learning information database 230 from the operation method learning unit 800 includes information about model inputs before and after the operation and model outputs obtained as a result of the operation.

  In the learning information database 230, the learning data 12 corresponding to the current driving condition is selected and input to the control signal generation unit 700 as learning information data 13.

  An operator of the plant 100 is provided in the control device 200 by using an external input device 900 including a keyboard 901 and a mouse 902, a maintenance tool 910 capable of transmitting and receiving data to and from the control device 200, and an image display device 920. You can access information stored in various databases.

  In addition, by using these devices, parameter setting values, learning constraints, and obtained learning used in the numerical analysis unit 400, the statistical model 500, the model correction unit 600, and the operation method learning unit 800 of the control device 200 are used. Setting information necessary for checking the result can be input.

  The maintenance tool 910 includes an external input interface 911, a data transmission / reception processing unit 912, and an external output interface 913, and can transmit / receive data to / from the control device 200 via the data transmission / reception processing unit 912.

  The maintenance tool input signal 91 generated by the external input device 900 is taken into the maintenance tool 910 via the external input interface 911. The data transmission / reception processing unit 912 of the maintenance tool 910 acquires the input / output data information 90 from the control device 200 according to the information of the maintenance tool input signal 92.

  Further, in the data transmission / reception processing unit 912, parameter setting values used in the numerical analysis unit 400, the statistical model 500, the model correction unit 600, and the operation method learning unit 800 of the control device 200 according to the information of the maintenance tool input signal 92, the learning method The input / output data information 90 including the constraint conditions and setting information necessary for visual recognition of the obtained learning result is output.

  The data transmission / reception processing unit 912 transmits a maintenance tool output signal 93 obtained as a result of processing the input / output data information 90 to the external output interface 913. The maintenance tool output signal 94 transmitted from the external output interface 913 is displayed on the image display device 920.

  In the control device 200, the measurement signal database 210, the model construction database 220, the learning information database 230, the control logic database 240, and the control signal database 250 are arranged inside the control device 200. Alternatively, a part can be arranged outside the control device 200.

  Further, although the numerical analysis unit 400 is disposed inside the control device 200, it can also be disposed outside the control device 200.

  For example, the numerical analysis unit 400 and the model construction database 220 may be arranged outside the control device 200, and the numerical analysis data 5 may be transmitted to the control device 200 via the Internet.

  FIG. 2 is a flowchart showing a control procedure in the plant control apparatus according to the first embodiment shown in FIG.

  In FIG. 2, the operation method learning unit 800 installed in the plant control apparatus 200 of the first embodiment learns the operation method, the control signal generation unit 700 generates the control signal 16, the control apparatus 200 operates the plant 100, 5 shows a flowchart representing the operation when the model construction data is modified by the model modification unit 600.

  The flowchart shown in FIG. 2 is executed by combining steps 1000, 1100, 1200, 1300, 1400, 1500, 1600, 1700, 1800, 1900, 2000. Hereinafter, each step will be described.

  After the operation of the control device 200 is started, first, in step 1000 for setting the model construction condition / learning condition, various parameters such as the execution condition at the time of model construction, the maximum number of times of learning at the time of learning, the maximum number of operations, and the constraint conditions Set the value.

  Next, in step 1100 of constructing a plant characteristic model, the statistical model 500 of the control device 200 is operated, and a plant characteristic model is constructed using the model construction data 6 stored in the model construction DB 220.

  Next, in step 1200 of learning the operation method, the operation method learning unit 800 is operated to learn the operation method of the model output 9 so that the model output 10 output from the statistical model 500 of the control device 200 becomes a desired value. To do. As the learning means, a known method such as reinforcement learning theory can be used.

  Next, in step 1300 for storing the learning result in the learning information database, the learning result 12 of the operation method is stored in the learning information database 230.

  Next, in step 1400 for generating a control signal, the control signal generator 700 is operated, and the control signal is generated using the control logic data 14 stored in the control logic DB 240 and the learning result 13 stored in the learning information DB 230. 15 is generated.

  Next, in step 1500 of operating the plant, the generated control signal 15 is output as the control signal 16 through the external output interface 202, and the operation is executed so that the plant 100 is in a desired state.

  Next, in step 1600 for storing the measurement data in the DB, the measurement signal data 3 input and stored in the control device 200 after the operation of the plant 100 is converted into model construction data 4 (measurement model data) in the measurement signal conversion unit 300. And stored in the model construction DB 220.

  Next, in step 1700 for setting model data correction conditions, various parameter values relating to execution conditions at the time of model correction are set.

  Next, in step 1800 for correcting the model data, the model data correction unit 600 is operated to update the model output value information included in the model construction data 7 and delete unnecessary data. Detailed functions and operations of the model correction unit 600 will be described later.

  The next step 1900 for determining the validity of the model data correction result is a branch. If the determination criteria are satisfied for the model data correction result, the process proceeds to step 2000; otherwise, the process returns to step 1700. Here, there are two types of determination means: automatic determination based on internal parameters, and manual determination in which the plant operator confirms the model data correction result displayed on the image display device 920 and determines the validity. Either may be used.

  The last step 2000 for determining whether the control is ON or OFF is a branch. Input relating to ON / OFF of control is executed through the external input device 900. If ON, the process returns to Step 1100. If OFF, the process proceeds to a step of ending the control operation of the plant 100 in the series of control apparatuses 200.

  With the above operation, the control of the plant 100 by the control device 200 autonomously learns the model input operation method for obtaining a desired model output based on the model adjustment conditions set by the operator of the plant 100 and the learning conditions. By operating the plant 100 with the control signal generated based on the learning result, the plant 100 can be brought into a desired operation state. Furthermore, by correcting the model construction data using the measurement model data obtained as a result of operating the plant 100, the accuracy of the statistical model can be improved in a short period of time, and the model construction time can be shortened.

  Next, the operation of the model correction unit 600 in the control device 200 will be described in detail with reference to FIG. FIG. 3 is a detailed configuration diagram of the model correction unit 600, and shows in detail the part including the model correction unit 600 and the model construction database 220 in the control device 200 shown in FIG.

  The model adjustment unit 600 includes an analysis data correction function 601 and a data deletion function 602. The analysis data correction function 601 uses the model construction data 7 stored in the model construction database 220 to update the model output value information included in each model construction data, and uses the model construction data 8 after the update as the model construction database 220. Save to

  The data deletion function 602 deletes unnecessary model construction data using the data creation time included in the model construction data 7 and the data input value information. The above operation corresponds to step 1800 for correcting the model data in the flowchart of FIG.

  FIG. 4 shows an example of data stored in the model construction database 220. In the data stored in the model construction database 220 shown in FIG. 4, the data ID 221 is an identification number of each model construction data. In the figure, i is a subscript of data. The data type identifier 222 identifies whether the type of each model construction data is measurement model data or analysis model data. For example, the type of data is identified by using measurement model data when the identifier Si = 0, and analysis model data when the identifier is 1. The model input condition 223 is coordinate information of the data in the model input space, and the model output condition 224 means a plant characteristic value in the input condition of the data.

  The model input is preset as an influencing factor of the plant characteristics, and the statistical model 500 is obtained by continuously interpolating the model construction data shown in FIG. A plant characteristic value (model output 10) for input 9) is estimated.

  The creation time 225 represents the time when each model construction data is created. By comparing the current time with the time information at a certain time after the start of plant operation, the elapsed time after data creation can be acquired.

  In the following, flowcharts (FIG. 5, FIG. 6, FIG. 8) and conceptual diagrams (FIG. 5) of the algorithm for model correction by the analysis data correction function 601 and the data deletion function 602 in the model correction unit 600 provided in the control device 200. This will be described with reference to FIG.

  FIG. 5 is a flowchart showing the algorithm operation of the model correcting unit 600, and corresponds to step 1800 for correcting the model data in the flowchart of FIG.

  The flowchart shown in FIG. 5 executes steps 1810 and 1820 in combination. Hereinafter, each step will be described.

  In step 1810 of correcting the analysis data after the model correction algorithm is started, the analysis data correction function 601 is operated to update the model output value information included in each model construction data. Details of this operation will be described later with reference to FIGS.

  Next, in step 1820 for deleting model data, the data deletion function 602 is operated to delete unnecessary model construction data. Details of this operation will be described later with reference to FIG.

  Next, the detailed operation of the analysis data correction function 601 will be described. The flowchart shown in FIG. 6 is executed by combining steps 1811, 1812, 1813, 1814, 1815, 1816, 1817, and 1819. Hereinafter, each step will be described.

  After starting the analysis data correction algorithm, first, in step 1811 for calculating the distance between the measurement data and the analysis data, all the analysis model data stored in the model construction DB 220 and the measurement obtained as a result of the immediately preceding plant operation. The distance from the measurement model data obtained based on the value is calculated. Here, the distance information is calculated by the Euclidean distance with respect to the input condition of each data.

  The next step 1812 for determining an analysis data selection method to be corrected for model output value information is a branch. There are two types of data selection methods: a method for selecting data based on a distance threshold value and a method for selecting N neighboring data for measurement model data. If the former is selected, the process proceeds to step 1813. If the latter is selected, go to Step 1815. This method can be selected using the image display device 920 and the external input device 900.

  Step 1813 is a branch, and the distance between each analysis model data and the measurement model data calculated in Step 1811 is compared with the distance threshold parameter determined in advance using the image display device 920 and the external input device 900. Then, it is determined whether there is analysis model data whose distance is equal to or less than a threshold value. If one or more analysis model data satisfies the above condition, the process proceeds to step 1814. If no data satisfying the previous period exists, the process proceeds to a step for ending the analysis data correction algorithm.

  In the next step 1814 of selecting analysis data whose distance is less than or equal to the threshold, all analysis model data satisfying the condition that the distance from the measurement model data is less than or equal to the threshold is selected.

  In step 1812, 2. In step 1815, which proceeds when the neighborhood is selected, N pieces of neighborhood data are selected in ascending order of distance from the measurement model data. The number N of neighbors can also be set using the image display device 920 and the external input device 900 in the same manner as the distance threshold. In this way, the data amount can be adjusted by limiting the data amount of the analysis model data by the distance threshold or the number of neighboring data.

  In the next steps 1816, 1817, and 1818, the model output value information is corrected for the analysis model data selected in step 1814 or 1815. Hereinafter, the correction algorithm of the model output value at each step will be described with reference to FIG. FIG. 7 shows the model construction data and statistical model characteristics when the model input and output dimensions are set to 1 for simplification of explanation. That is, the coordinate of the data with respect to the horizontal axis of the graph in the figure corresponds to the model input condition, and the coordinate with respect to the vertical axis corresponds to the model output value information. In FIG. 7, (a) is an operation mechanism of the algorithm in the case of data selection by a threshold, and r in the figure represents a distance threshold. (B) is an operation mechanism in the case of data selection by neighborhood, and represents the case where the number of neighborhoods N = 3. In FIG. 7, the latest measurement model data is expressed as dm, and the analysis model data is expressed as dic (i is a subscript of the data). That is, d1c and d2c are selection data in (a), and d1c, d2c, and d3c are selection data in (b).

  In step 1816, a model estimated value Zj (k) is calculated using the statistical model 500 for the model input condition of the measurement model data dm. Here, j is a subscript of the number of model outputs, k is the number of corrections of the model output value, and is initialized to k = 0 at the first execution of step 1816. Note that FIG. 7 shows only the estimated value when the number of model output dimensions is 1, but actually the estimated values for all outputs are calculated according to the preset model output number. From FIG. 7, the model output value information of the model output j component of the measurement model data dm is expressed as Zjm. Note that dm is not included in the model construction data used in calculating the model estimated value Zj (k).

  The next step 1817 for determining whether the estimation error is less than or equal to the threshold value is a branch. If the absolute value of the estimation error calculated by Zjm−Zj (k) is equal to or smaller than a preset threshold value, the process proceeds to step 1819, and if not, the process proceeds to 1818. If the number of model outputs is 2 or more, the process proceeds to step 1819 only when the estimation error for all outputs is equal to or less than the threshold value.

In the next step 1818 of correcting the analysis data selected based on the estimated value, the model output value information of the selected data is updated using [Equation 1] using the estimation error obtained in step 1817 as a clue. Here, the model output value information of the selected data is expressed as yij (k).
[Equation 1]
yij (k + 1) = yij (k) + α (Zjm−Zj (k))
In [Formula 1], the model output value information of the selected data is corrected by a value obtained by multiplying the estimation error by the correction rate α. In this way, by correcting the analysis model data output information in the vicinity of the measurement data point so as to reduce the estimation error of the measurement data point, estimation of the statistical model 500 using the corrected model data as shown in FIG. The effect of reducing the error is obtained.

  After updating the model output value information for all the selected analysis model data according to the above procedure, the number of corrections k is updated, and the process proceeds to Step 1816. Thereafter, the processing of steps 1816 to 1818 is repeatedly executed until the estimation error of the statistical model satisfies the condition equal to or less than the threshold value.

  Finally, in step 1819 for storing analysis data, the analysis model data corrected in the above-described procedure is stored in the model construction DB 220, and the process proceeds to a step for ending the analysis data correction algorithm.

  As is apparent from the above description, the analysis model correction algorithm is executed each time new measurement data is obtained, so that the plant in the statistical model 500 can be obtained even when the error between the initial statistical model characteristics and the actual plant characteristics is large. The estimation accuracy of characteristics can be improved in a short period of time, and deterioration of control performance can be avoided.

  Next, detailed operation of the data deletion function 602 will be described. The flowchart shown in FIG. 8 is executed by combining steps 1821, 1822, 1823, 1824 and 1825. Hereinafter, each step will be described.

  After starting the data deletion algorithm, first, in step 1821 for selecting the nearest neighbor data of the measurement data, the distance between the latest measurement data and the model construction data including the measurement data collected so far is calculated. Select the data that will be the smallest (the nearest neighbor).

  The next step 1822 for determining the distance of the selected nearest neighbor data is a branch. If the distance is equal to or smaller than a preset threshold value, the process proceeds to step 1823; otherwise, the process proceeds to a step of ending the data deletion algorithm. Here, it is desirable that the threshold value be sufficiently small.

  The next step 1823 for determining the type of the nearest neighbor data is a branch. If the nearest neighbor data is measurement model data, the process proceeds to step 1824. If the nearest neighbor data is analysis model data, the process proceeds to step 1825.

  The next step 1824 for determining the elapsed time since data creation is a branch. The elapsed time from data creation is calculated from the creation time 225 in FIG. 4 and the current time when the latest measurement data is collected. If it is equal to or greater than the preset time threshold value, the process proceeds to step 1825; otherwise, the process proceeds to a step of ending the data deletion algorithm.

  In the last step 1825 of deleting data, the nearest neighbor data of the latest selected measurement data is deleted from the model construction DB 220, and the process proceeds to a step of ending the data deletion algorithm.

  In the above-described algorithm, only when the model construction data nearest to the latest measurement data is sufficiently close to the measurement data, it is considered that the data can be replaced with the latest measurement data, and is deleted from the DB. However, in the case where the nearest data is measurement model data, in order to consider the influence of measurement errors included in the data, the data is excluded from being deleted within a certain period from the data generation. By providing the above deletion algorithm, it is possible to avoid an increase in the number of model construction data due to the accumulation of measurement data and an increase in the calculation time of the statistical model, and it is possible to learn the control logic within a practical time. Above, description of the detailed operation | movement of the model data correction part 600 is complete | finished.

  Next, in the plant control apparatus according to the first embodiment, the maintenance tool output signal 94 transmitted from the external output interface 913 of the maintenance tool 910 capable of transmitting and receiving data to and from the control apparatus 200 is displayed on the image display device 920. The displayed screen will be described with reference to FIGS. 9, 10 and 11. FIG. 9 to 11 are specific examples of screens displayed on the image display device 920. FIG.

  FIG. 9 is an example of a screen displayed on the image display device when setting the model input / output in the plant control apparatus of the first embodiment, and shows a control procedure in the plant control apparatus of the first embodiment. It is an example of the model construction condition setting screen of step 1000 which sets the model construction condition / learning condition in the flowchart of FIG.

  In the model input / output setting screen shown in FIG. 9, an arbitrary item can be selected from the input / output items based on the measurement data information of the plant, and the model input / output of the statistical model 500 in the control device 200 can be set. .

  With the screen shown in FIG. 9 displayed on the image display device 920, the mouse 902 of the external input device 900 is operated to move the focus to the numerical box on the screen, and a numerical value can be input by using the keyboard 901. Further, by operating the mouse 902 and clicking a button on the screen, the button can be selected (pressed). Similarly, a check can be made by operating the mouse 902 and clicking a check box on the screen.

  In the screen shown in FIG. 9, first, in the model input setting, the focus of the mouse 902 is moved to an arbitrary item with respect to the input item displayed in the input item list 3000, and the selection bar 3001 is selected by selecting a button. Can be matched with the input item. Then, by selecting a button 3002, the selected input item can be added to the model input item list 3003.

  Furthermore, the minimum value and the maximum value can be set for the added model input item by shifting the focus to the numerical boxes 3004 and 3005 and inputting the numerical values. When deleting an item from the already added model input item list, the item to be deleted can be deleted from the list by selecting the item to be deleted with the mouse 902 and selecting the button 3006.

  In the screen shown in FIG. 9, next, in the model output setting, the focus of the mouse 902 is moved to an arbitrary item with respect to the output item similarly displayed in the output item list 3007, and a selection bar is selected by selecting a button. 3008 can be matched with the selected output item. Then, by selecting a button 3009, the selected output item can be added to the model output item list 3010.

  When deleting an item from the already added model output item list, the item to be deleted can be deleted from the list by selecting the item to be deleted with the mouse 902 and selecting the button 3011. When the button 3012 is selected after the above model input / output setting is completed, the process proceeds to step 1100 in FIG. 2 for constructing a statistical model.

  FIG. 10 is a screen example displayed on the image display device when setting the model data correction condition in the plant control apparatus according to the first embodiment, and the control procedure in the plant control apparatus according to the first embodiment is shown. It is an example of the model data correction condition setting screen of step 1700 which sets the model data correction condition in the flowchart of FIG. 2 shown. In the model correction condition setting screen shown in FIG. 10, the check box 3100 and 3102 and the numerical value boxes 3101, 3103, 3014, 3105, 3106 and 3107 are moved to select the check box or input a numerical value. 6, the neighborhood data selection method (threshold / neighbor) in the flowchart of FIG. 6, the threshold parameter in the threshold method, the number of neighbors in the neighborhood method, the threshold of the estimation error, the data correction rate α in [Equation 1], and the flowchart in FIG. The neighborhood data distance threshold value and the data elapsed time threshold value can be respectively determined.

  After the above model data correction condition setting is completed, the model data correction can be started by selecting a button 3108.

  FIG. 11 is a screen displayed on the image display device when the model data correction result is confirmed in the plant control apparatus of the first embodiment, and shows a control procedure in the plant control apparatus of the first embodiment. It is an example of the screen used at the time of determination of the model data correction result in step 1900 of the flowchart of FIG.

  In the model adjustment result display screen shown in FIG. 11, a transition of the estimation error with respect to the number of iterations k of the analysis model data correction is displayed in a graph 3201 as the model estimation error display screen 3200 calculated in step 1816 in FIG. 6. .

  Here, the horizontal axis of the graph represents the number of iterations k of the analysis model data correction, and the vertical axis represents the estimation error at each iteration number. On this model estimation error display screen 3200, an estimation error threshold 3203 set on the model correction condition setting screen of FIG. 10 is displayed. Further, by selecting the tab 3202, it is possible to confirm the transition of the estimation error regarding the set arbitrary model output component.

  The operator of the plant can determine whether or not the correction of the model data is appropriately executed while looking at the model correction result displayed on the model estimation error display screen 3200 shown in FIG.

  When the estimated error 3201 at the end of model correction is below the estimated error threshold 3203 on the model estimated error display screen 3200, the model correction is completed by selecting the button 3204 on the assumption that a desired model correction result has been obtained. can do.

  On the other hand, if the model correction result does not satisfy any of the above-mentioned conditions, selecting the button 3205 returns to the model correction condition setting screen of FIG. 10, and the model correction can be re-executed.

  As shown in FIG. 10, a model data correction frame and a data deletion frame are provided separately. This is because the control device 200 having a statistical model such as FIG. 1 learns from measurement model data which is model construction data converted from measurement signals in order to reduce model construction data for constructing a statistical model. A method for correcting model construction data that accepts selection of model construction data by displaying whether to delete data or selecting data from analytical model data that is model construction data obtained by numerical analysis of a physical model simulating a plant This makes it possible to organize data in consideration of different types of model construction data. For example, since it is not necessary to confirm the type of data from a database as shown in FIG. 4, the workability of data selection or deletion is improved. In addition, it is effective to select or delete measurement model data and analysis model data as upstream processing of data selection, and since data organization is performed considering different types of model construction data, model accuracy based on different types of data Can be easily organized.

  In addition, when accepting the selection of model construction data, when accepting parameter input to search for model construction data, the parameter input indicates that data selection from measurement model data or data selection from analysis model data is displayed separately This should be done before. Since the types of data are displayed before being entered with specific parameters to be searched, it is possible to organize the data in consideration of model building data of different types. The parameter input and search method as search conditions for data selection and deletion performed for organizing data need not be limited to the above-described embodiments.

  Note that the measurement model data and the analysis model data can be distinguished from each other by displaying a numerical box for setting detailed parameters as shown in FIG. 10 so that the measurement model data and the analysis model data are separately enclosed. It is also possible to accept selection of measurement model data and analysis model data first without displaying for setting detailed parameters. When the model construction data shown in FIG. 4 is accessed after selecting the measurement model data and the analysis model data as the model construction data selection, the data sort by the data type identifier 222 is performed to display only the selected type of data. May be. If the data in FIG. 4 is in a different database depending on the type, only the selected type of data may be displayed by accessing the database of the selected type of data.

  As described above, the plant control apparatus according to the first embodiment has a function capable of determining whether or not to end the model correction according to the information displayed on the model correction result display screen illustrated in FIG. The model correction can be repeated until the statistical model performance is obtained. As a result, it is possible to construct a robust statistical model that guarantees a certain level of estimation accuracy even when there is a large error between the analysis model data and the actual machine characteristics of the plant.

  In addition, if the calculation error is repeated before the estimation error has converged before the target threshold is reached, the model correction conditions can be quickly reset by checking the screen. The correction can be performed efficiently.

  In the plant control apparatus of the above-described embodiment, the estimation accuracy of the statistical model can be improved by using the model construction data appropriately adjusted by the model correction unit. Further, since redundant data is deleted by the model correction unit, an increase in statistical model calculation time due to data accumulation can be avoided, and model construction and optimal control logic learning can be performed within a control cycle.

  Above, description about the screen displayed on the image display apparatus 920 in the control apparatus of the plant which is 1st Example is complete | finished.

  Next, a control device for a thermal power plant that is a second embodiment in which the control device 200 is applied to a thermal power plant will be described.

  Needless to say, the control device 200 can also be used when controlling a plant other than the thermal power plant.

  FIG. 12 is a schematic diagram illustrating a configuration of a thermal power plant 100a to which the control device 200 is applied. First, the mechanism of power generation by the thermal power plant 100a will be briefly described.

  In FIG. 12, pulverized coal that is fuel obtained by finely pulverizing coal in a mill 110, primary air for conveying pulverized coal, and secondary air for combustion adjustment are supplied to a boiler 101 that constitutes a thermal power plant 100a. A plurality of burners 102 are provided, and pulverized coal supplied through the burners 102 is burned in the boiler 101. The pulverized coal and the primary air are led from the pipe 134 and the secondary air is led from the pipe 141 to the burner 102, respectively.

  Further, the boiler 101 is provided with an after-air port 103 through which air for two-stage combustion is introduced into the boiler 101. The air for two-stage combustion is guided from the pipe 142 to the after air port 103.

  The high-temperature combustion gas generated by burning pulverized coal inside the boiler 101 flows downstream along the path inside the boiler 101 and is supplied to the heat exchanger 106 disposed inside the boiler 101. After generating steam by exchanging heat with the air, it becomes exhaust gas and flows into the air heater 104 installed on the downstream side of the boiler 101. The air supplied to the boiler 101 is heated by this air heater 104 and heated. To do.

  And the exhaust gas which passed this air heater 104 is discharged | emitted from the chimney to air | atmosphere after performing the exhaust gas process which is not shown in figure.

  The feed water circulating through the heat exchanger 106 of the boiler 101 is supplied to the heat exchanger 106 via the feed water pump 105 and is superheated by the combustion gas flowing down the boiler 101 in the heat exchanger 106 to become high-temperature and high-pressure steam. In this embodiment, the number of heat exchangers is one, but a plurality of heat exchangers may be arranged.

  The high-temperature and high-pressure steam generated in the heat exchanger 106 is guided to the steam turbine 108 via the turbine governor 107, and the steam turbine 108 is driven by the energy of the steam to generate power by the generator 109.

  In the thermal power plant 100a of the second embodiment, various measuring devices that detect state quantities indicating the operating state of the thermal power plant are arranged.

  Since the thermal power plant 100a corresponds to the plant 100 in FIG. 1, the measurement signal of the thermal power plant acquired from these measuring instruments is the control device 200 as the measurement signal 1 from the plant 100 as shown in FIG. To the external input interface 201.

  As a measuring instrument, for example, as shown in a thermal power plant 100a in FIG. 12, a temperature measuring instrument 151 that measures the temperature of high-temperature and high-pressure steam supplied from the heat exchanger 106 to the steam turbine 108, and measures the pressure of the steam. A pressure measuring device 152 and a power generation output measuring device 153 that measures the amount of power generated by the generator 9 are shown.

  The feed water generated by cooling the steam by the condenser (not shown) of the steam turbine 108 is supplied to the heat exchanger 106 of the boiler 101 by the feed water pump 105. The flow rate of this feed water is measured by the flow meter 150. It is measured.

In addition, the state quantity relating to the concentration of components (nitrogen oxide (NOx), carbon monoxide (CO), hydrogen sulfide (H ₂ S), etc.) contained in the exhaust gas that is the combustion gas discharged from the boiler 101 The measurement signal is measured by a concentration measuring device 154 provided on the downstream side of the boiler 101.

  That is, in the thermal power plant control apparatus of the second embodiment in which the control device 200 is applied to the thermal power plant 100a, the measurement data items of the thermal power plant 100a measured by the measuring instrument are measured by the respective measuring instruments. The flow rate of fuel supplied to the boiler 101, which is the state quantity of the thermal power plant 100a, the flow rate of air supplied to the boiler 101, the flow rate of feed water supplied to the heat exchanger 106 of the boiler 101, and the heat exchanger 106 of the boiler 101 Steam temperature generated and supplied to the steam turbine 108, feed water pressure supplied to the heat exchanger 106 of the boiler 101, gas temperature of the exhaust gas discharged from the boiler 101, gas concentration of the exhaust gas, and from the boiler 101 An exhaust gas recirculation flow rate for recirculating a part of the exhaust gas discharged to the boiler 101 is included.

  These measurement data items are measurement data items determined by the control signal 15 calculated and output by the control signal generation unit 700 in the control device 200 shown in FIG.

  In general, many measuring instruments other than those shown in FIG. 12 are arranged in the thermal power plant 100a, but the illustration is omitted here.

  Next, a path of air that is introduced into the boiler 101, that is, a path of primary air and secondary air that is introduced into the boiler 101 from the burner 102, and an interior of the boiler 101 from the after air port 103. The air path will be described with reference to FIG.

  In the boiler 101 shown in FIG. 12, the primary air is guided from the fan 120 to the pipe 130, and passes through the air heater 104 and the air heater 104 installed on the downstream side of the boiler 101 and the air heater 104. However, it is branched to the bypass pipe 131 but is joined again as the pipe 133 disposed on the downstream side of the air heater 104 and led to the mill 110 for producing pulverized coal installed on the upstream side of the burner 102. .

  The primary air passing through the air heater 104 is heated by exchanging heat with the combustion gas flowing down the boiler 101. The primary air that bypasses the air heater 104 together with the heated primary air conveys the differential coal crushed in the mill 110 to the burner 102.

  The air introduced from the pipe 140 using the fan 121 is heated in the same manner by the air heater 104 and then branched into a secondary air pipe 141 and an after-air port pipe 142, respectively. To the burner 102 and the after-air port 103.

  In the control apparatus for a thermal power plant according to the second embodiment, as an example of controlling the flow rate of air sent from the fan 121 and introduced into the boiler 101 from the burner 102 and the after air port 103, piping for secondary air 141 and an after-airport piping 142 are provided upstream of an air damper 162 and an air damper 163 as operation end devices, respectively, and the opening degree of these air damper 162 and air damper 163 is adjusted by the control device 200 to be installed inside the boiler 101. The flow rate of the supplied secondary air and after air can be controlled respectively.

  In addition, as an example of controlling the flow rate of air sent from the fan 120 and introduced into the boiler 101 together with the pulverized coal from the burner 102, an air damper serving as an operation end device is connected to the pipe 131 and the pipe 132 immediately before joining the pipe 133. 160 and an air damper 161 are provided, respectively, and the opening degree of the air damper 160 and the air damper 161 is adjusted by the control device 200 so that the flow rate of the air supplied into the boiler 101 can be controlled.

  Since the control device 200 can also control other measurement data items, the installation location of the operation terminal device may be changed according to the control target.

  FIG. 13 is an enlarged view of a piping section related to the air heater 104 installed on the downstream side of the boiler 101 of the thermal power plant 100a shown in FIG.

  As shown in FIG. 13, the air heater 104 is provided with a pipe 130 for supplying air and a pipe 140, and among these, the pipe 140 is disposed through the air heater 104, and the pipe 130 is The pipe 131 and the pipe 132 are branched from the middle, and the pipe 131 is arranged to bypass the air heater 104, and the pipe 132 is arranged to penetrate the air heater 104.

  The pipe 132 passes through the air heater 104 and then becomes a pipe 133 joined with the pipe 131 and is led to the mill 110, and is arranged so as to guide air along with the pulverized coal from the mill 110 to the burner 102 of the boiler 101. It is installed.

  In addition, the pipe 140 passes through the air heater 104 and then branches into a pipe 141 and a pipe 142. Of these, the pipe 141 guides air to the burner 102 of the boiler 101, and the pipe 142 leads to the after-air port 103 of the boiler 101, respectively. It is arranged like this.

  Also, an air damper 160 and an air damper 161 for adjusting the amount of air flowing are installed in the pipe 131 and the pipe 132 immediately before joining the pipe 133, respectively, and the upstream part of the pipe 141 and the pipe 142 flows. An air damper 162 and an air damper 163 for adjusting the amount of air are respectively installed.

  By operating these air dampers 160 to 163, the area through which air passes through the pipes 131, 132, 141, and 142 can be changed, so that the boiler 101 passes through the pipes 131, 132, 141, and 142. The flow rate of air supplied to the interior of each can be individually adjusted.

  The control signal 15 calculated by the control signal generation unit 700 of the control device 200 is output as the control signal 16 for the thermal power plant 100a via the external output interface 202, and installed in the pipes 131, 132, 141, 142 of the boiler 101, respectively. The devices at the operation end such as the air dampers 160, 161, 162, 163 are operated.

  In this embodiment, the devices such as the air dampers 160, 161, 162, and 163 are referred to as operation ends, and the control signal 15 calculated by the control device 200 necessary for operating the devices is transmitted from the control device 200 to the above-described operation end. The output signal commanded to the operation end is called a control signal 16.

  The control signal 16 calculated by the control signal generation unit 700 and output to the operation end includes an air flow rate supplied to the boiler 101 through pipes 131, 132, 141, and 142, and a pipe for supplying air to the boiler 101. 131, 132, 141, 142, air dampers 160 to 163 for adjusting the flow rate of air, the flow rate of pulverized coal supplied to the burner 102 of the boiler 101, and exhaust gas discharged from the boiler 101 For example, an exhaust gas recirculation flow rate for recirculating a part of the exhaust gas to the boiler 101 is included.

Thereafter, the control device of the first embodiment is applied to the thermal power plant 100a to adjust the amount of air supplied to the burner 102 installed in the boiler 101, and the air dampers 160 and 161 installed in the pipes 131 and 132, respectively. The amount of air supplied to the after-air port 103 installed in the boiler 101 is adjusted, and the air dampers 162 and 163 installed in the pipes 141 and 142 are used as operation ends, and CO, NOx, and H in exhaust gas discharged from the boiler 101 _The case where the concentration of ₂ S is the controlled amount will be described.

In this embodiment, the operation amount of the operation end of the boiler 101 (the opening degree of the air dampers 160, 161, 162, and 163) is the exhaust gas discharged from the boiler 101 as the model input of the statistical model 500 constituting the control device 200 The NOx, CO, and H ₂ S concentrations contained in are the model outputs of the statistical model 500, and the minimization of each model output is the learning objective.

  FIG. 14 shows an example of a screen displayed on the image display device 920 when used in the control device 200 of the thermal power plant 100a in the control device for the thermal power plant according to the second embodiment. Model input / output setting of the statistical model 500 constituting the control device 200 displayed on the image display device, corresponding to FIG. 9 showing an example of the screen displayed when setting the model input / output in the plant control device. It is an example of a screen.

  In the model input / output setting screen example shown in FIG. 14, the model of the statistical model 500 in the control device 200 while grasping the positional relationship with respect to each of the burner 102 and the after-air port 103 which are the operation ends of the boiler 101. Input / output can be set.

  Specifically, in the model input setting screen, the installation position of the item selected by the selection bar 3503 with respect to the input item displayed in the model input item list 3502 displayed on the boiler operation end display screen 3500 is set. A symbol on the boiler diagram before the can displayed on the boiler operation end display screen 3500 shown is indicated by a pointer 3501.

  Contrary to this operation, the selection bar is displayed by clicking the mouse 902 of the external input device 900 on the symbol of the specific operation terminal displayed on the boiler operation terminal display screen 3500 and focusing the pointer 3501. It is also possible to move the display position 3503 (select an input item). Then, by selecting a button 3504, the selected input item can be added to the model input item list 3503.

  In FIG. 14, the functions of the numerical value box 3506, the output item selection bar 3509, the screens 3508 and 3511, the buttons 3510, 3512 and 3513, and the button 3507 on the model input setting screen are the same as those in the screen of FIG. It is the same.

  In the air amount control for controlling the air supplied to the boiler 101 by the control device of the thermal power plant of the second embodiment, a priori knowledge exists regarding the adjustment method of the air damper of the specific burner and after-airport, and in many cases Control based on that is executed.

  Therefore, by setting the model input setting in the control device 200 of the present embodiment to the screen configuration as shown in FIG. 14, the plant operator confirms the position of the operation end of the boiler 101 and performs the prior experiment. The model input / output of the statistical model 500 in the control device 200 can be appropriately selected in consideration of the control method based on knowledge.

  In addition, since it is possible to understand the operation end and minimum / maximum values set using the screen shown in FIG. 14 in association with the plant design information, the model input setting can be made more efficient and setting errors can be reduced. Can contribute.

  FIG. 15 shows an example of a screen displayed on the image display device 920 when used in the control device 200 of the thermal power plant 100a in the thermal power plant control apparatus according to the second embodiment. The model of the model data correction unit 500 constituting the control device 200 displayed on the image display device corresponding to FIG. 10 showing an example of the screen displayed when setting the model data correction condition in the plant control device. It is an example of a screen of data correction condition setting.

  In the model data correction condition setting screen example shown in FIG. 15, a check box 3606 for associating the model output item display list 3605 with the measurement error influence is displayed on the estimation error threshold setting screen which is the model data correction condition. In the measurement data of the thermal power plant, there are data items such as the CO concentration where the temporal fluctuation of the measurement value is large even during steady operation. When such a data item is set as a model output, there is a possibility that the measurement error of the collected measurement data becomes large. As a result, the data variation becomes large, so that the data correction that satisfies the estimated error threshold value or less cannot be executed in the model correction algorithm shown in the first embodiment, and the algorithm may fall into a loop. Therefore, by associating the influence of the measurement error via the check box 3606 with such a measurement data item having a large measurement error, the threshold set in the numerical value box 3604 is determined when the estimated error threshold is determined in the step 1817. Add measurement error. As a result, even when the influence of the measurement error is large, the algorithm does not fall into a loop, and a desirable data correction result is always obtained.

  In FIG. 15, the functions of check boxes 3600 and 3602, numerical value boxes 3601, 3603, 3607, 3608 and 3609 and buttons 3610 and 3611 are the same as those in the screen of FIG.

As described above, when the plant control device 200 is applied to a thermal power plant, NOx, CO, and H discharged from the thermal power plant are learned by learning an operation method that satisfies the requirements for environmental regulations and operation costs. it is possible to achieve the target value of ₂ S concentration.

  According to the present embodiment, even when the error between the analysis model data used when constructing the statistical model and the plant characteristic is large, the model output value information of the analysis model data is corrected based on the collected measurement model data information. The accuracy of the statistical model can be improved in a short time. In addition, even when measurement model data is accumulated redundantly, by appropriately deleting the data, an increase in the statistical model construction time is avoided, and a control device for a thermal power plant that completes learning within a control cycle is realized. be able to.

  The control device of each embodiment described above can be implemented by a computer or the like having a memory or a CPU, and the processing means as a function of the device is a program module, and the module is read and executed by the computer. Each function can be implemented with.

  Each function can be implemented by causing a computer to read a recording medium on which a program module is recorded.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

  The present invention can be applied to a plant control device and a thermal power plant control device.

DESCRIPTION OF SYMBOLS 1 Measurement signal 16 Control signal 90 Input / output data information 100 Plant 100a Thermal power plant 101 Boiler 102 Burner 103 After air ports 130-133, 140-142 Piping 160-163 Air damper 200 Control device 201 External input interface 202 External output interface 210 Measurement signal Database 220 Model construction database 230 Learning information database 240 Control logic database 250 Control signal database 300 Measurement signal conversion unit 400 Numerical analysis unit 500 Statistical model 600 Model correction unit 700 Control signal generation unit 800 Operation method learning unit 900 External input device 901 Keyboard 902 Mouse 910 Maintenance tool 911 External input interface 912 Data transmission / reception processing unit 913 External output interface 920 image display device

Claims (12)

In a plant control device including a control device that takes in a measurement signal that is a state quantity of the plant from a plant and calculates a control signal for controlling the plant using the measurement signal,
The control device includes a measurement signal database that captures and stores a measurement signal that is a state quantity of the plant, measurement model data converted from the measurement data of the plant stored in the measurement signal database, and a physical model that simulates the plant A model construction database that stores model construction data including at least one of analysis model data that is an output of the plant obtained by numerical analysis, and a control signal to the plant using the model construction data stored in the model construction database A statistical model that simulates the control characteristics of the plant that estimates the value of the measurement signal, which is the state quantity of the plant, and the model output corresponding to the measurement signal using the statistical model achieves the target value. A model input generation method corresponding to the control signal given to the plant An operation method learning unit to learn, a learning information database for storing learning information data related to learning constraint conditions and learning results in the operation method learning unit, a measurement signal of the measurement signal database, and learning information data of the learning information database A control signal generation unit that calculates a control signal transmitted to the plant using
A model correction unit that selects and corrects a target to be corrected among analysis model data stored in the model construction database,
The plant control apparatus, wherein the operation method learning unit is configured so that the statistical model generates a model output using a correction result of model construction data by the model correction unit. The plant control apparatus according to claim 1,
The model building data stored in the model building database includes at least one of an identifier indicating whether each data is data based on measurement information or data based on numerical analysis, an input condition, an output condition, and a creation time thereof. A plant control device. The plant control apparatus according to claim 1,
The model correction unit that performs selection of the analysis model data has a function of selecting analysis model data candidates for correcting model output values based on a threshold based on a distance between the latest measurement data and analysis model data; A plant control apparatus comprising at least one of functions selected based on the presence of a predetermined number of neighbors from the latest measurement data. The plant control apparatus according to claim 1,
The model correction unit for correcting the analysis model data has a function of calculating a model output value for the latest measurement data, and an error between the calculated model output value and the actual measurement data value is not less than a threshold value. A plant control apparatus comprising a function of correcting model output value information of selected analysis model data based on an error thereof. The plant control apparatus according to claim 1,
The model correction unit that executes the deletion of the measurement model data has a function of selecting model construction data whose distance is equal to or less than a threshold based on the distance between the latest measurement data and model construction data, and the selection data is measurement data. In this case, the plant control apparatus is provided with a function of selecting data whose elapsed time after data creation is equal to or greater than a threshold time. The plant control apparatus according to claim 1,
The control device is connected to an image display device, and the image display function displays a function for displaying information stored in the model construction database on the image display device, and a model correction condition used in the model correction unit. A plant control apparatus comprising at least one of a function set via an apparatus and a function of displaying a correction result of model construction data by the model correction unit on an image display device. In a control device for a thermal power plant that includes a control device that takes in a measurement signal that is a state quantity of the plant from a thermal power plant including a boiler and calculates a control signal that controls the thermal power plant using the measurement signal ,
The measurement signal includes a state quantity signal representing at least one of the concentrations of nitrogen oxide, carbon monoxide, and hydrogen sulfide contained in the exhaust gas discharged from the boiler of the thermal power plant,
The control signal is used to recirculate the flow rate of air supplied to the boiler of the thermal power plant, the opening of an air damper that adjusts the air flow rate, the flow rate of fuel supplied to the boiler, and the exhaust gas discharged from the boiler. A signal representing at least one of the exhaust gas recirculation flow rates,
The control apparatus simulates the measurement signal database that captures and stores the measurement signal that is the state quantity of the thermal power plant, the measurement model data converted from the measurement data of the plant stored in the measurement signal database, and the thermal power plant A model construction database for storing model construction data including at least one of analysis model data that is an output of the plant obtained by numerical analysis of the physical model, and the model construction data stored in the model construction database A statistical model that simulates the control characteristics of the plant that estimates the value of the measurement signal that is the state quantity of the plant when the control signal is given to the plant, and the model output corresponding to the measurement signal using the statistical model is the target value A model corresponding to the control signal given to the plant to achieve An operation method learning unit that learns a force generation method, a learning information database that stores learning information data related to learning constraint conditions and learning results in the operation method learning unit, a measurement signal in the measurement signal database, and the learning information A control signal generator for calculating a control signal transmitted to the plant using the learning information data of the database;
A model correction unit that selects and corrects a target to be corrected among analysis model data stored in the model construction database,
The control method for a thermal power plant, wherein the operation method learning unit is configured so that the statistical model generates a model output using a correction result of model construction data by the model correction unit. The control device for a thermal power plant according to claim 7,
The model building data stored in the model building database includes at least one of an identifier indicating whether each data is data based on measurement information or data based on numerical analysis, an input condition, an output condition, and a creation time thereof. The control apparatus of the thermal power plant characterized by the above-mentioned. The control device for a thermal power plant according to claim 7,
The model correction unit that performs selection of the analysis model data has a function of selecting analysis model data candidates for correcting model output values based on a threshold based on a distance between the latest measurement data and analysis model data; A control apparatus for a thermal power plant, comprising at least one of functions selected based on the presence of a predetermined number of neighbors from the latest measurement data. The control device for a thermal power plant according to claim 7,
The model correction unit for correcting the analysis model data has a function of calculating a model output value for the latest measurement data, and an error between the calculated model output value and the actual measurement data value is not less than a threshold value. An apparatus for controlling a thermal power plant comprising a function of correcting model output value information of selected analysis model data based on the error. The control device for a thermal power plant according to claim 7,
The model correction unit that executes the deletion of the measurement model data has a function of selecting model construction data whose distance is equal to or less than a threshold based on the distance between the latest measurement data and model construction data, and the selection data is measurement data. In this case, the thermal power plant control apparatus is provided with a function of selecting data whose elapsed time after data creation is equal to or greater than a threshold time. The control device for a thermal power plant according to claim 7,
The control device is connected to an image display device, and the image display function displays a function for displaying information stored in the model construction database on the image display device, and a model correction condition used in the model correction unit. A plant control apparatus comprising at least one of a function set via an apparatus and a function of displaying a correction result of model construction data by the model correction unit on an image display device.

Images