KR20240024548A - Simulation apparatus of power plant - Google Patents

Abstract

The present invention discloses a simulation device for a power plant. The simulation device of the power plant of the present invention includes: a database; display; Memory to store the simulator program; and a processor that executes a simulator program, where the processor executes the simulator program and transmits and stores simulation data generated by simulating according to the process model to a database, and links with a big data platform that collects and manages actual operation data. In connection with the database, actual operation data is acquired from the big data platform, transmitted to the database and stored, and the fidelity is determined by comparing and analyzing the actual operation data and simulation data stored in the database, and the key parameters of the process model are adjusted. do.

Description

Simulation device of power plant {SIMULATION APPARATUS OF POWER PLANT}

The present invention relates to a simulation device for a power plant. More specifically, it not only determines the fidelity of the simulator in connection with a platform that provides actual operation data of the power plant, but also automatically calculates the parameters of the process model compared to the actual system and compares the results. It is about a simulation device for a power plant that can improve fidelity by adjusting it.

In general, a power plant simulator is a technology that can simulate the actual operation of a power plant by mathematically simulating the field equipment of the power plant. It is used for operation training using power plant design or operation data, or is used for control prior to the application of a digital control system. It is used for control verification such as logic testing, interlock testing, HMI (Human Machine Interface) testing, and control logic improvement testing. These simulators are not only used for driving training or control verification by being applied to nuclear power plants, thermal power plants, etc., but are also recently being used for engineering purposes such as engineering pre-verification and design use.

However, the existing power plant simulation device is constructed using power plant design data at the time of construction, and does not sufficiently reflect power plant modifications, deterioration, control logic modifications, and upgrades in a timely manner, so there are some differences from the actual operating system. There is a situation. For this reason, errors occur between the data of the actual operating system and the simulator, which limits its use for commissioning personnel training and control verification.

In the acceptance test that determines the fidelity of the simulator, the accuracy of the values and reference data calculated by the simulator is calculated to check the fidelity, which can be considered the performance standard of the simulator.

The background technology of the present invention is disclosed in Korean Patent Publication No. 10-2009-0032570 (published on April 1, 2009, monitoring system using a real-time simulator).

Currently, the fidelity of the simulator is judged by users directly comparing/analyzing data, and if fidelity standards are not met, parameters of the process model (pump, fan, valve, etc.) are adjusted through manual intervention by experts.

However, since the simulator consists of numerous process models, it takes a lot of time, effort, and cost to adjust parameters through manual intervention by experts. In addition, a lot of information is needed, such as the replacement/repair history of process equipment, and if it is difficult to obtain information, it can be obtained by comparing/analyzing the data one by one. Fidelity can be improved through trial and error, so fidelity can be checked immediately when necessary. There are problems that are difficult to improve.

The present invention was created to improve the problems described above, and the purpose of the present invention according to one aspect is not only to determine the fidelity of the simulator in connection with a platform that provides actual operation data of the power plant, but also to determine the fidelity of the process model compared to the actual system. It provides a simulation device for power plants that can improve fidelity by automatically calculating parameters and comparing and adjusting the results.

A simulation device for a power plant according to one aspect of the present invention includes a database; display; Memory to store the simulator program; and a processor that executes a simulator program, where the processor executes the simulator program and transmits and stores simulation data generated by simulating according to the process model to a database, and links with a big data platform that collects and manages actual operation data. In connection with the database, actual operation data is acquired from the big data platform, transmitted to the database and stored, and the fidelity is determined by comparing and analyzing the actual operation data and simulation data stored in the database, and the key parameters of the process model are adjusted. do.

In the present invention, the processor is linked to a big data platform and database through OPC (OLE for Process Control) communication or API (Application Programming Interface) to transmit, store, and read actual operation data and simulation data.

In the present invention, the processor acquires actual operational data using a big data platform and OPC communication, and stores it in a database using OPC communication.

In the present invention, the processor reads actual operation data and simulation data at a desired time and situation from a database, checks the average value, and checks fidelity.

In the present invention, the processor calculates key parameters based on existing parameter values of the process model and actual operation data, provides a characteristic graph, and updates key parameter values of the process model.

In the present invention, the processor uses nonlinear least squares (NLS), genetic algorithm (GA), particle swarm-based stochastic optimization (PSO: Particle Swarm Optimization), and differential evolution algorithm (DE). It is characterized by estimating and calculating key parameter values by optimizing the specified objective function using one of the techniques.

The power plant simulation device according to one aspect of the present invention not only determines the fidelity of the simulator in connection with a platform that provides actual operation data of the power plant, but also automatically calculates the parameters of the process model compared to the actual system, compares the results, and adjusts them. , not only can fidelity be improved through more accurate results based on data without trial and error, but model error improvement can be performed easily and quickly.

In addition, according to the present invention, by implementing a high-fidelity simulator, not only can the efficiency of data analysis work be improved and the development and maintenance costs of the fault diagnosis system be reduced, but optimal operation, maintenance, and management of the power plant can be performed more efficiently. In addition, it can be used for fault diagnosis/life prediction verification, driving training, control verification, and engineering in conjunction with digital twin applications such as performance evaluation/damage evaluation.

1 is a block diagram showing a simulation device for a power plant according to an embodiment of the present invention.
Figure 2 is a block diagram showing the configuration modules of a simulator program in a simulation device for a power plant according to an embodiment of the present invention.
Figure 3 is an exemplary diagram showing a screen implementing the model error minimization module of the simulator program in a simulation device for a power plant according to an embodiment of the present invention.
Figure 4 is an example diagram for explaining the processing process by the model error minimization module in the power plant simulation device according to an embodiment of the present invention.
Figure 5 is an example result showing the change in model error by the model error minimization module in the simulation device of a power plant according to an embodiment of the present invention.
Figure 6 is an example file defining a process model for driving a model error minimization module in a simulation device for a power plant according to an embodiment of the present invention.
Figure 7 is an example diagram showing a dual graph function linking actual operation data and simulation data in a simulation device for a power plant according to an embodiment of the present invention.

Hereinafter, a simulation device for a power plant according to the present invention will be described with reference to the attached drawings. In this process, the thickness of lines or sizes of components shown in the drawing may be exaggerated for clarity and convenience of explanation. In addition, the terms described below are terms defined in consideration of functions in the present invention, and may vary depending on the intention or custom of the user or operator. Therefore, definitions of these terms should be made based on the content throughout this specification.

Figure 1 is a block diagram showing a simulation device for a power plant according to an embodiment of the present invention, and Figure 2 is a block diagram showing a configuration module of a simulator program in a simulation device for a power plant according to an embodiment of the present invention. Figure 3 is an exemplary diagram showing a screen implementing the model error minimization module of the simulator program in the simulation device of a power plant according to an embodiment of the present invention, and Figure 4 is a view showing a model in the simulation device of a power plant according to an embodiment of the present invention. It is an example diagram for explaining the processing process by the error minimization module, and Figure 5 is an example result showing the change in model error by the model error minimization module in the simulation device of the power plant according to an embodiment of the present invention, and Figure 6 is this It is an example file defining a process model for driving a model error minimization module in a simulation device of a power plant according to an embodiment of the invention, and Figure 7 shows actual operation data and simulation data in a simulation device of a power plant according to an embodiment of the present invention. This is an example diagram showing the dual graph function linking .

As shown in FIG. 1, a simulation device for a power plant according to an embodiment of the present invention includes a database 50, a display 40, and a memory 20 that stores a simulator program; and a processor 30 that executes a simulator program.

Here, the database 50 may serve to store and manage both actual operation data and simulation data in the field.

At this time, the database 50 may receive, store, and provide data using OPC (OLE for Process Control) communication or API (Application Programming Interface).

In other words, when receiving actual operational data and simulation data, it operates as an OPC client to receive the data, and when providing it, it operates as an API server to provide the necessary data.

The display 40 may provide simulation results as well as display implementation results of a model error minimization module to improve fidelity.

As shown in FIG. 2, the simulator program running on the processor 30 may be configured and executed by including an operating data acquisition module 310, a model error minimization module 320, and a simulator module 330.

Therefore, when the processor 30 executes the simulator program, it transmits and stores simulation data generated by simulating according to the process model through the simulator module 330 in connection with the database 50.

In addition, the processor 30 links with the big data platform 10, which collects and manages actual operational data through the operational data acquisition module 310, and connects with the database 50 to collect actual operational data from the big data platform 10. It is acquired, transmitted to the database 50, and stored.

Here, the processor 30 can acquire actual operational data using the big data platform 10 and OPC communication through the operational data acquisition module 310, and store it in the database 50 using OPC communication.

In this way, when the actual operation data and simulation data are stored in the database 50, the processor 30 reads the actual operation data and simulation data stored in the database 50 through the model error minimization module 320 and compares and analyzes them to ensure fidelity. Determine and adjust key parameters of the process model.

Here, the processor 30 can read actual operation data and simulation data at a desired time and situation from the database 50 through the model error minimization module 320, check the average value, and check fidelity. In addition, the processor 30 can calculate key parameters based on existing parameter values and actual operating data of the process model, provide a characteristic graph, and update key parameter values of the process model.

For example, Figure 3 is an example diagram showing a screen implementing the model error minimization module of a simulator program in a simulation device of a power plant.

Here, the process model refers to devices such as pumps, fans, and valves in a large category, and the device name refers to a specific process of the power plant. There are numerous devices such as pumps, fans, and valves in the power plant process, and each process device has different tags and parameters such as inlet/outlet pressure and temperature, so you can select the device.

Additionally, basic data can be read from the database 50, including actual operational data and simulation data.

That is, the processor 30 can execute the model error minimization module 320 to load actual operation data and simulation data at a desired time and situation from the database 50 in a file written in a designated format.

In addition, the processor 30 may read the history times of actual operation data and simulation data differently to improve fidelity for each situation according to the user's intention.

In particular, the processor 30 can run the simulator module 330 to perform simulation while changing conditions as many times as desired according to the user's intention and purpose.

In this way, when actual operation data and simulation data are linked, the average value can be basically confirmed.

Additionally, in fidelity confirmation, fidelity calculations can be performed using linked actual operation data and simulation data, and whether fidelity is satisfied can be checked.

Here, when performing error calculations to determine fidelity, accuracy can be calculated using Equation 1. At this time, the reference data refers to actual operational data, and the measurement span can be calculated as Span-Zero using Zero and Span in the file defined for driving the model error minimization module 320.

Model fidelity in steady-state operation is judged based on model accuracy with reference data from the power plant and simulated calculations at 100% load and intermediate loads above 25% for which reference data from the power plant is available. can do.

At this time, the model error calculation formula for calculating accuracy is the same as Equation 1, and the standards for model fidelity in thermal balance calculation can be applied to the standards of ANSI/ANS-3,5-1998 and ANSI/ISA-77.20-1993. When referring to this, major parameters require a model error range within 1 to 2 [%], and other parameters require a model error range within 10 [%].

In this embodiment, the fidelity determination criteria can be added and supplemented according to the user's perspective, and satisfaction is determined based on the fidelity criteria (%). If the calculation result is smaller than the fidelity standard (%), it is indicated as O because it is satisfactory, and if the calculation result is greater than the fidelity standard (%), it is indicated as X because it is unsatisfactory.

After checking the fidelity in this way, key parameter calculations of the process model at City X can be performed.

The processor 30 calculates key parameters to reduce model errors compared to the real system by executing the model error minimization module 320, and improves fidelity by comparing/analyzing and applying existing/new result values and graphs of key parameters. can do.

In addition, the processor 30 retrieves existing parameters of the process model applied to the simulator module 330 as well as actual operating data to calculate key parameters through the model error minimization module 320.

Here, the reason why the parameter values applied to the simulator module 330 are needed is because they are performed and applied to key parameters rather than calculating and applying all parameters of the process model.

For example, since the parameter corresponding to the number of pumps does not change, there is no need to select and calculate it as a key parameter. This can save time and reduce computing load.

The processor 30 uses Nonlinear Least Squares (NLS), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Differential Evolution Algorithm (PSO) to calculate key parameters. Key parameter values can be estimated by optimizing the specified objective function using any one of DE: Differential Evolution) techniques.

For example, as shown in FIG. 4, the operation of the model error minimization module when applied to the pump process is explained as follows.

Here, the Condensate Boosting Pump (CBP) process device of the pump process model is selected, and actual operation data and simulation data appropriate for the situation are loaded to check whether the average value and fidelity are satisfied. At this time, if the simulator CBP discharge pressure value is different from the actual power plant CBP discharge pressure value and is judged to be unsatisfactory, the discharge pressure change based on the formula is analyzed and pre-specified key parameters are calculated.

At this time, if the slope of the pump characteristic curve is selected as a key parameter, the parameter value can be estimated using any one of the following techniques: nonlinear least squares, genetic algorithm, particle swarm-based stochastic optimization technique, and differential evolution algorithm.

Here, the objective function can be defined as in Equation 2, and can be changed as much as desired depending on the user's purpose.

Here, P _wl,realplant is the actual operating data for the outlet pressure of the pump, and P _wl,simulator is the simulation data for the outlet pressure of the pump.

This will be explained in more detail along with the implementation screen of the model error minimization module 320 shown in FIG. 3 as follows.

First, when PUMP is selected in the process model and CBP is selected in the device name, the processor 30 loads the corresponding actual operation data and simulation data from the database 50. Afterwards, the average value between actual operation data and simulation data is compared, and satisfaction is determined by checking fidelity. Here, if Satisfaction is marked with an

In the step of checking parameters, the current values of parameters applied to the existing simulator module are checked, and when the user's desired algorithm is selected, calculations are performed according to the selected algorithm.

Here, when calculating new parameters, the objective function that is suitable for the user's intention is optimized as shown in Equation 2. By comparing the current value on the left side of the parameter section with the calculated new value on the right side, you can check whether a reasonable result has been obtained by providing a characteristic curve graph at the bottom.

Based on these results, if application is selected if it meets the user's purpose, the processor 30 performs simulation by applying new parameter values to the simulator module 330.

In this way, when simulation is performed by applying new parameter values calculated through model error improvement, the effect of improving fidelity can be obtained as shown in FIG. 5.

In other words, when the CBP discharge pressure value of the actual power plant was 24.8 kg/cm2, the simulator discharge pressure value before improvement was 31.4 kg/cm2, but after model error improvement, the simulator discharge pressure value was 25.7 kg/cm2, which is the simulator model error. It can be seen that has improved from 13.2% to 1.8%.

Meanwhile, when the processor 30 loads simulation data through the model error minimization module 320, it can load and apply it from a file defining the process model and device name, as shown in FIG. 6.

For example, the simulator process model refers to processes such as pumps, fans, and valves in a large category, and the device name refers to devices of a specific process.

Specifically, BFP-A is a boiler feed water pump (BFP). In the classification, REAL refers to actual field data, SIM refers to simulation data from the simulator, and PARAMETER refers to parameters of the process model.

OPC Update is related to communication. R stands for read, W stands for write, and RW stands for read/write. In the case of variable names, P stands for pressure, T stands for temperature, Speed stands for speed, we stands for inlet, and wl stands for outlet. The tag name contains tags corresponding to the parameters of the device, such as inlet/outlet pressure, speed, etc. for each device. Since the units used in the field and the units used in the simulator may be different, information on the units was indicated, and Zero and Span were entered according to the physical quantity so that they could be used for fidelity calculations.

In addition, the physical quantity to determine fidelity was specified through fidelity calculation, and the fidelity standard was implemented so that it could be changed to suit the user's purpose. Since it is difficult for the user to input them individually according to the situation, a list of key devices is defined in advance and can be implemented so that the user can add and utilize them when necessary.

Meanwhile, the processor 30 may provide a dual graph function as shown in FIG. 7 to link actual operation data and simulation data to check and extract data. In this way, the upper graph is actual operational data, and the lower graph can simultaneously provide simulation data for comparison/analysis, and data can be loaded and saved for the tag at the time and state desired by the user.

In addition, the processor 30 stores records before and after applying key parameters, so that changes in process equipment can be checked, and when an error occurs, the existing applied parameters can be changed, and a report function that organizes data on key parameters can be provided. You can.

As described above, according to the power plant simulation device according to the embodiment of the present invention, not only is the fidelity of the simulator determined in connection with a platform that provides actual operation data of the power plant, but it also automatically calculates the parameters of the process model compared to the actual system. By comparing and adjusting the results, not only can fidelity be improved through more accurate results based on data without trial and error, but model error can be improved easily and quickly.

Implementations described herein may be implemented, for example, as a method or process, device, software program, data stream, or signal. Although discussed only in the context of a single form of implementation (eg, only as a method), implementations of the features discussed may also be implemented in other forms (eg, devices or programs). The device may be implemented with appropriate hardware, software, firmware, etc. The method may be implemented in a device such as a processor, which generally refers to a processing device that includes a computer, microprocessor, integrated circuit, or programmable logic device. Processors also include communication devices such as computers, cell phones, portable/personal digital assistants (“PDAs”) and other devices that facilitate communication of information between end-users.

The present invention has been described with reference to the embodiments shown in the drawings, but these are merely exemplary, and those skilled in the art will recognize that various modifications and other equivalent embodiments are possible therefrom. You will understand.

Therefore, the true technical protection scope of the present invention should be determined by the claims below.

10: Big data platform 20: Memory
30: Processor 40: Display
50: Database 310: Operational data acquisition module
320: model error minimization module 330: simulator module

Claims (6)

database;
display;
Memory to store simulator programs; and
Including a processor that executes the simulator program,
The processor executes the simulator program and transmits and stores simulation data generated by simulating according to the process model to the database,
Linking with a big data platform that collects and manages actual operational data, and linking with the database, acquires the actual operational data from the big data platform, transmits it to the database, and stores it,
A simulation device for a power plant, characterized in that fidelity is determined by comparing and analyzing the actual operation data and the simulation data stored in the database, and adjusting key parameters of the process model.
The method of claim 1, wherein the processor is linked to the big data platform and the database through OLE for Process Control (OPC) communication or an Application Programming Interface (API) to transmit and store the actual operation data and the simulation data. A simulation device for a power plant characterized by reading.
The simulation device of a power plant according to claim 1, wherein the processor acquires the actual operation data using the big data platform and OPC communication, and stores it in the database using OPC communication.
The simulation device of a power plant according to claim 1, wherein the processor reads the actual operation data and the simulation data at a desired time and situation from the database, checks the average value, and checks fidelity.
The power plant of claim 1, wherein the processor calculates key parameters based on existing parameter values of the process model and the actual operation data, provides a characteristic graph, and updates key parameter values of the process model. simulation device.
The method of claim 1, wherein the processor uses nonlinear least squares (NLS), genetic algorithm (GA), particle swarm-based stochastic optimization (PSO), and differential evolution algorithm (DE). A simulation device for a power plant characterized by estimating and calculating key parameter values by optimizing a specified objective function using any one of the following techniques: Differential Evolution).

Images