Auxiliary controller and control method using ai and FPGA

Abstract

The present invention relates to an auxiliary controller using AI and FPGA and a control method thereof, the auxiliary controller comprising: a first signal distributor for distributing control variables; a second signal distributor for distributing control signals output from a controller to a control target; an input module for receiving input signals of the first signal distributor, the second signal distributor, and a user terminal, and outputting the signals; an AI for receiving the output signals of the input module and learning the control variables and the output signals of the controller via machine learning; and an FPGA for performing learning, monitoring, and operation determination according to a predetermined flowchart to assist operation of the controller. According to the present embodiment, control variables, such as pressure, temperature, flow rate, water level, and output, may be received using signal distributors and an input module, the control variables and output signals may be learned via machine learning, an FPGA may be designed when a specific accuracy is exceeded, and learning, monitoring, and operation determination may be performed according to a predetermined flowchart to assist operation.

Description

Auxiliary controller using AI and FPGA and its control method

[0001] The present invention relates to an auxiliary controller using AI and an FPGA and a control method thereof, and more particularly, to an auxiliary controller using AI and an FPGA and a control method thereof, which receives control variables such as pressure, temperature, flow rate, water level, and output using a signal distributor and an input module, learns the control variables and output signals through machine learning, designs an FPGA when a certain accuracy is exceeded, and assists driving by judging learning, monitoring, and driving according to a predetermined flowchart.

[0002] In general, in the rapidly changing international energy environment, such as resource depletion, the enforcement of the Kyoto Protocol, and the surge in oil prices, the relative economic feasibility of new and renewable energy is becoming more advantageous, and the energy market based on new energy technologies such as hydrogen, fuel cells, and solar cells is expected to rapidly emerge as a huge industry that surpasses IT and BT. Accordingly, we have entered a global development competition system for the future new energy industry, and we need to prepare nationally to take the lead in the global market.

[0003] Nuclear power plants have demonstrated superior operating performance compared to hydroelectric or thermal power plants in terms of economic feasibility, safety, and environmental conservation, and have established themselves as an important means of power generation.

[0004] Nuclear power generation uses the energy generated during the nuclear fission process of fissile material to produce electricity. However, if an accident occurs in which radioactive materials generated during this process leak abnormally, there is a risk that it could develop into a large-scale disaster. Therefore, the safety of nuclear power plants has always been treated as a top priority.

[0005] Accordingly, although existing nuclear power plants are evaluated as having reasonable safety, the development of next-generation reactors with dramatically improved safety is actively underway worldwide.

[0006] Meanwhile, the power plant's controller controls the equipment using control variables such as pressure, temperature, flow rate, water level, and output.

[0007] In the case of critical equipment at power plants, hardware redundancy is used to assist, but the same failure can occur at similar times due to environmental factors such as dust or high temperatures, or software errors.

[0008] Since these power plants operate for more than 40 years, it may be difficult to obtain controllers if there is a problem with the supplier, such as a controller discontinuation.

[0009] Furthermore, while they can understand the blueprints, they cannot design as needed. Furthermore, designing power plant controllers or FPGAs requires specialized knowledge, making them impossible to produce on-site, and requesting small-scale production from outside sources is also difficult.

[0010]

[0011] The technical problem to be achieved by the present invention is to improve the conventional problems, and to provide an auxiliary controller using AI and FPGA and a control method thereof, which receives control variables such as pressure, temperature, flow rate, water level, and output using a signal distributor and an input module, learns the control variables and output signals through machine learning, designs an FPGA when a certain accuracy is exceeded, and assists driving by judging learning, monitoring, and driving according to a predetermined flowchart.

[0012] An auxiliary controller using AI and FPGA according to the features of the present invention to solve these problems is as follows:

[0013] As an auxiliary controller using AI and FPGA that assists the controller that controls the control target according to the control variable,

[0014] A first signal distributor for distributing the above control variable;

[0015] A second signal distributor that distributes a control signal output from the above controller to a control target;

[0016] An input module that receives and outputs input signals from the first signal distributor, the second signal distributor, and the user terminal;

[0017] AI that receives the output signal of the above input module and learns the control variable and the output signal of the controller through machine learning;

[0018] It includes an FPGA that assists the operation of the controller by judging learning, monitoring, and driving according to a predetermined flowchart.

[0019] The above AI designs the FPGA if it exceeds a certain accuracy.

[0020] The above control variables include at least one of pressure, temperature, flow rate, and water level.

[0021] It further includes an output module that converts the output signal of the above FPGA into an analog signal and outputs it to a user terminal and a second signal distributor.

[0022] The control method of the auxiliary controller using AI and FPGA according to the features of the present invention to solve these problems is as follows:

[0023] A step in which AI receives input / output signals of the controller from the input module (S201) and performs learning (S202);

[0024] Step (S203) of determining whether the accuracy of the above AI exceeds the set value;

[0025] If the accuracy exceeds the set value, the AI ​​determines whether the controller is outputting normally (S204);

[0026] If the above controller has a normal output, the AI ​​modifies the FPGA circuit (S205);

[0027] Step (S206) where the FPGA monitors the above controller;

[0028] The above FPGA determines whether the controller is out of the allowable error range (S207);

[0029] If the controller does not deviate from the allowable error range, the FPGA monitors the controller (S206), and if the controller deviates from the allowable error range, the FPGA transmits deviation information to the operator's user terminal to notify the operator (S208).

[0030] The above method,

[0031] The driver selects manual driving or assisted driving using the user terminal (S209);

[0032] When manual driving is selected, step (S211) of driving in manual driving;

[0033] When auxiliary driving is selected, the FPGA performs auxiliary driving (S210);

[0034] The above FPGA determines whether the controller is out of the allowable error range (S212);

[0035] If the controller is within the tolerance range, the FPGA notifies the user and receives the user's selection (S213), and if the controller is outside the tolerance range, the FPGA notifies the driver by sending deviation information to the driver's user terminal (S208);

[0036] If the driver selects incomplete, the FPGA performs assist driving (S210), and if the driver selects complete, the FPGA further includes a step of monitoring the controller (S206).

[0037]

[0038] According to one embodiment, an auxiliary controller and a control method thereof using AI and FPGA can be provided, which receives control variables such as pressure, temperature, flow rate, water level, and output using a signal distributor and an input module, learns the control variables and output signals through machine learning, designs an FPGA when a certain accuracy is exceeded, and assists driving by judging learning, monitoring, and driving according to a predetermined flowchart.

[0039] Figure 1 is a block diagram of an auxiliary controller using AI and FPGA according to an embodiment of the present invention.

[0040] FIG. 2 is a diagram illustrating a control method of an auxiliary controller using AI and FPGA according to an embodiment of the present invention.

[0041]

[0042] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present invention. The present invention may be implemented in various different forms and is not limited to the embodiments described herein.

[0043] Additionally, throughout the specification, whenever a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise specifically stated.

[0044] FIG. 1 is a schematic diagram of an auxiliary controller using AI and FPGA according to an embodiment of the present invention.

[0045] Referring to FIG. 1, an auxiliary controller using AI and FPGA according to an embodiment of the present invention,

[0046] An auxiliary controller using AI and FPGA that assists a controller (210) that controls a control target (220) according to a control variable.

[0047] A first signal distributor (110) that distributes the above control variable;

[0048] A second signal distributor (120) that distributes a control signal output from the above controller (210) to the control target (220);

[0049] An input module (130) that receives and outputs input signals from the first signal distributor (110), the second signal distributor (120), and the user terminal (230);

[0050]

[0051] *AI (140) that receives the output signal of the above input module (130) and learns the control variable and the output signal of the controller (210) through machine learning;

[0052] It includes an FPGA (150) that assists the operation of the controller (210) by judging learning, monitoring, and driving according to a predetermined flowchart.

[0053] The above AI (140) designs the FPGA (150) when it exceeds a certain accuracy.

[0054] The above control variables include at least one of pressure, temperature, flow rate, and water level.

[0055] It further includes an output module (160) that converts the output signal of the above FPGA (150) into an analog signal and outputs it to a user terminal (230) and a second signal distributor (120).

[0056]

[0057] A method for controlling an auxiliary controller using AI and FPGA according to an embodiment of the present invention having such a configuration is described as follows.

[0058] FIG. 2 is a diagram illustrating a control method of an auxiliary controller using AI and FPGA according to an embodiment of the present invention.

[0059] Referring to Fig. 2, first, AI (140) receives input / output signals of the controller (210) from the input module (S201) and performs learning (S202).

[0060] Here, AI (140) receives all inputs / outputs of the controller (210) connected to the control target (220) in analog, excluding data through data communication.

[0061] For example, the input of the controller (210) includes pressure, flow rate, temperature, control feedback, etc., and the output of the controller (210) includes control signals such as mA, mV, V, and contacts.

[0062] At this time, the first signal distributor (110) and the second signal distributor (120) must have hardware connections including high-speed relays that do not affect the input / output signals of the controller (210), and the first signal distributor (110) and the second signal distributor (120) have structures formed above and with the same width as the terminal terminals so that there is no interference with the operation of the adjacent terminals. In addition, the second signal distributor (120) can be configured as a relay circuit so that voltage signals can be connected in parallel and current signals can be connected in series.

[0063] And AI (140) learns by changing the constants of first-order or higher-order functions based on each input variable using machine learning (Python, TensorFlow, etc.).

[0064] Next, AI (140) determines whether the accuracy exceeds the set value (S203).

[0065] Here, AI (140) checks whether the accuracy of the function learned over a certain period of time and the input / output values ​​is 99% or higher.

[0066] If the accuracy exceeds the set value, AI (140) determines whether the controller (210) is outputting normally (S204). Here, AI (140) checks the status of the controller (210) itself to confirm its health.

[0067] If the controller (210) outputs normally, the AI ​​(140) modifies the FPGA (150) circuit (S205). Here, the AI ​​(140), which has learned the hardware language based on the learned function, programs and configures the FPGA (150) circuit.

[0068] Next, the FPGA (150) monitors the controller (210) (S206). Here, the FPGA (150) compares the simulation results with the input / output data of the controller (210).

[0069] Next, the FPGA (150) determines whether the controller (210) has exceeded the allowable error range (S207). Here, the allowable error can be, for example, set to 5% error and maintained for 1 second.

[0070] If the controller (210) does not deviate from the allowable error range, the FPGA (150) monitors the controller (210) (S206), and if the controller (210) deviates from the allowable error range, the FPGA (150) transmits deviation information to the operator's user terminal (230) to notify the operator (S208). Depending on the need, the operator notification or the automatic intervention followed by the operator notification can be selected.

[0071] Then, the driver selects manual driving or assisted driving using the user terminal (230) (S209).

[0072] If manual driving is selected, driving is performed manually (S211).

[0073] When auxiliary driving is selected, the FPGA (150) performs auxiliary driving (S210).

[0074] Next, the FPGA (150) determines whether the controller (210) is out of the allowable error range (S212).

[0075] If the controller (210) is within the allowable error range, the FPGA (150) notifies the user and receives the user's selection (S213), and if the controller (210) is outside the allowable error range, the FPGA (150) transmits deviation information to the driver's user terminal (230) to notify the driver (S208).

[0076] Afterwards, if the driver selects incomplete, the FPGA (150) performs assist driving (S210), and if the driver selects action completed, the FPGA (150) monitors the controller (210) (S206).

[0077] According to one embodiment, control variables such as pressure, temperature, flow rate, water level, and output are received using a signal distributor and an input module, the control variables and output signals are learned through machine learning, and when a certain accuracy is exceeded, an FPGA (150) is designed, and learning, monitoring, and driving can be judged according to a predetermined flowchart to assist driving.

[0078] Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

Claims

1. An auxiliary controller using AI and FPGA that assists the controller that controls the control target according to the control variable. A first signal distributor for distributing the above control variable; A second signal distributor that distributes a control signal output from the above controller to a control target; An input module that receives and outputs input signals from the first signal distributor, the second signal distributor, and the user terminal; AI that receives the output signal of the above input module and learns the control variable and the output signal of the controller through machine learning; An auxiliary controller using AI and an FPGA that assists the operation of the controller by judging learning, monitoring, and driving according to a predetermined flowchart.

2. In paragraph 1, The above AI is an auxiliary controller using the AI ​​that designs the FPGA and the FPGA when the above AI exceeds a certain accuracy.

3. In paragraph 2, The above control variable is an auxiliary controller using AI and FPGA including at least one of pressure, temperature, flow rate, and water level.

4. In paragraph 3, An auxiliary controller using AI and FPGA, further comprising an output module that converts the output signal of the above FPGA into an analog signal and outputs it to a user terminal and a second signal distributor.

5. A step in which AI receives input / output signals of the controller from the input module (S201) and performs learning (S202); Step (S203) of determining whether the accuracy of the above AI exceeds the set value; If the accuracy exceeds the set value, the AI ​​determines whether the controller is outputting normally (S204); If the above controller has a normal output, the AI ​​modifies the FPGA circuit (S205); Step (S206) where the FPGA monitors the above controller; The above FPGA determines whether the controller is out of the allowable error range (S207); A method for controlling an auxiliary controller using AI and an FPGA, including a step (S208) in which the FPGA monitors the controller if the controller does not deviate from the allowable error range and the FPGA sends deviation information to the operator's user terminal to notify the operator if the controller deviates from the allowable error range.

6. In paragraph 5, The driver selects manual driving or assisted driving using the user terminal (S209); When manual driving is selected, step (S211) of driving in manual driving; When auxiliary driving is selected, the FPGA performs auxiliary driving (S210); The above FPGA determines whether the controller is out of the allowable error range (S212); If the controller is within the tolerance range, the FPGA notifies the user and receives the user's selection (S213), and if the controller is outside the tolerance range, the FPGA notifies the driver by sending deviation information to the driver's user terminal (S208); A control method of an auxiliary controller using AI and FPGA, which further includes a step of the FPGA performing auxiliary driving when the driver selects incomplete (S210) and the FPGA monitoring the controller when the driver selects action complete (S206).

Images