Ammonia adsorption tower driving control device and control method using same

Abstract

The ammonia adsorption tower drive control apparatus according to the present disclosure comprises: a sensor unit that measures an internal state of an adsorption tower; a memory storing one or more instructions; a processor that executes the one or more instructions stored in the memory; the driving unit is used for driving the adsorption tower according to an adsorption period and a desorption period which are set based on the internal state of the adsorption tower, and the processor is configured to output the corresponding adsorption period and desorption period according to sensing data of the sensor unit by utilizing a learned artificial intelligence model, when the sensing data is within a preset optimal range, an instruction for driving the adsorption tower according to the output adsorption period and desorption period is transmitted to the driving unit, and when the sensing data exceeds the optimal range, the adsorption tower is driven by the driving unit. And transmitting an instruction for executing an abnormal condition processing process according to a preset abnormal adsorption period and an abnormal desorption period to the driving unit.

Description

Technical Field

[0001] The present invention relates to an ammonia adsorption tower drive control device and a control method using the device. Background Art

[0002] Ammonia is one of the substances that cause bad odors. It is not only produced in livestock farms, but also in composting facilities, factories and sewage treatment plants that use fertilizers or chemical fibers.

[0003] Conventionally, an adsorption tower is driven to adsorb and remove ammonia in air injected into the adsorption tower, and when the removal efficiency of the adsorption tower decreases due to adsorption of ammonia above a certain level, control is performed to desorb the ammonia to maintain a state capable of adsorption again.

[0004] In order to repeat the above process, it is very important to control the adsorption cycle and desorption cycle so that the adsorption and desorption processes can be carried out in time. Traditionally, the volatile organic compounds (VOCs) adsorbed on the adsorbent are desorbed by steam through the vapor recovery method, and then the desorbed vapor and VOCs are recovered and reused through a heat exchanger, and the desorption time of the adsorbent is sensed by a temperature sensor.

[0005] However, even if the desorption time is sensed or the desorption time has passed and an abnormal warning is issued, if the operator does not take timely action to resolve the abnormal state, the adsorption efficiency will decrease, and even if the operator responds in time, the adsorption efficiency cannot be maintained consistent because the method of responding to the abnormal state varies depending on the operator's proficiency.

[0006] In addition, even if an artificial intelligence model is used, the artificial intelligence model that outputs the optimal adsorption cycle and desorption cycle based on normal conditions may repeatedly output incorrect prediction values ​​based on abnormal conditions, resulting in incorrect learning and inaccurate predictions of the adsorption cycle and desorption cycle under normal conditions. Summary of the invention

[0007] 1. Technical issues to be resolved

[0008] According to one aspect of the present disclosure, an artificial intelligence-based ammonia adsorption tower drive control device and a control method using the device can be provided. When an abnormal state occurs, an abnormal situation handling process based on abnormal adsorption cycle and abnormal desorption cycle can be executed regardless of the predicted adsorption and desorption cycles, thereby contributing to environmental protection and producing clean energy.

[0009] (II) Technical solution

[0010] According to one embodiment of the present disclosure, an ammonia adsorption tower drive control device may include: a sensor unit for measuring the internal state of the adsorption tower; a memory for storing one or more instructions; a processor for executing the one or more instructions stored in the memory; and a drive unit for driving the adsorption tower according to an adsorption cycle and a desorption cycle set based on the internal state of the adsorption tower, and the processor may be configured to: output the corresponding adsorption cycle and desorption cycle according to the sensing data of the sensor unit using a learned artificial intelligence model, and when the sensing data is within a preset optimal range, transmit to the drive unit an instruction for driving the adsorption tower according to the outputted adsorption cycle and desorption cycle; when the sensing data exceeds the optimal range, transmit to the drive unit an instruction for executing an abnormal situation handling process according to a preset abnormal adsorption cycle and abnormal desorption cycle.

[0011] According to one embodiment, the abnormal adsorption period and the abnormal desorption period may be the shortest adsorption period and the longest desorption period within a settable period range of the driving unit.

[0012] According to one embodiment, when the processor transmits an instruction to the driving unit to execute the abnormal situation handling process, the instruction may include parameters based on at least one of the following: a minimum adsorption flow rate and a maximum desorption flow rate that the driving unit can drive; a minimum adsorption temperature and a maximum desorption temperature that the driving unit can drive; and a minimum adsorption pressure and a maximum desorption pressure that the driving unit can drive.

[0013] According to one embodiment, the processor may be configured to repeatedly self-diagnose at a preset cycle whether the sensing data measured by the sensor unit exceeds an optimal range or whether a pattern of previous sensing data and a pattern of subsequent sensing data are different.

[0014] According to one embodiment, after driving the adsorption tower according to the abnormal situation processing process, when the sensing data measured by the sensor unit is within the optimal range, an instruction to drive the adsorption tower using the adsorption cycle and desorption cycle output from the learned artificial intelligence model can be transmitted to the driving unit.

[0015] According to one embodiment, the processor can be configured to: when the previous adsorption cycle and the previous desorption cycle executed before driving the adsorption tower according to the abnormal situation handling process are different from the next adsorption cycle and the next desorption cycle to drive the adsorption tower after the abnormal situation handling process ends, the cost and time reduction data when the driving unit is driven are calculated using the next adsorption cycle and the next desorption cycle.

[0016] According to one embodiment, when the calculated cost and time reduction data do not exceed a preset threshold, the processor may transmit an instruction to the driving unit to maintain a previous adsorption cycle and a previous desorption cycle.

[0017] According to one embodiment, in order for the artificial intelligence model to detect an abnormal state, the processor may remove sensing data of an abnormal state at each time point of the process and may transmit only sensing data of a normal state to the artificial intelligence model.

[0018] According to one embodiment, the method may further include: a user interface, which provides an alarm to a user and receives information about a process of handling the abnormal situation when the abnormal state occurs.

[0019] According to one embodiment, the sensor unit may include at least one sensor among a pressure sensor, a temperature sensor, a humidity sensor, a gas sensor and an adsorbent sensor. When an error is detected in the at least one sensor through self-diagnosis, or a communication error is detected between the at least one sensor and the processor, the processor may transmit an instruction to the drive unit to stop the drive of the adsorption tower.

[0020] According to an ammonia adsorption tower drive control method according to an embodiment of the present disclosure, which is executed by a computing device, the computing device includes a memory storing one or more instructions and a processor executing the one or more instructions stored in the memory, the ammonia adsorption tower drive control method may include: a step of collecting sensing data for measuring the internal state of the adsorption tower; and a step of driving the adsorption tower according to the adsorption cycle and desorption cycle set based on the sensing data, the step of driving the adsorption tower may further include: a step of outputting the corresponding adsorption cycle and desorption cycle according to the sensing data using a learned artificial intelligence model; when the sensing data is within a preset optimal range, a step of transmitting to the drive unit an instruction to drive the adsorption tower according to the outputted adsorption cycle and desorption cycle; and when the sensing data exceeds the optimal range, a step of transmitting to the drive unit an instruction to execute an abnormal situation handling process according to a preset abnormal adsorption cycle and abnormal desorption cycle.

[0021] (III) Beneficial effects

[0022] According to one embodiment of the present disclosure, when an abnormal state occurs, an abnormal situation handling process is performed regardless of the predicted adsorption and desorption cycle, and the internal state of the adsorption tower is quickly restored to a normal state, thereby reducing costs and maximizing adsorption efficiency.

[0023] In addition, according to the present disclosure, the driving conditions are verified before the adsorption tower is driven according to the predicted adsorption and desorption cycles output by the artificial intelligence model, so as to change the driving conditions according to the costs incurred when changing the adsorption and desorption cycles and the operating performance efficiency to be improved, thereby reducing the costs and time caused by changing the driving conditions and maximizing the adsorption efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A conceptual diagram briefly illustrating an adsorption tower according to one embodiment of the present disclosure.

[0025] Figure 2 : is a block diagram briefly showing the configuration of an ammonia adsorption tower drive control device according to one embodiment of the present disclosure.

[0026] Figure 3 A flow chart showing a method for driving and controlling an ammonia adsorption tower according to an embodiment of the present disclosure is shown.

[0027] Figure 4 A detailed flow chart of an ammonia adsorption tower driving control method according to an embodiment of the present disclosure is shown.

[0028] Figure 5 A flow chart is shown when a state situation occurs in the ammonia adsorption tower driving control method according to one embodiment of the present disclosure.

[0029] Figure 6 A computing device according to one embodiment of the present disclosure is shown.

[0030] Description of Reference Numerals

[0031] 100: Ammonia adsorption tower drive control device

[0032] 110: Sensor unit

[0033] 120: Computing device

[0034] 14: Processor

[0035] 16: Memory

[0036] 130: Drive unit

[0037] 140: User Interface

[0038] 150: Display unit

[0039] 160: Artificial Intelligence Model DETAILED DESCRIPTION

[0040] Hereinafter, the present disclosure will be described in detail with reference to the accompanying drawings. However, this is merely exemplary, and the present disclosure is not limited to the specific embodiments exemplarily described.

[0041] In the present disclosure, the same reference numerals refer to the same components. Not all elements of the embodiments are described in the present disclosure, and general contents of the technical field to which the present disclosure belongs or contents repeated between embodiments are omitted. The terms "unit, module, component, block" used in the specification may be implemented by software or hardware, and depending on the embodiment, multiple "units, modules, components, blocks" may be implemented by a single component, or one "unit, module, component, block" may include multiple components.

[0042] Throughout the specification, when one part is referred to as being “connected” to another part, this includes not only the case of being directly connected but also the case of being indirectly connected including being connected through a wireless communication network.

[0043] In addition, when a section is referred to as “comprising” a component, unless specifically stated otherwise, it means that other components may be further included, rather than excluding other components.

[0044] Throughout the specification, when a member is located “on” another member, this includes not only a case where one member is in contact with another member but also a case where other members exist between the two members.

[0045] Terms such as first, second, etc. are used to distinguish one component from another component, and the components are not limited by the above terms.

[0046] Unless the context clearly indicates otherwise, an expression in the singular includes an expression in the plural.

[0047] The reference numerals of the steps are used for convenience of description and do not describe the order of the steps. Unless the context clearly indicates a specific order, the steps may be performed in an order different from the specified order.

[0048] In this specification, "devices according to the present disclosure" include all kinds of devices that can perform computing and provide results to users. For example, the devices according to the present disclosure may include computers, server devices, and portable terminals, or any one of them.

[0049] Here, the computer may include, for example, a notebook computer equipped with a web browser, a desktop computer, a laptop computer, a tablet PC, a touch-screen tablet PC, etc.

[0050] The server device is a server that processes information by communicating with an external device, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, a network server, etc.

[0051] The portable terminal is, for example, a wireless communication device that ensures portability and mobility, and may include all types of handheld wireless communication devices such as PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet), terminals, smart phones, and wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMD).

[0052] The functions related to artificial intelligence according to the present disclosure are operated by a processor and a memory. The processor may be composed of one or more processors. At this time, one or more processors may be general-purpose processors such as a CPU (central processing unit), an AP or a DSP (digital signal processor), a graphics-specific processor such as a GPU (graphics processing unit) or a VPU (vision processing unit), or an artificial intelligence-specific processor such as an NPU (neural network processor). One or more processors control to process input data according to predefined operating rules or artificial intelligence models stored in the memory. Alternatively, when one or more processors are artificial intelligence-specific processors, the artificial intelligence-specific processors can be designed as a hardware structure specifically for processing a specific artificial intelligence model.

[0053] The characteristics of the predefined operating rules or artificial intelligence models are established through learning. Here, established through learning means that the basic artificial intelligence model is learned by a learning algorithm using a large amount of learning data, thereby establishing predefined operating rules or artificial intelligence models to perform the desired characteristics (or purposes). This learning can be performed in the device itself that executes the artificial intelligence according to the present disclosure, or it can be performed by a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the above examples.

[0054] The artificial intelligence model can be composed of multiple neural network layers. Each of the multiple neural network layers has multiple weight values, and the neural network operation is performed by the operation result of the previous layer and the operation between the multiple weight values. The multiple weight values ​​of the multiple neural network layers can be optimized according to the learning results of the artificial intelligence model. For example, during the learning process, multiple weight values ​​can be updated to reduce or minimize the loss value or cost value obtained from the artificial intelligence model. The artificial neural network may include a deep neural network (Deep Neural Network, DNN), such as a convolutional neural network (Convolutional Neural Network, CNN), a deep neural network (Deep Neural Network, DNN), a recurrent neural network (Recurrent Neural Network, RNN), a restricted Boltzmann machine (Restricted Boltzmann Machine, RBM), a deep belief network (Deep Belief Network, DBN), a bidirectional recursive deep neural network (Bidirectional Recurrent Deep Neural Network, BRDNN) or a deep Q network (Deep Q-Networks), etc., but not limited to the above examples.

[0055] According to an exemplary embodiment of the present disclosure, the processor can implement artificial intelligence. Artificial intelligence refers to a machine learning method based on an artificial neural network (ANN), which enables a machine to learn by imitating human biological neurons (biological neuron). According to the learning method, the artificial intelligence method can be divided into supervised learning, unsupervised learning, and reinforcement learning. The supervised learning is to provide input data and output data as learning data at the same time, so that the answer (output data) to the question (input data) has been determined. The unsupervised learning is to only provide input data without providing output data, so that the answer (output data) to the question (input data) has not been determined. The reinforcement learning is that whenever an action (Action) is taken in the current state (State), the external environment will give a reward (Reward) so that the learning proceeds in the direction of maximizing the reward. In addition, the artificial intelligence method can be distinguished according to the architecture of the structure of the learning model. The architecture of the widely used deep learning technology can be distinguished into convolutional neural networks (CNN), recursive neural networks (RNN), Transformer, generative adversarial networks (GAN), etc.

[0056] The device and system according to the present disclosure may include an artificial intelligence model. The artificial intelligence model may be one artificial intelligence model or multiple artificial intelligence models. The artificial intelligence model may be composed of a neural network (or artificial neural network), and may include a statistical learning algorithm that imitates biological neurons in machine learning and cognitive science. A neural network may refer to an overall model in which artificial neurons (nodes) that form a network through the combination of synapses change the binding strength of synapses through learning, and have problem-solving capabilities. Neurons in a neural network may include a combination of weight values ​​or biases. A neural network may include more than one layer consisting of one or more neurons or nodes. For example, a device may include an input layer, a hidden layer, and an output layer. The neural network constituting the device can change the weight values ​​of neurons through learning, so that the result to be predicted (output) can be inferred from any input (input).

[0057] The processor may generate a neural network, train or learn a neural network, perform operations based on received input data, generate information signals based on execution results, or retrain a neural network. The neural network model may include various types of models such as CNN (convolutional neural network) such as GoogleNet, AlexNet, VGG network, R-CNN (Region with Convolution Neural Network), RPN (Region Proposal Network), RNN (Recurrent Neural Network), S-DNN (Stacking-based deep Neural Network), S-SDNN (State-Space Dynamic Neural Network), Deconvolution Network, DBN (Deep Belief Network), RBM (Restrcted Boltzman Machine), Fully Convolutional Network, LSTM (Long Short-Term Memory) network, Classification Network, etc., but is not limited thereto. The processor may include more than one processor for performing operations according to a model of a neural network. For example, the neural network may include a deep neural network.

[0058] Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward) and LSTM (Long Short-Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Deep Belief Network), DCN (Deep Convolutional Network), etc. The present invention relates to a neural network comprising: a neural network (CNN), a deep neural network (ANN), a deep neural network (DLN), a deep neural network (DNC ...

[0059] According to an exemplary embodiment of the present disclosure, the processor can use CNN (Convolutional Neural Network) such as GoogleNet, AlexNet, VGG network, R-CNN (Region with Convolution Neural Network), RPN (Region Proposal Network), RNN (Recurrent Neural Network), S-DNN (Stacking-based deep Neural Network), S-SDNN (State-Space Dynamic Neural Network), Deconvolution Network, DBN (Deep Belief Network), RBM (Restrcted Boltzman Machine), Fully Convolutional Network, LSTM (Long Short-Term Memory) Network, Classification Network, Generative Modeling, Explainable AI, Continuous AI, Representation Learning. Learning), Material Design AI, BERT for Natural Language Processing, SP-BERT, MRC / QA, Text Analysis, Dialog System, GPT-3, GPT-4, Visual Analytics for Visual Processing, Visual Understanding, Video Synthesis, Anomaly Detection for ResNet Data Intelligence, Prediction, Time-Series Forecasting, Optimization, Recommendation, and Data Creation, but not limited thereto. Below, the embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0060] Figure 1 is a conceptual diagram briefly showing an adsorption tower 10 according to an embodiment of the present disclosure. Figures 2 to 5The ammonia adsorption tower drive control device based on the above-mentioned artificial intelligence and the control method using the device are described in detail.

[0061] like Figure 1 As shown, the adsorption tower 10 receives air (feed) containing ammonia, volatile organic compounds (VOC), etc., and during the process from T1 to T4, the air (product) from which ammonia, volatile organic compounds (VOC), etc. can be removed by applying steam recovery methods such as recompression, condensation, and activated carbon adsorption.

[0062] The adsorption tower 10 can be divided into regions T1, T2, T3, and T4 from a position close to the suction port for receiving the feed to a position of the discharge port for discharging the product, and a sensor unit 110 can be installed in each of the divided regions (refer to Figure 2 The sensor unit 110 may include at least one of a pressure sensor measuring the internal pressure of the adsorption tower 10, a temperature sensor measuring the internal temperature of the adsorption tower 10, a humidity sensor measuring the internal humidity of the adsorption tower 10, a gas sensor measuring the internal gas composition of the adsorption tower 10 to check the purity of the extracted hydrogen, and an adsorbent sensor.

[0063] Whether the pressure or temperature is maintained within a predetermined range to maintain adsorption efficiency may be confirmed using the sensor unit 110 , and a result value regarding hydrogen extraction efficiency may be calculated using humidity or hydrogen purity.

[0064] In addition, the adsorbent sensor of the sensor unit 110 can measure the state of the adsorbent inside the adsorption tower, evaluate the life and performance of the adsorbent, and obtain information that can predict the replacement time of the adsorbent.

[0065] like Figure 1 As shown, the more active the adsorption, the higher the temperature and the lower the adsorption efficiency. The closer the adsorbent replacement time is, the smaller the temperature rise. Therefore, the change in the adsorbent state can be seen based on the temperature change.

[0066] At this time, it can be seen that, from T1 to T4, the later the contact time with the feed is, the later the adsorbent replacement time occurs, and it can be seen that the adsorbent replacement times from T1 to T4 do not occur together.

[0067] In addition, when an abnormal state occurs, even if the adsorbent replacement time has not yet arrived, the internal temperature variation range of the adsorption tower 10 decreases, so it may be mistakenly believed that the adsorbent replacement time has occurred.

[0068] Therefore, if Figure 2As shown, according to an embodiment of the present disclosure, an artificial intelligence-based ammonia adsorption tower drive control device 100 may include: a sensor unit 110, measuring the internal state of the adsorption tower 10; a memory 16, storing one or more instructions; a processor 14, executing the one or more instructions stored in the memory 16; and a drive unit 130, driving the adsorption tower 10 according to an adsorption cycle and a desorption cycle set based on the internal state of the adsorption tower 10.

[0069] The sensor units 110 may be disposed at locations T1 to T4 inside the adsorption tower 10 , respectively.

[0070] As one example, the interior of the adsorption tower 10 may be separated and partitioned into a plurality of beds, and the sensor units 110 may be respectively installed in the separated and partitioned beds.

[0071] As another example, a plurality of adsorption towers 10 may be connected, and the sensor unit 110 may be installed in each of the adsorption towers 10 , respectively.

[0072] In other words, in at least one adsorption tower 10 separated into separate spaces by dividing one adsorption tower 10 into a plurality of beds or connecting a plurality of adsorption towers 10 , the sensor unit 110 may be provided in each separated space.

[0073] The processor 14 and the memory 16 may be disposed together or separately in one or more computing devices 120. The computing device 120 may be used to control the driving of the adsorption tower 10 so that the efficiency of the adsorption tower 10 is maintained above a predetermined level.

[0074] The artificial intelligence model 160 may be installed in the computing device 120 or may be connected via a separate server or cloud.

[0075] In addition, the artificial intelligence-based ammonia adsorption tower drive control device 100 according to an embodiment of the present disclosure may further include a user interface (User Interface) 140 , and the user interface 140 is used to receive user input to drive the adsorption tower 10 .

[0076] When an abnormal state occurs, the user interface 140 may provide the user with an alert and receive information about the abnormal situation handling process.

[0077] In addition, the artificial intelligence-based ammonia adsorption tower drive control device 100 according to an embodiment of the present disclosure may further include a display unit 150, which provides an alarm for driving the adsorption tower 10 or monitors the operating status of the adsorption tower 10 in real time and displays the monitoring results.

[0078] As an embodiment, the processor 14 uses the learned artificial intelligence model 160 to output the corresponding adsorption cycle and desorption cycle according to the sensing data of the sensor unit 110. When the sensing data is within a preset optimal range, the instruction for driving the adsorption tower according to the output adsorption cycle and desorption cycle is transmitted to the driving unit 130. When the sensing data exceeds the optimal range, the instruction for executing the abnormal situation handling process according to the preset abnormal adsorption cycle and abnormal desorption cycle is transmitted to the driving unit 130.

[0079] The artificial intelligence-based ammonia adsorption tower drive control device 100 identifies abnormal conditions based on data collected by the sensor unit 110, and can utilize an artificial intelligence model 160 that learns the normal conditions of the pressure, temperature, humidity, concentration, etc. of the gas in the adsorption tower 10 at each time point of the adsorption process to detect abnormal conditions in real time when they occur.

[0080] The processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 can repeatedly self-diagnose at a preset cycle whether the sensing data measured by the sensor unit 110 exceeds the optimal range or whether the pattern of the previous sensing data and the pattern of the subsequent sensing data are different, and can use the artificial intelligence model 160 to detect abnormal conditions in real time when abnormal conditions occur.

[0081] In addition, when performing self-diagnosis, the processor 14 can not only detect abnormal conditions inside the adsorption tower 10, but also detect whether a sensor error occurs in at least one sensor of the sensor unit 110, or detect whether a communication error occurs between the sensor unit 110 and the processor 14. When a sensor error or a communication error is detected, an instruction to stop driving the adsorption tower 10 can be transmitted to the drive unit 130.

[0082] As an example, when a normal state is identified, the processor 14 may perform the adsorption process at optimal efficiency using the adsorption cycle and desorption cycle output by the artificial intelligence model 160. At this time, optimal efficiency refers to a state in which the adsorption efficiency can be maintained above a predetermined level while reducing costs by delaying the replacement time of the adsorbent to the maximum extent.

[0083] As an embodiment, when an abnormal state is identified, the processor 14 may not execute the adsorption process with the adsorption cycle and desorption cycle output by the artificial intelligence model 160, but instead start a preset abnormal situation handling process and command the adsorption tower 10 to drive with the adsorption cycle and desorption cycle that recover to the normal state as quickly as possible.

[0084] Here, the adsorption cycle and desorption cycle that are restored to the normal state fastest are preset abnormal adsorption cycle and abnormal desorption cycle, and the abnormal adsorption cycle and the abnormal desorption cycle may be the shortest adsorption cycle and the longest desorption cycle within the settable cycle range of the driving unit 130 .

[0085] Alternatively, when the processor 14 transmits an instruction to the driving unit 130 to execute the abnormal situation handling process, the instruction may include parameters based on at least one of the minimum adsorption flow rate and the maximum desorption flow rate that can be driven by the driving unit 130, the minimum adsorption temperature and the maximum desorption temperature, and the minimum adsorption pressure and the maximum desorption pressure.

[0086] Specifically, in addition to the abnormal adsorption cycle and the abnormal desorption cycle, the processor 14 may also transmit abnormal parameters corresponding to the abnormal state to the driving unit 130. The processor 14 may transmit to the driving unit 130 the minimum adsorption flow rate and the maximum desorption flow rate that the driving unit 130 can drive, measure or calculate the minimum adsorption flow rate that can be handled when the adsorbent reaches the best adsorption efficiency before the replacement cycle, and measure or calculate the maximum desorption flow rate that can be handled when the adsorbent reaches the best desorption efficiency before the replacement cycle.

[0087] Alternatively, the processor 14 can transmit to the driving unit 130 the lowest adsorption temperature and the highest desorption temperature or the minimum adsorption pressure and the maximum desorption pressure that the driving unit 130 can drive, measure or calculate the lowest adsorption temperature or the minimum adsorption pressure that occurs when the adsorbent maintains the best adsorption efficiency before the replacement cycle, and measure or calculate the highest desorption temperature or the maximum desorption pressure that occurs when the adsorbent maintains the best desorption efficiency before the replacement cycle.

[0088] At this time, the adsorption cycle and desorption cycle output by the artificial intelligence model 160 can be stored separately for use when returning to normal state later.

[0089] As an embodiment, after an abnormal state is identified, the adsorption tower 10 is driven according to the abnormal situation processing process. When the sensing data measured by the sensor unit 110 is within the optimal range again, the adsorption cycle and desorption cycle output by the learned artificial intelligence model 160 can be used to drive the adsorption tower 10.

[0090] At this time, the artificial intelligence model 160 may be operated again to output the adsorption period and the desorption period, or the adsorption period and the desorption period output by the artificial intelligence model 160 previously stored separately may be called.

[0091] Therefore, the ammonia adsorption tower drive control device 100 based on artificial intelligence can identify the normal state and the abnormal state in advance, and control the driving of the adsorption tower 10 accordingly to perform a separate adsorption process.

[0092] On the other hand, the processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 is configured to calculate the cost and time reduction data when the drive unit is driven using the next adsorption cycle and the next desorption cycle when the previous adsorption cycle and the previous desorption cycle executed before driving the adsorption tower 10 according to the abnormal situation handling process are different from the next adsorption cycle and the next desorption cycle to be driven by the adsorption tower after the abnormal situation handling process is completed.

[0093] Specifically, when there is an adsorption and desorption cycle P1 of the driving adsorption tower 10 before abnormal state recognition, an abnormal adsorption and desorption cycle P2, and an adsorption and desorption cycle P2 output again by the artificial intelligence model 160 after the normal state is restored, the processor 14 may not immediately use P2 to perform the adsorption process, but perform the adsorption process after finally verifying the adsorption process according to P1 and P2.

[0094] For example, the cost incurred when the driving condition of the adsorption tower 10 is changed from P1 to P2 and the operating performance efficiency to be improved may be calculated and compared, and then the adsorption and desorption cycle may be changed.

[0095] In other words, the calculated expected cost and efficiency of changing the adsorption and desorption cycle are compared with the current adsorption and desorption cycle cost and efficiency. Only when there is a cost reduction and efficiency increase effect above a preset threshold, can the adsorption tower 10 be controlled to change the adsorption and desorption cycle.

[0096] Considering the loss caused by frequently changing the adsorption and desorption cycles and the process delay during the cycle change, the adsorption tower 10 can be controlled by reflecting the comprehensive driving conditions of the adsorption tower 10 .

[0097] Therefore, when the calculated cost and time reduction data do not exceed the preset threshold, the processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 can transmit instructions to the drive unit 130 to maintain the previous adsorption cycle and the previous desorption cycle.

[0098] The processor 14 of the ammonia adsorption tower drive control device 100 based on artificial intelligence can transmit the adsorption tower 10 drive data as the learning data of the artificial intelligence model 160. In order for the artificial intelligence model 160 to detect the abnormal state, the sensing data of the abnormal state can be removed at each time point of the process, and only the sensing data of the normal state can be transmitted to the artificial intelligence model.

[0099] On the other hand, Figure 3As shown, the control method using the artificial intelligence-based ammonia adsorption tower drive control device 100 may include: a step (S310) of collecting sensing data for measuring the internal state of the adsorption tower 10; and a step (S320) of driving the adsorption tower 10 according to the adsorption cycle and desorption cycle set based on the sensing data.

[0100] The step of driving the adsorption tower 10 may further include: a step of using the learned artificial intelligence model 160 to output the corresponding adsorption cycle and desorption cycle according to the sensing data; when the sensing data is within a preset optimal range, a step of transmitting to the driving unit 130 an instruction to drive the adsorption tower 10 according to the outputted adsorption cycle and desorption cycle; when the sensing data exceeds the optimal range, a step of transmitting to the driving unit 130 an instruction to execute an abnormal situation handling process according to a preset abnormal adsorption cycle and abnormal desorption cycle.

[0101] Specifically, Figure 4 As shown, in the data collection step (S410), the sensing data inside the initial adsorption tower 10 can be collected by the sensor unit 110. Then, in the data preprocessing step (S420), the sensing data can be preprocessed, such as data normalization, missing value processing, and removal of abnormal values ​​measured under abnormal conditions.

[0102] In the artificial intelligence model learning step (S430), the artificial intelligence model 160 can be learned using the preprocessed data to generate a model that predicts the optimal driving conditions of the adsorption tower 10. Machine learning or deep learning algorithms can be used for learning to output the optimal adsorption and desorption cycle corresponding to the sensing data, and in particular, the adsorbent replacement cycle and cost, the optimal adsorption efficiency, and the various process times and gas purity for maintaining the optimal adsorption efficiency can be considered for learning to output the optimal adsorption and desorption cycle.

[0103] After the learning is completed, the optimal adsorption and desorption cycle corresponding to the internal state of the current adsorption tower 10 may be output in the adsorption and desorption cycle output step ( S440 ) using the learned artificial intelligence model 160 .

[0104] Instead of immediately performing the adsorption and desorption process at the expected adsorption and desorption cycle, the abnormal state detection step (S450) is performed, and if no abnormal state is detected, the driving condition verification step (S460) may be performed.

[0105] In the abnormal state detection step (S450), it can be detected whether the temperature, humidity or pressure range predicted by the artificial intelligence model 160 based on the current amount of adsorbent and the internal process of the adsorption tower 10 is exceeded, and the difference between the predicted value and the current measured value is judged. If the difference is greater than or equal to the threshold, it can be identified as an abnormal state and an alarm is provided to the display unit 150.

[0106] The difference between the predicted value and the current measured value is judged. When the difference is less than the threshold and does not exceed the normal state range, it can be determined in the driving condition verification step (S460) whether to maintain or change the existing adsorption and desorption cycle based on the amount of adsorbent required to generate gas, the amount of energy, the application time of each step, etc.

[0107] Finally, after the driving condition verification is completed, the adsorption tower 10 driving step (S470) may be performed according to the determined adsorption and desorption cycles.

[0108] Or, if Figure 5 As shown, if an abnormal state is identified in the abnormal state detection step (S510), an abnormal situation processing process execution step (S520) according to an abnormal adsorption and desorption cycle may be performed. The abnormal state detection step (S510) is the same step as step S410.

[0109] In the abnormal situation handling process execution step (S520), the adsorption and desorption processes may be performed according to the preset abnormal adsorption cycle and abnormal desorption cycle to restore the internal state of the adsorption tower 10 to the normal state as soon as possible.

[0110] Afterwards, in the normal state recovery detection step (S530), the adsorption and desorption process is performed according to the abnormal adsorption cycle and the abnormal desorption cycle and self-diagnosis is performed. When the sensing data measured by the sensor unit 110 enters the normal state range, it is detected as the normal state is restored, and the adsorption and desorption cycle output step (S540) using the learned artificial intelligence model 160 can be performed. The adsorption and desorption cycle output step (S540) using the learned artificial intelligence model 160 is the same step as step S440. After step S540, steps S450, S460, and S470 can be performed.

[0111] In addition, after the processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 drives the adsorption tower 10 according to the output adsorption cycle and desorption cycle, when the actual process results measured by the sensor unit 110 exceed the predicted process results and the preset range, instructions for executing the abnormal situation handling process according to the abnormal adsorption cycle and the abnormal desorption cycle can be transmitted to the drive unit 130.

[0112] Alternatively, after the processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 drives the adsorption tower 10 according to the output adsorption cycle and desorption cycle, when the actual process results measured by the sensor unit 110 exceed the predicted process results and the preset range, the artificial intelligence model 160 can transmit re-learning instructions.

[0113] When the sensing data of the sensor unit 110 does not exceed the normal range but the adsorption efficiency does not reach the prediction, the abnormal situation handling process can be driven to reset the internal state of the adsorption tower 10, or re-learn to eliminate the possibility of erroneous learning of the artificial intelligence model 160.

[0114] As another embodiment, the processor 14 of the artificial intelligence-based ammonia adsorption tower drive control device 100 can determine whether the abnormal state is temporary or permanent, and apply different abnormal situation processing procedures according to the abnormal state.

[0115] When the sensor unit 110 temporarily mismeasures or a communication network fails due to external factors, since this is a temporary abnormal state, the adsorption and desorption process may be performed using a preset abnormal adsorption and desorption period.

[0116] When the abnormal state is due to a failure of the sensor unit 110 or a communication error between at least one sensor unit 110 and the processor 14 caused by a failure of the communication unit, since this is a permanent abnormal state, the abnormal state cannot be eliminated unless repaired from the outside, so the drive of the adsorption tower 10 can be stopped.

[0117] Alternatively, when the abnormal state is permanent, the average value of the adsorption and desorption cycles of the outputs stored in the memory 16 may be calculated, and the adsorption and desorption process may be performed according to the calculated adsorption and desorption cycles. In order to prevent costs and losses from occurring during the waiting period for the abnormal state to be eliminated, the adsorption tower 10 may be driven at any efficiency.

[0118] In another embodiment, the abnormal state is an abnormal pattern that deviates from the normal state, which can be learned by patterning the parameters measured under the normal state. For example, the artificial intelligence model 160 can pattern the temperature change inside the adsorption tower 10 for learning. When the pattern changes, it can be detected as an abnormal state.

[0119] like Figure 1 As shown, when a temperature pattern different from the normal state appears in an abnormal state, the artificial intelligence model 160 can detect the difference from the learned pattern and perform an abnormal situation processing process. At this time, the pattern may include a temperature pattern, a pressure pattern, a humidity pattern, a gas concentration pattern, etc.

[0120] Thus, the desorption process can be performed in time, so that the adsorbent can be continuously used while the efficiency of the adsorption tower 10 is maintained above a predetermined level, so that the adsorption tower 10 can be stably driven not only in normal conditions but also in abnormal conditions.

[0121] In addition, by simulation before and after the adsorption and desorption cycle control, the adsorption tower 10 is driven in consideration of the cost and the adsorption efficiency, so that the cost can be reduced and the optimal adsorption efficiency can be maintained for a long time.

[0122] On the other hand, Figure 6 As shown, the artificial intelligence-based ammonia adsorption tower drive control device 100 according to the present disclosure can be implemented in the form of a recording medium storing instructions executable by a computer. The instructions can be stored in the form of program code, and when the instructions are executed by a processor, a program module can be generated to perform the operations of the disclosed embodiment. The recording medium can be implemented in the form of a computer-readable recording medium.

[0123] The artificial intelligence-based ammonia adsorption tower drive control device according to the present disclosure may correspond to a computing device 12, which may include at least one processor 14, a computer-readable storage medium 16 including a program 20, and a communication bus 18. In addition, the computing device 12 may further include one or more input / output interfaces 22 and one or more network communication interfaces 26 that provide interfaces for the input / output device 24.

[0124] The computing device 120 of the present disclosure may correspond to the computing device 12. The user interface 140 and the display unit 150 of the present disclosure may correspond to one or more input / output interfaces 22 that provide interfaces for the input / output device 24. Therefore, each component of the artificial intelligence-based ammonia adsorption tower drive control device 100 according to the present disclosure may be fully included in the above-mentioned one computing device 12 or each component may be implemented in a separate device.

[0125] As described above, the disclosed embodiments are described with reference to the accompanying drawings. Those skilled in the art to which the present disclosure belongs will understand that the present disclosure can be implemented in a form different from the disclosed embodiments without changing the technical ideas or essential features of the present disclosure. The disclosed embodiments are exemplary and should not be interpreted as limiting.

Claims

1. An ammonia adsorption tower drive control device, comprising: A sensor unit, measuring the internal state of the adsorption tower; a memory storing one or more instructions; a processor, executing the one or more instructions stored in the memory; as well as a driving unit, driving the adsorption tower according to an adsorption cycle and a desorption cycle set based on an internal state of the adsorption tower, The processor is configured to: Outputting the corresponding adsorption cycle and desorption cycle according to the sensing data of the sensor unit using the learned artificial intelligence model, When the sensing data is within a preset optimal range, an instruction for driving the adsorption tower according to the output adsorption cycle and desorption cycle is transmitted to the driving unit, When the sensing data exceeds the optimal range, an instruction for executing an abnormal situation processing process according to a preset abnormal adsorption cycle and abnormal desorption cycle is transmitted to the driving unit.

2. The ammonia adsorption tower drive control device according to claim 1, wherein: The abnormal adsorption period and the abnormal desorption period are the shortest adsorption period and the longest desorption period within a settable period range of the driving unit.

3. The ammonia adsorption tower drive control device according to claim 2, wherein: When the processor transmits an instruction to the drive unit to execute the abnormal situation handling process, the instruction includes a parameter based on at least one of the following: The minimum adsorption flow rate and the maximum desorption flow rate that can be driven by the driving unit; The lowest adsorption temperature and the highest desorption temperature that can be driven by the driving unit; as well as The driving unit can drive the minimum adsorption pressure and the maximum desorption pressure.

4. The ammonia adsorption tower drive control device according to claim 2, wherein: The processor is configured to repeatedly self-diagnose at a preset cycle whether the sensing data measured by the sensor unit exceeds an optimal range or whether a pattern of previous sensing data and a pattern of subsequent sensing data are different.

5. The ammonia adsorption tower drive control device according to claim 4, wherein: After driving the adsorption tower according to the abnormal situation processing process, when the sensing data measured by the sensor unit is within the optimal range, an instruction to drive the adsorption tower using the adsorption cycle and desorption cycle output from the learned artificial intelligence model is transmitted to the driving unit.

6. The ammonia adsorption tower drive control device according to claim 5, wherein: The processor is configured to: When the previous adsorption cycle and the previous desorption cycle executed before driving the adsorption tower according to the abnormal situation handling process are different from the next adsorption cycle and the next desorption cycle to be driven by the adsorption tower after the abnormal situation handling process is completed, the cost and time reduction data when the driving unit is driven are calculated using the next adsorption cycle and the next desorption cycle.

7. The ammonia adsorption tower drive control device according to claim 6, wherein: When the calculated cost and time reduction data do not exceed a preset threshold, the processor transmits an instruction to the driving unit to maintain a previous adsorption cycle and a previous desorption cycle.

8. The ammonia adsorption tower drive control device according to claim 1, wherein: In order for the artificial intelligence model to detect an abnormal state, the processor removes sensing data of the abnormal state at each time point of the process and transmits only sensing data of the normal state to the artificial intelligence model.

9. The ammonia adsorption tower drive control device according to claim 8, further comprising: A user interface is provided to provide an alert to a user when the abnormal state occurs and to receive information about the abnormal situation handling process.

10. The ammonia adsorption tower drive control device according to claim 4, wherein: The sensor unit includes at least one sensor among a pressure sensor, a temperature sensor, a humidity sensor, a gas sensor, and an adsorbent sensor, When an error in the at least one sensor is detected through self-diagnosis, or a communication error between the at least one sensor and the processor is detected, the processor transmits an instruction to stop driving the adsorption tower to the driving unit.

11. An ammonia adsorption tower drive control method, which is executed by a computing device, the computing device comprising a memory storing one or more instructions and a processor executing the one or more instructions stored in the memory, the ammonia adsorption tower drive control method comprising: the step of collecting sensor data measuring the internal state of the adsorption tower; as well as a step of driving the adsorption tower according to an adsorption period and a desorption period set based on the sensing data, The step of driving the adsorption tower further comprises: The step of outputting corresponding adsorption cycles and desorption cycles according to the sensing data using the learned artificial intelligence model; When the sensing data is within a preset optimal range, a step of transmitting to a driving unit of the adsorption tower an instruction for driving the adsorption tower according to the output adsorption cycle and desorption cycle; and When the sensing data exceeds the optimal range, a step of transmitting an instruction for executing an abnormal situation processing process according to a preset abnormal adsorption cycle and abnormal desorption cycle to a driving unit of the adsorption tower.

Images