High-efficiency acetic acid solution storage and control system

Abstract

The invention discloses a high-efficiency acetic acid solution storage and control system, and relates to the technical field of chemical storage and control systems.The high-efficiency acetic acid solution storage and control system comprises a monitoring module, a classification module, a data processing module, an execution module and an alarm module; different data processing units in the data processing module are used for processing to obtain corresponding execution schemes, and the execution module is arranged to execute the obtained execution schemes, so that appropriate execution schemes can be effectively provided for various problems in the acetic acid solution storage process. By setting an adjustment value, when the pressure exceeds the adjustment value, whether the reason of pressure generation is caused by temperature and liquid level changes or not is analyzed, if not, it is indicated that the reason of pressure abnormity may be abnormity caused by mechanical problems such as small cracks of a storage container, and early warning information can be sent to remind manual inspection; small problems are prevented from being developed into large problems, and the possibility of serious consequences is reduced.

Description

Technical Field

[0001] This invention relates to the field of chemical storage and control system technology, specifically to a high-efficiency acetic acid solution storage and control system. Background Technology

[0002] Acetic acid solution is a widely used chemical substance in many fields, playing an important role in food processing, pharmaceutical preparation, and the chemical industry. The main component of this solution is acetic acid (CH3COOH), an organic acid commonly found in natural vinegars and fruit vinegars. In food processing, acetic acid is widely used as a condiment and food preservative. It not only imparts a unique sour taste to food but also inhibits the growth of bacteria and fungi, extending the shelf life of food. In drug synthesis and preparation, acetic acid is often used as a catalyst or solvent. It plays a crucial role in organic synthesis, used to synthesize drugs and pharmaceutical intermediates. Acetic acid is also an important raw material for many organic synthesis reactions, widely used in the preparation of acetate esters, aluminum acetate, and other chemicals. It is also a solvent in many industrial processes.

[0003] While acetic acid solutions play a crucial role in numerous applications, their storage and handling present several challenges. Acetic acid is corrosive and can damage some materials, necessitating careful selection of storage equipment and piping. Furthermore, acetic acid poses certain hazards; leaks during storage can lead to serious safety incidents. Therefore, monitoring of relevant data is often required during acetic acid solution storage. Existing technologies typically monitor data such as temperature, liquid level, and pressure, and are equipped with corresponding control functions.

[0004] However, existing monitoring and control equipment is often unconnected. When abnormal pressure data is detected, the nitrogen sealing system is often used immediately for adjustment without analysis. This makes it impossible to determine whether the pressure abnormality is caused by changes in temperature and liquid level, and consequently, the pressure cannot be adjusted by regulating temperature and liquid level. As a result, the cause of the pressure abnormality remains unresolved. The nitrogen sealing system needs to operate continuously, which greatly increases nitrogen consumption and reduces the lifespan of the nitrogen sealing system. Furthermore, directly using the nitrogen sealing system for adjustment without analysis may mask pressure abnormalities caused by mechanical problems such as small cracks in storage containers, causing small problems to develop into big problems and resulting in serious consequences. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide a high-efficiency acetic acid solution storage and control system to solve the problems in the existing solutions.

[0006] The objective of this invention can be achieved through the following technical solution: a high-efficiency acetic acid solution storage and control system, comprising:

[0007] The monitoring module is used to collect environmental data during the storage of acetic acid solution at intervals T and send the collected environmental data to the classification module. The environmental data includes important data and secondary data. The important data includes pressure data, and the secondary data includes liquid level data and solution temperature data.

[0008] The classification module is used to classify environmental data after receiving it. The classification types include Class 1, Class 2, Class 3, Class 4 and Class 5 data, and then send the classified data to the data processing module.

[0009] The data processing module includes a type 1 data processing unit, a type 2 data processing unit, a type 3 data processing unit, a type 4 data processing unit, and a type 5 data processing unit, which are used to process the received data of the corresponding type.

[0010] The control module is used to control the storage process of acetic acid solution by executing the received execution plan. It includes a liquid level control unit, a pressure control unit, and a temperature control unit. The liquid level control unit adjusts the liquid level through input and output devices. The pressure control unit adjusts the pressure by charging and discharging nitrogen. The temperature control unit adjusts the temperature through heating and cooling devices.

[0011] The alarm module is used to send received alarm and warning information to the corresponding user terminal and to store the received alarm and warning information.

[0012] Furthermore, the type 1 data processing unit is used to process type 1 data, and the method for processing type 1 data is as follows: sending a preset execution plan to the control unit and generating alarm information and sending it to the alarm unit;

[0013] The two types of data processing units are used to process two types of data. The method for processing the two types of data is as follows: obtain the corresponding execution plan based on the secondary data, send the execution plan to the execution unit, and generate alarm information and send it to the alarm unit.

[0014] The three types of data processing units are used to process three types of data. The method for processing the three types of data is to determine whether the pressure data change is related to the secondary data. If so, an execution plan is generated through a pre-trained execution strategy model and the execution plan is sent to the control unit. Otherwise, an early warning information is generated and sent to the alarm unit.

[0015] The four data processing units are used to process four types of data. The method for processing the four types of data is to generate an execution plan through a pre-trained execution strategy model and generate a corresponding control plan based on the excess value and send it to the execution unit.

[0016] Five types of data processing units are used to process five types of data. The method for processing the five types of data is to timestamp the data and store it as historical data.

[0017] Furthermore, the classification of the collected environmental data includes:

[0018] S1. Establish data standards, which include ideal values, adjustment values, and alarm values ​​for each data item;

[0019] S2. Based on the data standard, determine whether the key data exceeds the corresponding alarm value. If the key data exceeds the corresponding alarm value, classify the data into category 1 data and end the classification. Otherwise, proceed to step S3.

[0020] S3. Based on the data standard, determine whether there is any minor data exceeding the corresponding alarm value. If there is minor data exceeding the corresponding alarm value, classify the data into two categories and end the classification. Otherwise, proceed to step S4.

[0021] S4. Based on the data standard, determine whether the key data exceeds the corresponding adjustment value. If the key data exceeds the corresponding adjustment value, classify the data into 3 categories and end the classification. Otherwise, proceed to step S5.

[0022] S5. Based on the data standard, determine whether there is any minor data that exceeds the corresponding adjustment value. If there is minor data that exceeds the corresponding adjustment value, classify the data into 4 categories and end the classification. Otherwise, proceed to step S6.

[0023] S6. Divide the data into 5 categories and end the classification.

[0024] Furthermore, the preset execution plan is to start the nitrogen blanketing system for a duration of T; the alarm message indicates that the equipment has malfunctioned.

[0025] The step of obtaining the corresponding execution plan based on secondary data includes:

[0026] When the temperature data exceeds the positive alarm value, the execution plan is to adjust the opening degree of the heating equipment to 0%, adjust the opening degree of the cooling equipment to 100%, and set the execution time to T.

[0027] When the temperature data is less than the negative alarm value, the execution plan is to adjust the opening degree of the cooling equipment to 0%, adjust the opening degree of the heating equipment to 100%, and set the execution time to T;

[0028] When the liquid level data is greater than the positive alarm value, the execution plan is to adjust the opening degree of the input device to 0%, adjust the opening degree of the output device to 100%, and set the execution time to T;

[0029] When the liquid level data is less than the negative alarm value, the execution plan is to adjust the opening degree of the output device to 0%, adjust the opening degree of the input device to 100%, and set the execution time to T;

[0030] By setting alarm values, when various data reach the alarm value, the corresponding execution plan is immediately implemented, and all adjustment equipment is activated to reduce the possibility of major safety accidents. At the same time, alarm information is sent to remind manual handling.

[0031] The opening degree indicates the operating status of the equipment. An opening degree of 0% indicates that the equipment is off, and an opening degree of 100% indicates that the equipment is operating at full capacity. The time T is the maximum time limit for the required staff to be on-site to handle the situation.

[0032] Furthermore, the warning information indicates that the equipment may malfunction, and the method for determining whether the pressure data change is related to secondary data includes:

[0033] Retrieve the most recent liquid level data h1, temperature data t1, and pressure data p1 from historical data;

[0034] Obtain the liquid level data h2, temperature data t2, and pressure data p2 for this test.

[0035] The theoretical pressure change is calculated using the following formula:

[0036]

[0037] In the formula, ΔP represents the theoretical pressure change, ρ is the density of the acetic acid solution, and g is the gravitational acceleration.

[0038] The coefficient of variation, B, is calculated using the following formula:

[0039]

[0040] If B ≥ 0.8, the result is yes; otherwise, the result is no.

[0041] By setting an adjustment value, when the pressure exceeds the adjustment value, the system analyzes whether the pressure is caused by changes in temperature and liquid level. If so, it generates a corresponding execution plan to adjust the temperature and liquid level, thereby indirectly adjusting the pressure value to the normal range. This avoids directly using the nitrogen sealing system, saving nitrogen resources and extending the service life of the nitrogen sealing system. If not, it indicates that the abnormal pressure may be caused by mechanical problems such as small cracks in the storage container. In this case, the invention sends an early warning message to remind manual inspection, preventing small problems from developing into big problems and reducing the possibility of serious consequences.

[0042] Furthermore, the pre-trained execution strategy model is a machine learning model, and the training method of the machine learning model is as follows:

[0043] The data on adjustment requirements and the corresponding execution strategies for achieving those adjustment requirements will be used as the sample set.

[0044] The sample set is divided into a training set and a test set. A convolutional neural network model is constructed with historical adjustment demand data as the input layer and execution strategy data as the output layer.

[0045] The convolutional neural network model is trained using the training set to obtain the initial convolutional network learning model;

[0046] The initial machine learning model was tested using a test set to obtain the accuracy of the convolutional neural network model.

[0047] Repeatedly train the convolutional neural network until the accuracy of the convolutional neural network model is greater than the preset accuracy threshold.

[0048] A trained convolutional neural network model is obtained, which is used to obtain an execution strategy based on the adjustment requirement data.

[0049] By using machine learning technology to obtain execution plans through pre-trained models, and then controlling the operation of the liquid level and temperature control units according to the execution plans, the accuracy of temperature and liquid level control during the storage process can be effectively improved. This can ensure that the temperature and liquid level are kept within a reasonable range to the greatest extent possible when there are no equipment failures such as damage or corrosion.

[0050] Furthermore, the adjustment requirement data includes temperature adjustment, liquid level adjustment, liquid level adjustment unit data, and temperature adjustment unit data. The liquid level adjustment unit data includes the opening degree of the input device and the opening degree of the output device. The temperature adjustment unit data includes the opening degree of the heating device and the opening degree of the cooling device. The execution strategy data includes liquid level unit execution strategy data and temperature adjustment unit execution data. The liquid level unit execution strategy data includes execution time, input device adjustment degree, and output device adjustment degree. The temperature adjustment unit execution data includes execution time, heating device adjustment degree, and cooling device adjustment degree. The adjustment degree is used to adjust the opening degree.

[0051] Furthermore, the ratio of the training set to the test set is 8:2, and the convolutional neural network model includes convolutional layers, pooling layers, LSTM layers, and fully connected layers; the convolutional neural network model is compiled, and the optimizer and loss function optimizer are configured using the Adam algorithm; the Adam algorithm is based on optimizing stochastic gradient descent and adaptively adjusts the learning rate;

[0052] The training of the convolutional neural network model using the training set includes:

[0053] The training set is input into the convolutional neural network model in batches. The convolutional neural network model takes the adjustment requirement data in the training set as input and outputs a batch of execution policy data. The mean squared error function value of this batch of execution policy data and the corresponding execution policy data in the training set is calculated. Using the backpropagation algorithm, the loss error is propagated back along the model computation graph in the reverse direction to update the parameters of each layer of the convolutional neural network model. The mean squared error function is expressed as:

[0054]

[0055] Where z is the sample size; Y pred,i Y represents the data for the i-th execution strategy predicted by the model. true,i This represents the execution strategy data corresponding to the i-th sample.

[0056] Furthermore, the generation of execution schemes through pre-trained execution strategy models in the three types of processing units includes:

[0057] The liquid level Δh is calculated using the formula Δh = h2 - h1, and the temperature is calculated using the formula Δt = t2 - t1.

[0058] Obtain the current data from the liquid level control unit and the temperature control unit;

[0059] Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data.

[0060] The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy.

[0061] The aforementioned execution strategy is used as the execution plan.

[0062] The generation of execution plans through pre-trained execution strategy models in the four types of processing units includes:

[0063] The liquid level Δh is calculated using the formula Δh = h2 - h0, and the temperature is calculated using the formula Δt = t2 - t0, where h0 represents the preset ideal liquid level value and t0 represents the preset ideal temperature value.

[0064] Obtain the current data from the liquid level control unit and the temperature control unit;

[0065] Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data.

[0066] The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy.

[0067] The aforementioned execution strategy is used as the execution plan.

[0068] This invention provides a high-efficiency acetic acid solution storage and control system. It has the following beneficial effects:

[0069] 1. By setting an adjustment value, when the pressure exceeds the adjustment value, the system analyzes whether the pressure is caused by changes in temperature and liquid level. If so, it generates a corresponding execution plan to adjust the temperature and liquid level, thereby indirectly adjusting the pressure value to the normal range. This avoids directly using the nitrogen sealing system, saving nitrogen resources and extending the service life of the nitrogen sealing system. If not, it indicates that the abnormal pressure may be caused by mechanical problems such as small cracks in the storage container. This invention will send a warning message to remind manual inspection, preventing small problems from developing into big problems and reducing the possibility of serious consequences.

[0070] 2. By setting alarm values, when various data reach the alarm value, the corresponding execution plan will be executed immediately, and all adjustment equipment will be activated to reduce the possibility of major safety accidents. At the same time, alarm information will be sent to remind manual handling.

[0071] 3. By using machine learning technology to obtain execution plans through pre-trained models, and controlling the operation of liquid level and temperature control units through execution plans, the accuracy of temperature and liquid level control during the storage process can be effectively improved. This can ensure that the temperature and liquid level are kept within a reasonable range to the greatest extent possible when there are no equipment failures such as damage or corrosion. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of a high-efficiency acetic acid solution storage and control system according to the present invention.

[0073] Figure 2 This is a schematic diagram of the classification method of a high-efficiency acetic acid solution storage and control system according to the present invention;

[0074] Figure 3 This is a schematic diagram illustrating the process of generating an execution plan through a pre-trained execution strategy model in a high-efficiency acetic acid solution storage and control system of the present invention. Detailed Implementation

[0075] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] Example 1

[0077] Please refer to Figure 1-3 This invention provides a high-efficiency acetic acid solution storage and control system, comprising:

[0078] The monitoring module is used to collect environmental data during the storage of acetic acid solution at intervals T and send the collected environmental data to the classification module. The environmental data includes important data and secondary data. The important data includes pressure data, and the secondary data includes liquid level data and solution temperature data.

[0079] In this example, the interval T can be set to ten minutes, indicating that the system collects environmental data every ten minutes and takes corresponding subsequent actions based on the environmental data. The pressure data represents the pressure value inside the acetic acid solution storage container. The environmental data is collected by sensors. The temperature data represents the temperature inside the acetic acid solution storage container. The liquid level data represents the height of the acetic acid solution inside the storage container.

[0080] During the storage of acetic acid solution, the pressure value in the storage container is an important data point. Abnormal pressure values ​​may damage the storage container, leading to leakage of the acetic acid solution and causing safety accidents. Therefore, this invention monitors pressure data as an important data point. Liquid level data and solution temperature data are related to pressure data. For example, an increase in liquid level or an increase in temperature may cause an increase in pressure. Moreover, liquid level and temperature are easy to control during the storage process. Therefore, this invention monitors liquid level and temperature as secondary data points.

[0081] The classification module is used to classify the collected environmental data. The classification types include Class 1, Class 2, Class 3, Class 4 and Class 5 data, and then send the classified data to the data processing module.

[0082] The collected environmental data is categorized as follows:

[0083] S1. Establish data standards, which include ideal values, adjustment values, and alarm values ​​for each data item;

[0084] The ideal, adjustment, and alarm values ​​for each data point are all preset by staff. The adjustment values ​​include positive adjustment values ​​that are greater than the ideal value and negative adjustment values ​​that are less than the ideal value. The alarm values ​​include positive alarm values ​​that are greater than the positive adjustment values ​​and negative alarm values ​​that are less than the negative adjustment values.

[0085] In this embodiment, the ideal value for pressure data can be the median value of the storage container's pressure-bearing capacity; the positive adjustment value can be 60% of the storage container's maximum positive pressure tolerance; the positive alarm value can be 80% of the storage container's maximum positive pressure tolerance; the negative adjustment value can be 60% of the storage container's maximum negative pressure tolerance; and the negative alarm value can be 80% of the storage container's maximum negative pressure tolerance. Similarly, the ideal value for liquid level data can be 60% of the storage container's height; the positive adjustment value can be 80% of the storage container's internal height; the positive alarm value can be 90% of the storage container's internal height; the negative adjustment value can be 40% of the storage container's internal height; and the negative alarm value can be 20% of the storage container's internal height. Finally, the ideal value for temperature data can be 25 degrees Celsius; the positive adjustment value can be 30 degrees Celsius; the positive alarm value can be 38 degrees Celsius; the negative adjustment value can be 20 degrees Celsius; and the negative alarm value can be 18 degrees Celsius.

[0086] S2. Based on the data standard, determine whether the key data exceeds the corresponding alarm value. If the key data exceeds the corresponding alarm value, classify the data into category 1 data and end the classification. Otherwise, proceed to step S3.

[0087] S3. Based on the data standard, determine whether there is any minor data exceeding the corresponding alarm value. If there is minor data exceeding the corresponding alarm value, classify the data into two categories and end the classification. Otherwise, proceed to step S4.

[0088] S4. Based on the data standard, determine whether the key data exceeds the corresponding adjustment value. If the key data exceeds the corresponding adjustment value, classify the data into 3 categories and end the classification. Otherwise, proceed to step S5.

[0089] S5. Based on the data standard, determine whether there is any minor data that exceeds the corresponding adjustment value. If there is minor data that exceeds the corresponding adjustment value, classify the data into 4 categories and end the classification. Otherwise, proceed to step S6.

[0090] S6. Divide the data into 5 categories and end the classification.

[0091] The data processing module includes a data processing unit of type 1, a data processing unit of type 2, a data processing unit of type 3, a data processing unit of type 4, and a data processing unit of type 5. The data processing units of type 1, type 2, type 3, type 4, and type 5 are used to process the received data of the corresponding type.

[0092] A type 1 data processing unit is used to process type 1 data. The method for processing type 1 data is as follows: send the preset execution plan to the control unit and generate alarm information and send it to the alarm unit.

[0093] The two types of data processing units are used to process two types of data. The method for processing the two types of data is as follows: obtain the corresponding execution plan based on the secondary data, send the execution plan to the execution unit, and generate alarm information and send it to the alarm unit.

[0094] The preset execution plan is to start the nitrogen blanketing system for a duration of T; an alarm message indicates that the equipment has malfunctioned.

[0095] The execution plan obtained based on secondary data includes:

[0096] When the temperature data exceeds the positive alarm value, the execution plan is to adjust the opening degree of the heating equipment to 0%, adjust the opening degree of the cooling equipment to 100%, and set the execution time to T.

[0097] When the temperature data is less than the negative alarm value, the execution plan is to adjust the opening degree of the cooling equipment to 0%, adjust the opening degree of the heating equipment to 100%, and set the execution time to T;

[0098] When the liquid level data is greater than the positive alarm value, the execution plan is to adjust the opening degree of the input device to 0%, adjust the opening degree of the output device to 100%, and set the execution time to T;

[0099] When the liquid level data is less than the negative alarm value, the execution plan is to adjust the opening degree of the output device to 0%, adjust the opening degree of the input device to 100%, and set the execution time to T;

[0100] The opening degree indicates the operating status of the equipment. An opening degree of 0% indicates that the equipment is closed, and an opening degree of 100% indicates that the equipment is operating at full capacity. The time T is the maximum time limit for the required staff to arrive on site to handle the situation.

[0101] Because this invention has an adjustment function, when the data is classified as Class 1 or Class 2, it usually indicates that the storage device has experienced abnormalities such as corrosion or damage. At this time, it is necessary to fully activate all adjustment devices to reduce the possibility of a major safety accident, and at the same time, it is necessary to send alarm information to remind manual handling.

[0102] In this example, the nitrogen sealing system is a prior art technology that uses nitrogen to regulate the internal pressure of the storage container. In this example, the heating equipment, cooling equipment, input equipment and output equipment can all adopt prior art. The heating equipment can be hot water pipe heating, the cooling equipment can be spray cooling equipment, and the input and output equipment can be a combination of pipeline equipment and pump equipment.

[0103] The three data processing units are used to process three types of data. The method for processing the three types of data is to determine whether the pressure data change is related to the secondary data. If so, an execution plan is generated through the pre-trained execution strategy model and sent to the control unit. Otherwise, an early warning message is generated and sent to the alarm unit.

[0104] Warning messages indicate that equipment may malfunction. Methods for determining whether changes in pressure data are related to minor data include:

[0105] Retrieve the most recent liquid level data h1, temperature data t1, and pressure data p1 from historical data;

[0106] Obtain the liquid level data h2, temperature data t2, and pressure data p2 for this test.

[0107] The theoretical pressure change is calculated using the following formula:

[0108]

[0109] In the formula, ΔP represents the theoretical pressure change, ρ is the density of the acetic acid solution, and g is the gravitational acceleration.

[0110] The formula for calculating the coefficient of variation, B, is as follows:

[0111]

[0112] If B ≥ 0.8, the result is yes; otherwise, the result is no.

[0113] The three types of data indicate that the pressure exceeds the control range and needs to be adjusted. This invention analyzes temperature and liquid level data to determine whether the pressure abnormality is caused by changes in temperature and liquid level. If so, the pressure is indirectly controlled by adjusting the temperature and liquid level. Otherwise, the pressure abnormality may be caused by equipment failure. Therefore, reminding employees to handle the problem in a timely manner can effectively prevent the problem from escalating.

[0114] The four data processing units are used to process four types of data. The method for processing the four types of data is to obtain the excess values ​​in the secondary data and generate the corresponding control scheme based on the excess values ​​and send it to the execution unit.

[0115] The execution plans generated by the pre-trained execution strategy model in the three types of processing units include:

[0116] The liquid level Δh is calculated using the formula Δh = h2 - h1, and the temperature is calculated using the formula Δt = t2 - t1.

[0117] Obtain the current data from the liquid level control unit and the temperature control unit;

[0118] Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data.

[0119] The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy.

[0120] The execution strategy is used as the execution plan.

[0121] The execution plans generated by the pre-trained execution strategy model in the four types of processing units include:

[0122] The liquid level Δh is calculated using the formula Δh = h2 - h0, and the temperature is calculated using the formula Δt = t2 - t0, where h0 represents the preset ideal liquid level value and t0 represents the preset ideal temperature value.

[0123] Obtain the current data from the liquid level control unit and the temperature control unit;

[0124] Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data.

[0125] The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy.

[0126] The execution strategy is used as the execution plan.

[0127] In this embodiment, machine learning technology is used to obtain an execution plan through a pre-trained model. The execution plan is used to regulate the operation of the liquid level regulation unit and the temperature regulation unit, which can effectively improve the accuracy of temperature and liquid level regulation during the storage process. This can ensure that the temperature and liquid level are kept within a reasonable range to the greatest extent possible when there are no equipment failures such as damage or corrosion.

[0128] Five types of data processing units are used to process five types of data. The method for processing the five types of data is to timestamp the data and store the data as historical data.

[0129] The five data points indicate that all data are normal; simply record the data.

[0130] The control module is used to control the storage process of acetic acid solution by executing the received execution plan. It includes a liquid level control unit, a pressure control unit, and a temperature control unit. The liquid level control unit adjusts the liquid level through input and output devices, the pressure control unit adjusts the pressure by charging and discharging nitrogen, and the temperature control unit adjusts the temperature through heating and cooling devices.

[0131] The alarm module is used to send received alarm and warning information to the corresponding user terminal and to store the received alarm and warning information.

[0132] In this example, the user terminal can be a computer, mobile phone, tablet, or other terminal device.

[0133] The pre-trained execution strategy model is a machine learning model, and the training method for the machine learning model is as follows:

[0134] The data on adjustment requirements and the corresponding execution strategies for achieving those adjustment requirements will be used as the sample set.

[0135] The adjustment requirement data includes temperature adjustment, liquid level adjustment, liquid level adjustment unit data, and temperature adjustment unit data. The liquid level adjustment unit data includes the opening degree of the input device and the opening degree of the output device, and the temperature adjustment unit data includes the opening degree of the heating device and the opening degree of the cooling device. The execution strategy data includes liquid level unit execution strategy data and temperature adjustment unit execution data. The liquid level unit execution strategy data includes execution time, input device adjustment degree, and output device adjustment degree, and the temperature adjustment unit execution data includes execution time, heating device adjustment degree, and cooling device adjustment degree. The adjustment degree is used to adjust the opening degree.

[0136] The sample set is divided into a training set and a test set. A convolutional neural network model is constructed with historical adjustment demand data as the input layer and execution strategy data as the output layer.

[0137] The convolutional neural network model is trained using the training set to obtain the initial convolutional network learning model;

[0138] The initial machine learning model was tested using a test set to obtain the accuracy of the convolutional neural network model.

[0139] Repeatedly train the convolutional neural network until the accuracy of the convolutional neural network model is greater than the preset accuracy threshold.

[0140] The training set and test set are divided in an 8:2 ratio. The convolutional neural network model includes convolutional layers, pooling layers, LSTM layers, and fully connected layers. The convolutional neural network model is compiled, and the optimizer and loss function are configured. The Adam algorithm is selected as the optimizer. The Adam algorithm is based on optimizing stochastic gradient descent and adaptively adjusts the learning rate.

[0141] Training a convolutional neural network model using a training set includes:

[0142] The training set is input into the convolutional neural network model in batches. The convolutional neural network model takes the adjustment requirement data in the training set as input and outputs a batch of execution policy data. The mean squared error function value of this batch of execution policy data and the corresponding execution policy data in the training set is calculated. Using the backpropagation algorithm, the loss error is propagated back along the model computation graph in the reverse direction to update the parameters of each layer of the convolutional neural network model. The mean squared error function is expressed as:

[0143]

[0144] Where z is the sample size; Y pred,i Y represents the data for the i-th execution strategy predicted by the model. true,i This represents the execution strategy data corresponding to the i-th sample.

[0145] The trained convolutional neural network model is obtained, and the convolutional neural network model is used to obtain the execution strategy based on the adjustment requirement data.

[0146] In this embodiment, the data used to train the machine model can be real data recorded in history, or data simulated in a theoretical environment through experiments.

[0147] Furthermore, according to embodiments of this application, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, a data communication method for a train operation control system. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0148] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0149] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division of a waterway underwater topography change analysis system and method. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0151] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0152] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0153] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0154] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high performance acetic acid solution storage and control system, characterized by, include: The monitoring module is used to collect environmental data during the storage of acetic acid solution at intervals T and send the collected environmental data to the classification module. The environmental data includes important data and secondary data. The important data includes pressure data, and the secondary data includes liquid level data and solution temperature data. The classification module is used to classify environmental data after receiving it. The classification types include Class 1, Class 2, Class 3, Class 4 and Class 5 data, and then send the classified data to the data processing module. The data processing module includes a type 1 data processing unit, a type 2 data processing unit, a type 3 data processing unit, a type 4 data processing unit, and a type 5 data processing unit, which are used to process the received data of the corresponding type. The control module is used to control the storage process of acetic acid solution by executing the received execution plan. It includes a liquid level control unit, a pressure control unit, and a temperature control unit. The liquid level control unit adjusts the liquid level through input and output devices. The pressure control unit adjusts the pressure by charging and discharging nitrogen. The temperature control unit adjusts the temperature through heating and cooling devices. The alarm module is used to send received alarm and warning information to the corresponding user terminal and to store the received alarm and warning information.

2. The high-efficiency acetic acid solution storage and control system according to claim 1, characterized in that, The first type of data processing unit is used to process a type of data. The method for processing the first type of data is to send a preset execution plan to the control unit and generate alarm information and send it to the alarm unit. The two types of data processing units are used to process two types of data. The method for processing the two types of data is as follows: obtain the corresponding execution plan based on the secondary data, send the execution plan to the execution unit, and generate alarm information and send it to the alarm unit. The three types of data processing units are used to process three types of data. The method for processing the three types of data is to determine whether the pressure data change is related to the secondary data. If so, an execution plan is generated through a pre-trained execution strategy model and the execution plan is sent to the control unit. Otherwise, an early warning information is generated and sent to the alarm unit. The four data processing units are used to process four types of data. The method for processing the four types of data is to generate an execution plan through a pre-trained execution strategy model and generate a corresponding control plan based on the excess value and send it to the execution unit. Five types of data processing units are used to process five types of data. The method for processing the five types of data is to timestamp the data and store it as historical data.

3. The high performance acetic acid solution storage and control system of claim 2, wherein, The classification of the collected environmental data includes: S1. Establish data standards, which include ideal values, adjustment values, and alarm values ​​for each data item; S2. Based on the data standard, determine whether the key data exceeds the corresponding alarm value. If the key data exceeds the corresponding alarm value, classify the data into category 1 data and end the classification. Otherwise, proceed to step S3. S3. Based on the data standard, determine whether there is any minor data exceeding the corresponding alarm value. If there is minor data exceeding the corresponding alarm value, classify the data into two categories and end the classification. Otherwise, proceed to step S4. S4. Based on the data standard, determine whether the key data exceeds the corresponding adjustment value. If the key data exceeds the corresponding adjustment value, classify the data into 3 categories and end the classification. Otherwise, proceed to step S5. S5. Based on the data standard, determine whether there is any minor data that exceeds the corresponding adjustment value. If there is minor data that exceeds the corresponding adjustment value, classify the data into 4 categories and end the classification. Otherwise, proceed to step S6. S6. Divide the data into 5 categories and end the classification.

4. The high performance acetic acid solution storage and control system of claim 3, wherein, The ideal value, adjustment value, and alarm value of each data item are all obtained through preset by staff. The adjustment value includes a positive adjustment value that is greater than the ideal value and a negative adjustment value that is less than the ideal value. The alarm value includes a positive alarm value that is greater than the positive adjustment value and a negative alarm value that is less than the negative adjustment value.

5. The high performance acetic acid solution storage and control system of claim 4, wherein, The preset execution plan is to start the pressure regulating unit for a duration of T. The alarm message indicates that the device has malfunctioned; The step of obtaining the corresponding execution plan based on secondary data includes: When the temperature data exceeds the positive alarm value, the execution plan is to adjust the opening degree of the heating equipment to 0%, adjust the opening degree of the cooling equipment to 100%, and set the execution time to T. When the temperature data is less than the negative alarm value, the execution plan is to adjust the opening degree of the cooling equipment to 0%, adjust the opening degree of the heating equipment to 100%, and set the execution time to T; When the liquid level data is greater than the positive alarm value, the execution plan is to adjust the opening degree of the input device to 0%, adjust the opening degree of the output device to 100%, and set the execution time to T; When the liquid level data is less than the negative alarm value, the execution plan is to adjust the opening degree of the output device to 0%, adjust the opening degree of the input device to 100%, and set the execution time to T; The opening degree indicates the operating status of the equipment. An opening degree of 0% indicates that the equipment is off, and an opening degree of 100% indicates that the equipment is operating at full capacity. The time T is the maximum time limit for the required staff to be on-site to handle the situation.

6. The high performance acetic acid solution storage and control system of claim 5, wherein, The warning message indicates that the equipment may malfunction, and the method for determining whether the pressure data change is related to the secondary data includes: Retrieve the most recent liquid level data h1, temperature data t1, and pressure data p1 from historical data; Obtain the liquid level data h2, temperature data t2, and pressure data p2 for this test. The theoretical pressure change is calculated using the following formula: In the formula, ΔP represents the theoretical pressure change, ρ is the density of the acetic acid solution, and g is the gravitational acceleration. The coefficient of variation, B, is calculated using the following formula: If B ≥ 0.8, the result is yes; otherwise, the result is no.

7. The high performance acetic acid solution storage and control system of claim 6, wherein, The pre-trained execution strategy model is a machine learning model, and the training method for the machine learning model is as follows: The data on adjustment requirements and the corresponding execution strategies for achieving those adjustment requirements will be used as the sample set. The sample set is divided into a training set and a test set. A convolutional neural network model is constructed with historical adjustment demand data as the input layer and execution strategy data as the output layer. The convolutional neural network model is trained using the training set to obtain the initial convolutional network learning model; The initial machine learning model was tested using a test set to obtain the accuracy of the convolutional neural network model. Repeatedly train the convolutional neural network until the accuracy of the convolutional neural network model is greater than the preset accuracy threshold. A trained convolutional neural network model is obtained, which is used to obtain an execution strategy based on the adjustment requirement data.

8. The high performance acetic acid solution storage and control system of claim 7, wherein, The adjustment requirement data includes temperature adjustment, liquid level adjustment, liquid level adjustment unit data, and temperature adjustment unit data. The liquid level control unit data includes the opening degree of the input device and the opening degree of the output device, and the temperature control unit data includes the opening degree of the heating device and the opening degree of the cooling device. The execution strategy data includes liquid level unit execution strategy data and temperature regulation unit execution data; The liquid level unit execution strategy data includes execution time, input device adjustment degree, and output device adjustment degree; The temperature control unit executes data including execution time, heating device adjustment degree, and cooling device adjustment degree; The adjustment degree is used to adjust the opening degree.

9. The high performance acetic acid solution storage and control system of claim 8, wherein, The training set and test set are divided in an 8:2 ratio. The convolutional neural network model includes convolutional layers, pooling layers, LSTM layers, and fully connected layers. The convolutional neural network model is compiled, and the optimizer and loss function are configured using the Adam algorithm. The Adam algorithm is based on optimized stochastic gradient descent and adaptively adjusts the learning rate. The training of the convolutional neural network model using the training set includes: The training set is input into the convolutional neural network model in batches. The convolutional neural network model takes the adjustment requirement data in the training set as input and outputs a batch of execution policy data. The mean squared error function value of this batch of execution policy data and the corresponding execution policy data in the training set is calculated. Using the backpropagation algorithm, the loss error is propagated back along the model computation graph in the reverse direction to update the parameters of each layer of the convolutional neural network model. The mean squared error function is expressed as: wherein z is the number of samples; Y pred,i is the i-th execution strategy data predicted by the model; Y true,i is the execution strategy data corresponding to the i-th sample.

10. The high performance acetic acid solution storage and control system of claim 9, wherein, The generation of execution plans through pre-trained execution strategy models in the three types of processing units includes: The liquid level Δh is calculated using the formula Δh = h2 - h1, and the temperature is calculated using the formula Δt = t2 - t1. Obtain the current data from the liquid level control unit and the temperature control unit; Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data. The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy. The aforementioned execution strategy will be used as the execution plan. The generation of execution plans through pre-trained execution strategy models in the four types of processing units includes: The liquid level Δh is calculated using the formula Δh = h2 - h0, and the temperature is calculated using the formula Δt = t2 - t0, where h0 represents the preset ideal liquid level value and t0 represents the preset ideal temperature value. Obtain the current data from the liquid level control unit and the temperature control unit; Combine the data from the liquid level control unit, the temperature control unit, and the temperature unit into the control demand data. The adjustment demand data is input into the execution strategy model, and the execution strategy model outputs the execution strategy. The aforementioned execution strategy is used as the execution plan.

Images