Method, system and equipment for acquiring intelligent regulating and controlling model for alkalinity of circulating water and medium

Abstract

The invention belongs to the technical field of thermal power plant circulating cooling water treatment, and relates to a method, a system, equipment and a medium for acquiring a circulating water alkalinity intelligent regulation and control model. According to the method, the model is constructed by combining the water quality data of the circulating water return pipeline, the acid liquor adding amount data of the cooling tower pool, the season data, the thermal power plant working condition data and the set alkalinity expected value data with the long-short-term memory network, and the data covers full-dimensional dynamic factors influencing the alkalinity, so that the accuracy of the model is ensured; the accuracy of controlling the alkalinity of the circulating water is favorably improved. Through a small number of sensors and automatic data acquisition and processing, automatic control over the alkalinity of the circulating water can be achieved, the maintenance cost is reduced, real-time monitoring and control over the alkalinity of the circulating water can be achieved, and the safety and economical efficiency of a circulating water system in the operation process can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of circulating cooling water treatment in thermal power plants, and relates to a method, system, device and medium for obtaining an intelligent regulation model of circulating water alkalinity. Background Art

[0002] During the operation of the circulating water system in a thermal power plant, alkalinity control is the core link to maintain water quality stability and prevent scaling and corrosion. At present, the traditional control methods have the following problems: First, relying on manual experience or adding acid at fixed intervals, it cannot respond to water quality fluctuations in real time, resulting in an increased risk of scaling or corrosion; Second, the control cost is relatively high. The fixed-frequency acid addition strategy is prone to over-addition or under-addition of acid solution, increasing the chemical agent cost. Manual inspection and manual adjustment increase the workload of maintenance personnel, and the maintenance cost is relatively high.

[0003] In the prior art, pH meters, alkalinity meters and flow meters are installed on the circulating water make-up water pipeline and the circulating water drainage pipeline, and the alkalinity of the circulating water system is intelligently controlled by adjusting the regulating valves of the circulating water make-up water pipeline, the circulating water drainage pipeline and the sulfuric acid regulating valve. Since multiple devices are scattered at different pipeline nodes, a large amount of manpower and material resources are still required for equipment calibration, fault troubleshooting and replacement. Moreover, the highly corrosive water quality environment is likely to shorten the service life of sensors. The more sensors there are, the higher the maintenance cost, further driving up the maintenance cost.

[0004] In summary, in the existing circulating water alkalinity control methods, there are many and scattered devices, which is not conducive to later maintenance. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, system, device and medium for obtaining an intelligent regulation model of circulating water alkalinity, so as to solve the technical problem that there are many and scattered devices, which is not conducive to later maintenance.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a method for obtaining an intelligent regulation model of circulating water alkalinity, including the following steps: Obtain initial data, where the initial data includes water quality data of the circulating water return pipeline, acid addition amount data of the cooling tower basin, seasonal data, thermal power plant operating conditions data and set alkalinity expected value data; Establish a circulating water alkalinity law data set and a quantitative influence factor data set according to the initial data; Construct a model using a long short-term memory network based on the circulating water alkalinity law data set and the quantitative influence factor data set to obtain an initial intelligent regulation model of circulating water alkalinity; Annotate the water quality data, perform data augmentation on the annotated water quality data through random time-domain transformation, and use the augmented data to train the initial intelligent control model for circulating water alkalinity to obtain the optimal intelligent control model for circulating water alkalinity. The intelligent control model for circulating water alkalinity is used for regulating the circulating water alkalinity.

[0007] In a second aspect, the present invention provides a method for controlling the circulating water alkalinity based on alkalinity pH measurement, including the following steps: Obtain initial data, where the initial data includes water quality data of the circulating water return pipeline, acid addition amount data of the cooling tower basin, seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data; According to the seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data, divide the preprocessed data into a suitable alkalinity range, a high alkalinity range, and a low alkalinity range to obtain a quantified influencing factor data set; Use time series to align the data in the quantified influencing factor data set, seasonal data, thermal power plant operating conditions data, and set alkalinity expectation data to obtain a multi-source data set; Based on the intelligent control model for circulating water alkalinity, use natural language processing technology to obtain the predicted value of the circulating water alkalinity and the amount of acid solution to be added based on the multi-source data set; Adjust the acid addition amount of the cooling tower basin according to the amount of acid solution to be added.

[0008] In a third aspect, the present invention provides a system for obtaining an intelligent control model for circulating water alkalinity, including: Data acquisition module: used to obtain initial data, where the initial data includes water quality data of the circulating water return pipeline, acid addition amount data of the cooling tower basin, seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data; Data set construction module: used to establish a circulating water alkalinity law data set and a quantified influencing factor data set according to the initial data; Model construction module: used to construct a model using a long short-term memory network based on the circulating water alkalinity law data set and the quantified influencing factor data set to obtain an initial intelligent control model for circulating water alkalinity; Model training module: used to annotate the water quality data, perform data augmentation on the annotated water quality data through random time-domain transformation, and use the augmented data to train the initial intelligent control model for circulating water alkalinity to obtain the optimal intelligent control model for circulating water alkalinity. The intelligent control model for circulating water alkalinity is used for regulating the circulating water alkalinity.

[0009] In a fourth aspect, the present invention provides an electronic device, including: a processor; a memory for storing computer program instructions; and when executing the computer program, it realizes the steps of the method for obtaining an intelligent control model for circulating water alkalinity.

[0010] In a fifth aspect, the present invention provides a storage medium storing computer program instructions. When the computer program instructions are loaded and run by a processor, the processor executes a method for obtaining an intelligent regulation model for circulating water alkalinity.

[0011] Compared with the prior art, the present invention has the following beneficial effects: The method of the present invention constructs a model by combining water quality data of the circulating water return pipeline, acid liquid addition amount data of the cooling tower basin, seasonal data, thermal power plant operating conditions data, and set alkalinity expected value data with a long short-term memory network. The data covers all-dimensional dynamic factors affecting alkalinity, which is beneficial to ensuring the accuracy of the model and improving the accuracy of circulating water alkalinity control.

[0012] The method of the present invention can effectively improve the accuracy of circulating water alkalinity control by integrating various different types of data such as water quality data, acid liquid addition amount data, and set alkalinity expected value data. Through a small number of sensors and automated data collection and processing, automatic control of circulating water alkalinity can be achieved, which is beneficial to later maintenance, reduces maintenance costs, can realize real-time monitoring and control of circulating water alkalinity, and helps to improve the safety and economy during the operation of the circulating water system.

[0013] The system of the present invention includes a data acquisition module, a data set construction module, a model construction module, and a model training module. The data acquisition module is used to acquire initial data, and the initial data includes water quality data of the circulating water return pipeline, acid liquid addition amount data of the cooling tower basin, seasonal data, thermal power plant operating conditions data, and set alkalinity expected value data; the data set construction module is used to establish a circulating water alkalinity rule data set and a quantitative influence factor data set according to the initial data; the model construction module is used to construct a model using a long short-term memory network based on the circulating water alkalinity rule data set and the quantitative influence factor data set to obtain an initial intelligent regulation model for circulating water alkalinity; the model training module is used to label the water quality data, perform data augmentation on the labeled water quality data through random time-domain transformation, and use the augmented data to train the initial intelligent regulation model for circulating water alkalinity to obtain an optimal intelligent regulation model for circulating water alkalinity. The intelligent regulation model for circulating water alkalinity is used for circulating water alkalinity regulation. Each module cooperates with each other, which is beneficial to ensuring the accuracy of the model and improving the accuracy of circulating water alkalinity control.

[0014] The electronic device and storage medium of the present invention can also ensure the accuracy of the model and are beneficial to improving the accuracy of circulating water alkalinity control. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a control system module diagram of an embodiment of the present invention; Figure 2Schematic diagram of the overall structure of the embodiment of the present invention; Figure 3 Logic block diagram of the embodiment of the present invention; Figure 4 Flowchart of the method for obtaining the intelligent regulation model of circulating water alkalinity according to the embodiment of the present invention; Figure 5 Flowchart of the circulating water alkalinity control method according to the embodiment of the present invention; Figure 6 System module diagram of the embodiment of the present invention.

[0016] Wherein: 1. Cooling tower basin; 2. Cooling tower basin liquid level gauge; 3. Sulfuric acid storage tank; 4. Sulfuric acid pipeline inlet valve; 5. Variable frequency booster pump inlet valve; 6. Variable frequency booster pump; 7. Variable frequency booster pump outlet valve; 8. Variable frequency booster pump bypass valve; 9. Sulfuric acid regulating valve; 10. Sulfuric acid regulating valve bypass valve; 11. Sulfuric acid pipeline flowmeter; 12. Circulating water return pipeline; 13. Circulating water sampling pipeline; 14. Intelligent regulating device for circulating water alkalinity; 15. Sewage pipeline; 16. Circulating water sampling valve; 17. Circulating water online pH meter; 18. Circulating water online alkalinity meter; 19. Circulating water sampling valve bypass valve; 20. Controller and actuator; 21. Display and input interface. Detailed implementation manners

[0017] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] It should be noted that the terms "first", "second", etc. in the specification of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0019] The present invention will be further described in detail below with reference to the accompanying drawings: See Figure 4, this embodiment discloses a method for obtaining an intelligent control model of circulating water alkalinity, including the following steps: S1, obtain initial data, and the obtained initial data includes water quality data of the circulating water return pipeline 12, acid liquid addition amount data of the cooling tower basin 1, seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data. The data covers all-dimensional dynamic factors affecting alkalinity, which is beneficial to ensuring the accuracy of the model.

[0020] Preferably, the water quality data of the circulating water return pipeline 12 includes: circulating water pH data, circulating water alkalinity data, and liquid level data of the cooling tower basin 1.

[0021] Preferably, the acid liquid addition amount data of the cooling tower basin 1 includes: operating current data of the variable frequency booster pump 6, opening data of the sulfuric acid regulating valve 9, and data of the sulfuric acid pipeline flowmeter 11.

[0022] S2, establish a circulating water alkalinity rule data set and a quantitative influence factor data set according to the initial data. By integrating multi-source heterogeneous data, the problem of time series misalignment caused by sensor acquisition delay is eliminated, and data consistency is improved. In addition, through long-term data accumulation, covering different seasons and load conditions, the universal rules of dynamic alkalinity change are extracted, such as the accelerated decline of alkalinity caused by high temperature in summer, which is beneficial to enhancing the adaptability of the model to complex conditions. Specifically as follows: Use the normalization method to preprocess the water quality data and the acid liquid addition amount data to obtain preprocessed data with the same dimension; Preferably, the specific calculation formula of the normalization is as follows: X norm = ( X - μ ) / σ Among them, X norm is the preprocessed data after normalization, X is the original data value, and the original data value includes water quality data and acid liquid addition amount data, μ is the mean value of the corresponding original data group value, σ is the standard deviation of the original data group value.

[0023] Establish a circulating water alkalinity rule data set that correlates water quality data, acid liquid addition amount data, seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data according to seasonal data, thermal power plant operating conditions data, set alkalinity expectation value data, and preprocessed data; Divide the preprocessed data into a suitable alkalinity range, a high alkalinity range, and a low alkalinity range according to seasonal data, thermal power plant operating conditions data, and set alkalinity expectation value data to obtain a quantitative influence factor data set.

[0024] S3. Based on the circulating water alkalinity rule dataset and the quantified influencing factor dataset, a long short-term memory network is used to build a model to obtain an initial intelligent regulation model for circulating water alkalinity. The hidden state update mechanism of the LSTM can adaptively adjust the contribution weights of the power plant operating conditions data and the seasonal data (for example, the evaporation rate of the cooling tower decreases at low temperatures in winter, weakening the influence of seasonal data), improving the physical rationality of the control strategy; Preferably, the intelligent regulation model for circulating water alkalinity includes: An input layer that accepts a continuous sequence of water quality datasets; A normalization layer that accelerates the training process by batch-normalizing the water quality data; The first layer of LSTM that accepts the normalized water quality dataset and processes the time series data to capture the dynamic changes over time; The second layer of LSTM that further processes the output of the first layer of LSTM, deepening the model's understanding of the time series and capturing more complex time-dependent relationships; A fully connected layer that maps the fused spatio-temporal features to specific alkalinity values; An output layer that outputs the alkalinity values at the set future time steps and the required amount of acid solution to be added.

[0025] S4. Label the water quality data, perform data augmentation on the labeled water quality data through random time domain transformation, and use the augmented data to train the initial intelligent regulation model for circulating water alkalinity to obtain the optimal intelligent regulation model for circulating water alkalinity. The intelligent regulation model for circulating water alkalinity is used for regulating the circulating water alkalinity. Through random time domain transformation, such as time window translation and local scaling of the water quality data sequence, disturbances such as sensor noise and communication delay in the actual system are simulated to enhance the robustness of the model.

[0026] Based on the above method, this embodiment also discloses a method for controlling the circulating water alkalinity based on alkalinity pH measurement. See Figure 5 , including the following steps: S10. Obtain initial data. The initial data includes the water quality data of the circulating water return pipe 12, the data of the acid solution addition amount in the cooling tower sump 1, the seasonal data, the power plant operating conditions data, and the set alkalinity expected value data. The data covers all-dimensional dynamic factors affecting alkalinity, which is beneficial to ensuring the accuracy of the regulation result.

[0027] S20. According to the seasonal data, the power plant operating conditions data, and the set alkalinity expected value data, divide the preprocessed data into a suitable alkalinity range, a high alkalinity range, and a low alkalinity range to obtain a quantified influencing factor dataset; S30. Align the data in the quantified influencing factor dataset, seasonal data, thermal power plant operating condition data, and the set expected alkalinity data using time series to obtain a multi-source dataset. S40. Based on the intelligent regulation model for circulating water alkalinity, use natural language processing technology to obtain the predicted value of the circulating water alkalinity and the amount of acid solution to be added based on the multi-source dataset.

[0028] S50. Adjust the amount of acid solution added to the cooling tower basin 1 according to the amount of acid solution to be added.

[0029] The method of the present invention can effectively improve the accuracy of circulating water alkalinity control by integrating various different types of data such as water quality data, acid solution addition amount data, and set expected alkalinity value data. Through a small number of sensors and automated data collection and processing, the automatic control of circulating water alkalinity can be achieved, reducing the maintenance cost. The real-time monitoring and control of circulating water alkalinity can be realized, which helps to improve the safety and economy during the operation of the circulating water system.

[0030] Based on the acquisition method of the intelligent regulation model for circulating water alkalinity, this embodiment also discloses a system for acquiring the intelligent regulation model for circulating water alkalinity. Refer to Figure 6 , including: a data acquisition module, a dataset construction module, a model construction module, and a model training module; Data acquisition module: used to acquire initial data, and the initial data includes the water quality data of the circulating water return pipe 12, the amount of acid solution added to the cooling tower basin 1, seasonal data, thermal power plant operating condition data, and the set expected alkalinity data; Dataset construction module: used to establish a circulating water alkalinity rule dataset that correlates water quality data, acid solution addition amount data, seasonal data, thermal power plant operating condition data, and the set expected alkalinity data; Model construction module: used to construct a model using a long short-term memory network based on the circulating water alkalinity rule dataset to obtain an initial intelligent regulation model for circulating water alkalinity; Model training module: used to label the water quality data, perform data augmentation on the labeled water quality data through random time domain transformation, and train the initial intelligent regulation model for circulating water alkalinity using the augmented data to obtain the optimal intelligent regulation model for circulating water alkalinity, and the intelligent regulation model for circulating water alkalinity is used for circulating water alkalinity regulation.

[0031] Each module of the present invention cooperates with each other, which is beneficial to ensuring the accuracy of the model and improving the accuracy of circulating water alkalinity control.

[0032] Embodiment 2: Refer to Figure 2 and Figure 3, this embodiment discloses a method and system for controlling the alkalinity of circulating water based on alkalinity pH measurement. Refer to Figure 2 , which is part of the circulating water alkalinity control system, including a cooling tower basin 1, a sulfuric acid storage tank 3, a circulating water return pipe 12, a circulating water sampling pipe 13, a circulating water alkalinity intelligent control device 14, and a sewage discharge pipe 15; The outlet of the sulfuric acid storage tank 3 is communicated with the cooling tower basin 1, the circulating water return pipe 12 is communicated with the cooling tower basin 1, a cooling tower basin liquid level gauge 2 is arranged in the cooling tower basin 1, the circulating water return pipe 12 is communicated with the circulating water sampling pipe 13, the circulating water sampling pipe 13 is communicated with the circulating water alkalinity intelligent control device 14, and the sewage discharge pipe 15 is communicated with the cooling tower basin 1.

[0033] Preferably, the circulating water return pipe 12 is communicated with the sewage discharge pipe 15 through the circulating water sampling pipe 13, a circulating water sampling valve 16, a circulating water on-line pH meter 17, and a circulating water on-line alkalinity meter 18, and the circulating water on-line pH meter 17 and the circulating water on-line alkalinity meter 18 are connected in parallel.

[0034] Preferably, the outlet of the sulfuric acid storage tank 3 is successively communicated with the cooling tower basin 1 through an inlet valve 5 of a variable-frequency booster pump, a variable-frequency booster pump 6, an outlet valve 7 of the variable-frequency booster pump, a sulfuric acid regulating valve 9, and a sulfuric acid pipeline flowmeter 11.

[0035] A branch formed by the inlet valve 5 of the variable-frequency booster pump, the variable-frequency booster pump 6, and the outlet valve 7 of the variable-frequency booster pump is connected in parallel with a bypass valve 8 of the variable-frequency booster pump, and the sulfuric acid regulating valve 9 is connected in parallel with a bypass valve 10 of the sulfuric acid regulating valve.

[0036] Preferably, the circulating water return passes through the circulating water sampling valve 16, the circulating water on-line pH meter 17, and the circulating water on-line alkalinity meter 18 and is communicated with the sewage discharge pipe 15, and the circulating water on-line pH meter 17 and the circulating water on-line alkalinity meter 18 are connected in parallel.

[0037] A branch formed by the circulating water sampling valve 16, the circulating water on-line pH meter 17, and the circulating water on-line alkalinity meter 18 is connected in parallel with a bypass valve 19 of the circulating water sampling valve.

[0038] It also includes a controller and an actuator 20, and the controller and the actuator 20 are electrically connected to the variable-frequency booster pump 6, the sulfuric acid regulating valve 9, the sulfuric acid pipeline flowmeter 11, the cooling tower basin liquid level gauge 2, the circulating water on-line pH meter 17, and the circulating water on-line alkalinity meter 18. The controller and the actuator 20 are used to adjust the alkalinity of the circulating water according to the circulating water alkalinity control method of the present invention.

[0039] In this embodiment, the control system includes a control and data processing unit, a water quality data acquisition unit, and an acid addition amount data acquisition unit. The control method includes the following steps: S1. Data collection: Collect water quality data through the water quality data collection unit, and collect the data of the acid addition amount through the acid addition amount data collection unit. Obtain seasonal data, the operating conditions data of the thermal power plant, and the set alkalinity expected value data through the control and data processing unit. The water quality data includes the data of the on-line pH meter 17 of the circulating water, the data of the on-line alkalinity meter 18 of the circulating water, and the data of the cooling tower basin level meter 2. The acid addition amount data collection unit collects the operating current data of the variable frequency booster pump 6, the opening degree data of the sulfuric acid regulating valve 9, and the data of the sulfuric acid pipeline flowmeter 11.

[0040] S2. Data preprocessing: Use the normalization method to preprocess the collected water quality data and acid addition amount data to obtain data with the same dimension, and perform time series segmentation and data set division, specifically as follows: Regarding the specific values of the water quality data and acid addition amount data obtained in step S1, divide them into the appropriate alkalinity range, the high alkalinity range, and the low alkalinity range according to the seasonal data, the operating conditions data of the thermal power plant, and the set alkalinity expected value data to obtain a quantitative influence factor data set; establish a circulating water alkalinity rule data set that correlates the water quality data, the acid addition amount data, the seasonal data, the operating conditions data of the thermal power plant, and the set alkalinity expected value data according to the seasonal data, the operating conditions data of the thermal power plant, and the set alkalinity expected value data obtained in step S1.

[0041] S3. Data fusion: Perform data fusion on the data obtained in steps S1 and S2. Use the long short-term memory network (LSTM) to build a model to obtain a circulating water alkalinity intelligent regulation model. Select a general large model, use the data to fine-tune the model, and use the circulating water alkalinity intelligent regulation model to give the amount of acid to be added based on the data after data fusion by using natural language processing technology, specifically as follows: Step 31: Align the data in the quantitative influence factor data set, the seasonal data, the operating conditions data of the thermal power plant, and the set alkalinity expected value data by using time series to obtain a multi-source data set, ensuring the synchronization of different data in time. Step 32: Use the long short-term memory network (LSTM) to build a model to obtain a circulating water alkalinity intelligent regulation model, specifically as follows: Input layer: Accept a continuous water quality data set sequence.

[0042] Normalization layer: Accelerate the training process and improve the generalization ability of the model through batch normalization.

[0043] The first layer of LSTM: Accept the normalized water quality data set and process the time series data to capture the dynamic changes in time, such as the increase, decrease, and stability of the alkalinity value.

[0044] The second - layer LSTM further processes the output of the first - layer LSTM, deepening the model's understanding of the time series and enabling it to capture more complex temporal dependencies, such as the co - directional or asynchronous changes in alkalinity values and pH values.

[0045] Fully - connected layer: A fully - connected layer is set after the second - layer LSTM to map the fused spatio - temporal features to specific alkalinity values.

[0046] Output layer: The output layer outputs the alkalinity values at the set future time steps ΔT and the amount of acid solution to be added required to control the value of the on - line alkalinity meter 18 of the circulating water within the range of the set alkalinity expected value ± 0.5 mmol / L.

[0047] Step 33: Model training. Using the training data set, which includes the water quality data that has been labeled, the labeled data is augmented through random time - domain transformation. Step 34: Using the intelligent control model for circulating water alkalinity obtained in Step 32 and natural language processing technology, based on the feature values of the multi - source data set in Step 31, give the predicted value of the circulating water alkalinity and the amount of acid solution to be added.

[0048] S4: Controlling acid addition. Control the operating current of the variable - frequency booster pump and the valve opening of the sulfuric acid regulating valve according to the amount of acid solution to be added to adjust the amount of acid added to the circulating water, including the following steps: Control the operating current of the variable - frequency booster pump 6 and the valve opening of the sulfuric acid regulating valve 9 according to the amount of acid solution to be added obtained in Step S3 to adjust the amount of acid added to the circulating water. When the measured value of the on - line alkalinity meter 18 of the circulating water is less than the set alkalinity expected value or the measured value of the sulfuric acid pipeline flowmeter 11 is greater than the amount of acid solution to be added, then control to close the variable - frequency booster pump 6 and the sulfuric acid regulating valve 9.

[0049] This solution can effectively improve the accuracy of circulating water alkalinity control by integrating various types of data such as water quality data, acid solution addition amount data, and set alkalinity expected value data. In specific implementation, through a small number of sensors and automated data acquisition and processing, the maintenance cost is reduced, and real - time monitoring and control of the circulating water alkalinity can be achieved, which helps to improve the safety and economy during the operation of the circulating water system.

[0050] Embodiment 3: As shown in the appendix Figure 1 to Figure 2 This embodiment discloses a method and system for controlling the alkalinity of circulating water based on alkalinity - pH measurement.

[0051] As shown in the appendix Figure 1As shown in the figure, the control system includes a control and data processing unit, a water quality data acquisition unit, and an acid addition amount data acquisition unit. The water quality data acquisition unit is used to collect the data of the on-line pH meter of the circulating water and the data of the on-line alkalinity meter of the circulating water. The acid addition amount data acquisition unit is used to collect the operating current data of the variable frequency booster pump, the opening data of the sulfuric acid regulating valve, the data of the sulfuric acid pipeline flowmeter, and the data of the cooling tower sump level meter.

[0052] As shown in the Figure 2 figure, the intelligent circulating water alkalinity regulation system of the present invention includes a cooling tower sump 1, a cooling tower sump level meter 2, a sulfuric acid storage tank 3, a sulfuric acid pipeline inlet valve 4, a variable frequency booster pump inlet valve 5, a variable frequency booster pump 6, a variable frequency booster pump outlet valve 7, a variable frequency booster pump bypass valve 8, a sulfuric acid regulating valve 9, a sulfuric acid regulating valve bypass valve 10, a sulfuric acid pipeline flowmeter 11, a circulating water return pipeline 12, a circulating water sampling pipeline 13, an intelligent circulating water alkalinity regulation device 14, a sewage pipeline 15, a circulating water sampling valve 16, an on-line pH meter 17 of the circulating water, an on-line alkalinity meter 18 of the circulating water, a circulating water sampling valve bypass valve 19, a controller and actuator 20, and a display and input interface 21.

[0053] A cooling tower sump level meter 2 is arranged in the cooling tower sump 1.

[0054] The outlet of the sulfuric acid storage tank 3 is connected to the cooling tower sump 1 through a sulfuric acid pipeline inlet valve 4, a variable frequency booster pump inlet valve 5, a variable frequency booster pump 6, a variable frequency booster pump outlet valve 7, a sulfuric acid regulating valve 9, and a sulfuric acid pipeline flowmeter 11 in sequence. A variable frequency booster pump bypass valve 8 is connected in parallel to the branch formed by the variable frequency booster pump inlet valve 5, the variable frequency booster pump 6, and the variable frequency booster pump outlet valve 7. A sulfuric acid regulating valve bypass valve 10 is connected in parallel to the sulfuric acid regulating valve 9.

[0055] The circulating water return pipeline 12 is connected to the circulating water sampling pipeline 13. The circulating water return is connected to the cooling tower sump 1 through a circulating water sampling valve 16. An on-line pH meter 17 of the circulating water and an on-line alkalinity meter 18 of the circulating water are arranged between the circulating water sampling valve 16 and the cooling tower sump 1. Among them, the on-line pH meter 17 of the circulating water and the on-line alkalinity meter 18 of the circulating water are connected in parallel. A circulating water sampling valve bypass valve 19 is connected in parallel to the branch formed by the circulating water sampling valve 16, the on-line pH meter 17 of the circulating water, and the on-line alkalinity meter 18 of the circulating water.

[0056] The controller and actuator 20 is electrically connected to the cooling tower sump level meter 2, the variable frequency booster pump 6, the sulfuric acid regulating valve 9, the sulfuric acid pipeline flowmeter 11, the on-line pH meter 17 of the circulating water, and the on-line alkalinity meter 18 of the circulating water. The controller and actuator 20 is communicatively connected to the display and input interface.

[0057] As shown in the Figure 3As shown in the figure, a method for precisely controlling the alkalinity of circulating water based on pH alkalinity measurement according to the present invention specifically includes the following steps: S1. Data collection: Collect water quality data and acid addition amount data through a water quality data collection unit and an acid addition amount data collection unit, and obtain seasonal data, thermal power plant operating condition data, and set alkalinity expected value data through a control and data processing unit.

[0058] Specifically, the water quality data includes the data of the on-line pH meter 17 of the circulating water, the data of the on-line alkalinity meter 18 of the circulating water, and the data of the cooling tower pool level meter 2; the acid addition amount data includes the operating current data of the variable frequency booster pump 6 and the data of the sulfuric acid pipeline flowmeter 11; the seasonal data includes, but is not limited to, month, season, and solar term; the thermal power plant operating condition data includes, but is not limited to, the main steam flow of the boiler, the power generation load, and the circulating water temperature rise.

[0059] S2. Data preprocessing: Normalize the water quality data and acid addition amount data obtained through step S1 into values with unified dimensions. The normalization calculation formula is as follows: X norm = ( X - μ ) / σ Wherein, X norm is the normalized value, X is the original data value, μ is the mean value of the original data group value, σ is the standard deviation of the original data value group.

[0060] For the specific values of the water quality data and acid addition amount data obtained in step S1, divide them into an appropriate alkalinity range, a high alkalinity range, and a low alkalinity range according to the seasonal data, thermal power plant operating condition data, and set alkalinity expected value data to obtain a quantitative influence factor data set.

[0061] Taking the data of the on-line alkalinity meter 18 of the circulating water as an example, the result of subtracting the set alkalinity expected value from the data value of the on-line alkalinity meter 18 of the circulating water is within the range of ±0.5 mmol / L for the appropriate alkalinity range, the result is greater than 0.5 mmol / L for the high alkalinity range, and the result is less than 0.5 mmol / L for the low alkalinity range.

[0062] Based on the seasonal data, thermal power plant operation condition data, and the set expected alkalinity data obtained in step S1, a circulating water alkalinity rule dataset that correlates water quality data, acid addition amount data, seasonal data, thermal power plant operation condition data, and the set expected alkalinity data is established. Taking seasonal data as an example, during summer, the temperature is relatively high, and the evaporation rate of circulating water in the cooling tower accelerates, resulting in continuous concentration of salts dissolved in the water, and the alkalinity of the circulating water is usually on the high side.

[0063] The preprocessing of the water quality data and acid addition amount data effectively reduces the complexity of the data, facilitating subsequent analysis.

[0064] S3. Data fusion: Perform data fusion on the data obtained in steps S1 and S2. Use a long short-term memory network (LSTM) to build a model to obtain a circulating water alkalinity intelligent control model. Select a general large model and use the data to fine-tune the model. Use the circulating water alkalinity intelligent control model to utilize natural language processing technology to give the predicted value of the circulating water alkalinity and the amount of acid solution to be added based on the data after data fusion.

[0065] Step 31: Align the data in the quantitative influence factor dataset, seasonal data, thermal power plant operation condition data, and the set expected alkalinity data using time series to obtain a multi-source dataset, ensuring the synchronization of different data in time. Among them, the data in the quantitative influence factor dataset and seasonal data are often relatively stable. For example, the quarter and month in the seasonal data can directly obtain the corresponding values of the time series.

[0066] Step 32: Build a model using a long short-term memory network (LSTM). The model input is the water quality dataset, including the circulating water online pH meter 17 numerical dataset, the circulating water online alkalinity meter 18 numerical dataset, and the cooling tower basin liquid level meter 2 data. The model makes a judgment based on the above input and outputs the alkalinity value at the set future time step ΔT and the amount of acid solution required to control the circulating water online alkalinity meter 18 numerical value within the range of the set expected alkalinity value ± 0.5 mmol / L.

[0067] The specific structure of the model is as follows: Input layer, accepting a continuous sequence of water quality datasets.

[0068] Normalization layer, accelerating the training process and improving the generalization ability of the model through batch normalization.

[0069] The first layer of LSTM, accepting the normalized water quality dataset and processing time series data to capture dynamic changes in time, such as the increase, decrease, and stability of alkalinity values.

[0070] The second - layer LSTM further processes the output of the first - layer LSTM, deepening the model's understanding of the time series and enabling it to capture more complex temporal dependencies, such as the co - variation or asynchronous variation of the alkalinity value and the pH value.

[0071] Fully - connected layer: A fully - connected layer is set after the second - layer LSTM to map the fused spatio - temporal features to specific alkalinity values.

[0072] Output layer: It outputs the alkalinity values for the set future time steps ΔT and the amount of acid solution to be added required to control the value of the on - line alkalinity meter 18 of the circulating water within the range of the set alkalinity expected value ±0.5 mmol / L.

[0073] Step 33: Model training. Through the training dataset, the training dataset includes the water quality data that has been labeled. To enhance the generalization ability of the model, the labeled data is augmented by random time - domain transformation, such as intercepting different time segments by sliding windows for data augmentation. In terms of the loss function, the root - mean - square error (RMSE) loss function is used, and the Adam optimizer is selected during the training process. This optimizer can automatically adjust the learning rate to ensure the efficiency and stability of parameter updates. During the training process, the validation set is used to evaluate the model, monitoring the model performance and overfitting situation. According to the performance of the validation set, the hyperparameters of the model are adjusted, including the learning rate, batch size, and the number of LSTM units. After training, a circulating water alkalinity intelligent regulation model is obtained.

[0074] Step 34: Select a general large - model and fine - tune the model using the data. Using the circulating water alkalinity intelligent regulation model obtained in Step 33 and natural language processing technology, based on the feature values of the multi - source dataset in Step 31, the predicted value of the circulating water alkalinity and the amount of acid solution to be added are given.

[0075] S4: Control acid addition. According to the amount of acid solution to be added obtained in Step S3, control the operating current of the variable - frequency booster pump 6 and the valve opening of the sulfuric acid regulating valve 9 to adjust the amount of acid added to the circulating water. When the measured value of the on - line alkalinity meter 18 of the circulating water is less than the set alkalinity expected value or the measured value of the sulfuric acid pipeline flowmeter 11 is greater than the amount of acid solution to be added, control the variable - frequency booster pump 6 and the sulfuric acid regulating valve 9 to close.

[0076] An electronic device includes: a processor; a memory for storing computer program instructions; and is used to implement the steps of the method for obtaining the circulating water alkalinity intelligent regulation model when executing the computer program.

[0077] A storage medium stores computer program instructions. When the computer program instructions are loaded and run by a processor, the processor executes the method for obtaining the circulating water alkalinity intelligent regulation model.

[0078] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0079] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0080] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realize the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0082] The above content is only to illustrate the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the present invention.

Claims

1. A method for obtaining an intelligent control model of circulating water alkalinity, characterized in that It includes the following steps: Obtain initial data, where the initial data includes water quality data of the circulating water return pipe (12), acid addition amount data of the cooling tower basin (1), seasonal data, power plant operating conditions data, and set alkalinity expectation value data; Establish a circulating water alkalinity law data set and a quantified influencing factor data set according to the initial data; Construct a model using a long short-term memory network based on the circulating water alkalinity law data set and the quantified influencing factor data set to obtain an initial intelligent control model for circulating water alkalinity; Label the water quality data, perform data augmentation on the labeled water quality data through random time-domain transformation, and use the augmented data to train the initial intelligent control model for circulating water alkalinity to obtain an optimal intelligent control model for circulating water alkalinity, and the intelligent control model for circulating water alkalinity is used for regulating the circulating water alkalinity.

2. The method for obtaining the intelligent control model of circulating water alkalinity according to claim 1, characterized in that The water quality data of the circulating water return pipe (12) includes: circulating water pH data, circulating water alkalinity data, and liquid level data of the cooling tower basin (1).

3. The method for obtaining the intelligent regulation model of circulating water alkalinity according to claim 1, characterized in that The acid addition amount data of the cooling tower basin (1) includes: operating current data of the variable frequency booster pump (6), opening data of the sulfuric acid regulating valve (9), and data of the sulfuric acid pipeline flowmeter (11).

4. The method for obtaining the intelligent control model of circulating water alkalinity according to claim 1, characterized in that The intelligent control model for circulating water alkalinity includes: An input layer that accepts a sequence of continuous water quality data sets; A normalization layer that accelerates the training process by batch-normalizing the water quality data; The first layer of LSTM that accepts the normalized water quality data set and processes time series data to capture dynamic changes over time; The second layer of LSTM that further processes the output of the first layer of LSTM to deepen the model's understanding of the time series and can capture more complex time-dependent relationships; A fully connected layer that maps the fused spatio-temporal features to specific alkalinity values; An output layer that outputs the alkalinity value at the set future time step and the required acid addition amount.

5. The method for obtaining the intelligent regulation model of circulating water alkalinity according to claim 1, wherein The establishment of the circulating water alkalinity law data set and the quantified influencing factor data set according to the initial data is specifically as follows: Use the normalization method to preprocess the water quality data and the acid addition amount data to obtain preprocessed data with the same dimension; Establish a circulating water alkalinity law data set that correlates water quality data, acid addition amount data, seasonal data, power plant operating conditions data, and set alkalinity expectation value data according to the seasonal data, power plant operating conditions data, set alkalinity expectation value data, and preprocessed data; Divide the preprocessed data into a suitable alkalinity range, a high alkalinity range, and a low alkalinity range according to the seasonal data, power plant operating conditions data, and set alkalinity expectation value data to obtain a quantified influencing factor data set.

6. The method for obtaining the intelligent control model of circulating water alkalinity according to claim 1, wherein, The specific calculation formula for the normalization is as follows: X norm = （ X - μ ） / σ Among them, X norm is the preprocessed data after normalization, X is the original data value, and the original data value includes water quality data and acid addition amount data, μ is the mean value of the corresponding original data group value, σ is the standard deviation of the original data group value.

7. A method for controlling the alkalinity of circulating water based on the measurement of alkalinity pH, characterized in that, It includes the following steps: Obtain initial data, where the initial data includes water quality data of the circulating water return pipe (12), acid addition amount data of the cooling tower basin (1), seasonal data, power plant operating conditions data, and set alkalinity expectation value data; Divide the preprocessed data into a suitable alkalinity range, a high alkalinity range, and a low alkalinity range according to the seasonal data, power plant operating conditions data, and set alkalinity expectation value data to obtain a quantified influencing factor data set; Align the data in the quantified influencing factor dataset, seasonal data, thermal power plant operating condition data, and the set expected alkalinity data using time series to obtain a multi-source dataset; According to the circulating water alkalinity intelligent regulation model described in any one of claims 1 to 6, use natural language processing technology to obtain the predicted value of the circulating water alkalinity and the amount of acid solution to be added based on the multi-source dataset; Adjust the amount of acid solution added to the cooling tower basin (1) according to the amount of acid solution to be added.

8. Acquisition system for intelligent regulation model of circulating water alkalinity, characterized in that, It includes: Data acquisition module: used to acquire initial data, and the initial data includes the water quality data of the circulating water return pipeline (12), the amount of acid solution added to the cooling tower basin (1), seasonal data, thermal power plant operating condition data, and the set expected alkalinity value data; Dataset construction module: used to establish a circulating water alkalinity rule dataset and a quantified influencing factor dataset according to the initial data; Model construction module: used to construct a model using a long short-term memory network based on the circulating water alkalinity rule dataset and the quantified influencing factor dataset to obtain an initial circulating water alkalinity intelligent regulation model; Model training module: used to label the water quality data, perform data augmentation on the labeled water quality data through random time-domain transformation, and use the augmented data to train the initial circulating water alkalinity intelligent regulation model to obtain an optimal circulating water alkalinity intelligent regulation model, and the circulating water alkalinity intelligent regulation model is used for circulating water alkalinity regulation.

9. An electronic device, comprising: A processor; a memory, and the electronic device is used to store computer program instructions; characterized in that when executing the computer program, it realizes the steps of the method for obtaining the circulating water alkalinity intelligent regulation model described in any one of claims 1-6.

10. A storage medium storing computer program instructions, characterized in that, When the computer program instructions are loaded and run by the processor, the processor executes the method for obtaining the circulating water alkalinity intelligent regulation model described in any one of claims 1-6.

Images