A flow control system and method for residual pressure power generation in water plants

Abstract

This invention belongs to the field of power generation control technology, and provides a flow control system and method for wastewater treatment plant residual pressure power generation. It acquires data sets of purified water flow rate and pressure from the original discharge pipeline of a wastewater treatment plant over a specified period, predicts the stability level of the purified water discharge, and derives a flow control model based on the stability level. It receives first real-time flow rate data from the main pipeline flow meter and real-time water pressure data from the pressure gauge, as well as second real-time flow rate data from the bypass pipeline flow meter. The flow control model outputs predicted first opening degree of the main pipeline electric valve and second opening degree of the bypass pipeline outlet electric valve, and executes the first and second opening degrees. This invention ensures the stability of the power output of the wastewater residual pressure power generation device.

Description

Technical Field

[0001] This invention relates to the field of power generation control technology, and more specifically, to a flow control system and method for power generation from residual pressure in water plants. Background Technology

[0002] In today's society, with the acceleration of industrialization and the continuous growth of the population, the demand for water resources and the scale of water treatment are becoming increasingly enormous. As a key facility for ensuring urban water safety, water purification plants involve a large amount of water pressure operation in their operation. Currently, in the water purification process, raw water undergoes a series of complex purification treatments to produce purified water with a certain pressure. However, the residual pressure carried by this purified water is often directly consumed or wasted during its delivery to the user end, and is not effectively utilized.

[0003] Currently, technologies for generating electricity using residual pressure from purified water have emerged, such as... Figure 1 As shown, the system comprises two pipelines: a main pipeline (the original discharge pipeline for purified water from the wastewater treatment plant) and a bypass pipeline (the installation pipeline for the wastewater treatment plant's residual pressure power generation device). The electricity generated by the residual pressure power generation device needs to meet grid connection requirements, which necessitates ensuring the stability of the head (the vertical height difference (potential energy) during purified water discharge, the core energy source driving the turbine). Stable head control is achieved by adjusting the opening of the electric valves on the main pipeline and the outlet electric valve of the bypass pipeline. Currently, existing technologies primarily rely on rule-based control strategies for adjusting the opening of these valves. This involves selecting appropriate valve control parameters based on the correspondence between the real-time flow detected by the flowmeter and the rules. However, since the established rules often lack sufficient refinement, this control method results in a coarse control effect and unstable power generation from the residual pressure power generation device.

[0004] Therefore, how to more accurately regulate the flow of the water purification residual pressure power generation system to ensure the stability of power generation is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] To address this issue, the present invention provides a flow control method, system, electronic device, computer storage medium, and computer program product for power generation from residual pressure in water plants, thereby solving the aforementioned technical problems.

[0006] This invention discloses a flow control method for residual pressure power generation in water plants, the method comprising the following steps:

[0007] Obtain the clean water flow rate and clean water pressure datasets from the original clean water discharge pipeline of the wastewater treatment plant within a specified time period, and predict the clean water discharge stability level based on the clean water flow rate dataset and the clean water pressure dataset.

[0008] A flow control model is derived based on the stability level of the purified water discharge. Specifically, if the stability level of the purified water discharge is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model.

[0009] The system receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It then inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

[0010] In some embodiments, the specified duration is determined in the following manner:

[0011] Obtain the inner wall roughness of the original pipeline and the layout information of the original pipeline, and evaluate the pipeline layout complexity based on the layout information.

[0012] The pipe inner wall roughness and the pipe layout complexity are simultaneously matched with a matching relationship to obtain the specified duration; wherein, in the matching relationship, both the pipe inner wall roughness and the pipe layout complexity are positively correlated with the specified duration.

[0013] In some embodiments, predicting the water discharge stability level based on the water flow rate dataset and the water pressure dataset includes:

[0014] Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge.

[0015] Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

[0016] In some embodiments, the second flow control model is trained in the following manner:

[0017] Collect and construct a training dataset. Each piece of training data in the training dataset includes: real-time flow rate and real-time water pressure of the original pipeline, real-time flow rate of the bypass pipeline, opening degree of the electric valve of the main pipeline and opening degree of the electric valve of the bypass pipeline outlet, and label data; wherein, the label data is the stability evaluation value of the preset head.

[0018] The training data in the training dataset is divided into a training set and a test set. The second traffic control model is trained using the training set. After the training reaches a preset level, the second traffic control model is tested using the test set. If the test is satisfactory, the training ends; if the test is unsatisfactory, the training continues using other training data in the training set until the test is satisfactory.

[0019] In some embodiments, the training data used for training the prediction model is derived from multiple sets of historical discharge data from the original water discharge pipeline of the wastewater treatment plant. The historical discharge data includes water flow rate data and water pressure data corresponding to the specified duration and a period of time after the specified duration.

[0020] The present invention also discloses a flow control system for residual pressure power generation in water plants, the system comprising a water discharge stability level prediction unit, a flow control model screening unit, and a flow control unit;

[0021] The water discharge stability level prediction unit acquires the water flow rate dataset and water pressure dataset of the original water discharge pipeline of the sewage treatment plant within a specified time period, and predicts the water discharge stability level based on the water flow rate dataset and water pressure dataset.

[0022] The flow control model screening unit matches the flow control model according to the water discharge stability level. Specifically, if the water discharge stability level is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model.

[0023] The flow control unit receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

[0024] In some embodiments, the water discharge stability level prediction unit is used for:

[0025] Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge.

[0026] Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

[0027] The present invention also discloses an electronic device comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program, when executed, implements the method as described in any of the preceding claims.

[0028] The present invention also discloses a computer storage medium storing a computer program that, when executed, implements the method as described in any of the preceding claims.

[0029] The present invention also discloses a computer program product having computer program code, which, when executed, implements the method described in the preceding claim.

[0030] The beneficial effects of this invention are as follows:

[0031] The flow control method of the present invention can significantly improve the targeting of flow control by flexibly selecting an appropriate flow regulation model through accurate assessment of the stability of purified water discharge. It can better adapt to the fluctuations in purified water discharge, stabilize power generation, improve power quality, facilitate grid connection, reduce equipment impact losses, and extend equipment service life. Attached Figure Description

[0032] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a schematic diagram illustrating the basic principle of a water purification residual pressure power generation system;

[0034] Figure 2 This is a schematic flowchart of a flow control method for residual pressure power generation in a water plant, as disclosed in an embodiment of the present invention.

[0035] Figure 3 This is a schematic diagram of the structure of a flow control system for residual pressure power generation in a water plant, as disclosed in an embodiment of the present invention. Detailed Implementation

[0036] The following specific embodiments illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0037] Furthermore, the technical features involved in the different embodiments of this application described below can be combined with each other as long as they do not conflict with each other.

[0038] The control method and control system of the present invention are based on Figure 1 The water purification residual pressure power generation system shown herein and the principles described in the background technology.

[0039] like Figure 2 As shown in the figure, an embodiment of the present invention discloses a flow control method for power generation from residual pressure in a water plant, the method comprising the following steps:

[0040] S1, obtain the clean water flow rate dataset and clean water pressure dataset of the original clean water discharge pipeline of the sewage treatment plant within a specified time period, and predict the clean water discharge stability level based on the clean water flow rate dataset and the clean water pressure dataset.

[0041] Pressure gauges and main pipeline flow meters are installed in the original pipeline of the wastewater treatment plant for the discharge of purified water. During the specified time period after the start of the discharge of purified water, multiple sets of data are collected by the pressure gauges and the main pipeline flow meters, which constitute the purified water flow data set and the purified water pressure data set, respectively. These time series data reflect the dynamic changes of the purified water discharge.

[0042] Then, a prediction model is pre-built, using the aforementioned purified water flow rate and purified water pressure datasets as input data. The prediction model then predicts the purified water discharge stability level for this operation. The purified water discharge stability level reflects the predicted fluctuation range of purified water flow in the original pipeline in terms of unit flow rate and water pressure during this operation, i.e., stability (the corresponding purified water discharge stability level for subsequent time periods can be obtained through multi-step prediction). A higher purified water discharge stability level indicates that the fluctuation range of purified water flow in terms of unit flow rate and water pressure will be smaller in the following time period, i.e., higher stability. Conversely, a lower stability level indicates that the fluctuation range of purified water flow in terms of unit flow rate and water pressure will be larger, i.e., lower stability.

[0043] It should be noted that the prediction model can be built based on algorithms such as LSTM and CNN, and this invention does not impose any specific limitations on it.

[0044] S2, a flow control model is obtained by matching the water discharge stability level. Specifically, if the water discharge stability level is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model.

[0045] Once the stable water discharge level is obtained, it is compared with a pre-set preset level. The preset level is a benchmark determined based on long-term practical experience and expectations for system performance. If the predicted stable water discharge level is higher than the preset level, it indicates that the current water discharge status is relatively stable, and fluctuations in water flow and pressure are within an acceptable range. In this case, the first flow control model is selected. This model has relatively low control precision, but its advantages lie in its relatively simple calculation process, rapid response, and low requirements for equipment computing power. It also reduces wear and tear on equipment caused by frequent fine-tuning and is sufficient to meet flow control requirements under stable operating conditions, thereby reducing system operating costs.

[0046] Conversely, if the water discharge stability level is lower than the preset level, it indicates that the discharge status is unstable and the changes in water flow and pressure are complex. In this case, a second flow control model with higher control precision is activated. This model can more sensitively capture subtle changes in flow and pressure and provide more precise control strategies to cope with unstable discharge conditions, ensuring that the system can operate stably under complex and ever-changing conditions and guaranteeing that the generated electricity can meet grid connection requirements.

[0047] It should be noted that the first flow control model can be constructed based on classification control algorithms such as SVM and random forest, while the second flow control model can be constructed based on deep Q-networks.

[0048] S3 receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

[0049] During the power generation process of the water purification residual pressure power generation unit, the main pipeline flow meter continuously measures and reports the first real-time flow, the pressure gauge collects and reports water pressure data in real time, and the bypass pipeline flow meter provides the second real-time flow. Simultaneously, a water head is pre-specified based on the power generation target, for example, 30 meters. These real-time data and the preset water head are input together into the flow control model selected in the preceding steps. The flow control model fully considers the interrelationship between flow, pressure, and water head, ultimately outputting the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve. These two opening values ​​precisely determine the flow distribution within the two pipelines. Through precise control of the water flow, stable control of the water head is achieved, ensuring the stability of the power generated by the water purification residual pressure power generation unit.

[0050] Taking the second flow control model as an example using a deep Q-network: a deep neural network is constructed as the Q-function estimator. The input is the current system state (including the first real-time flow rate, real-time water pressure data, the second real-time flow rate, and the preset head), and the output is the Q-value of each possible action (different valve opening combinations) in that state. Through continuous iterative training, the network learns the valve opening adjustment strategy that can obtain the maximum cumulative reward (such as making the power generation most stable) under different states.

[0051] The flow control method of the present invention can significantly improve the targeting of flow control by flexibly selecting an appropriate flow regulation model through accurate assessment of the stability of purified water discharge. It can better adapt to the fluctuations in purified water discharge, stabilize power generation, improve power quality, facilitate grid connection, reduce equipment impact losses, and extend equipment service life.

[0052] In some embodiments, the specified duration is determined in the following manner:

[0053] Obtain the inner wall roughness of the original pipeline and the layout information of the original pipeline, and evaluate the pipeline layout complexity based on the layout information.

[0054] The pipe inner wall roughness and the pipe layout complexity are simultaneously matched with a matching relationship to obtain the specified duration; wherein, in the matching relationship, both the pipe inner wall roughness and the pipe layout complexity are positively correlated with the specified duration.

[0055] In this embodiment of the invention, the roughness of the pipe inner wall is one of the key factors affecting the water flow state. The original pipe inner wall is periodically inspected beforehand using, for example, a roughness tester. The inner wall roughness reflects the unevenness of the pipe's inner surface, directly affecting the friction between the water flow and the pipe wall. The rougher the inner wall (the more severe the corrosion and scaling), the greater the energy loss of the water flow, the worse the water flow stability, and the longer it takes to reach a stable state.

[0056] Pipeline layout information includes the pipeline's direction, the presence of bends, tees, and other components, as well as their quantity and distribution. This information is obtained through on-site surveys and reviewing pipeline design drawings, and the complexity of the pipeline layout is then assessed. A complex pipeline layout means that the water flow needs to frequently change direction and velocity during transport, which can easily generate vortices and impacts, requiring more time for the water flow to stabilize.

[0057] A set of matching relationships is pre-established, which clarifies the connection between pipe inner wall roughness, pipe layout complexity, and a specified duration. In this set of matching relationships, both pipe inner wall roughness and pipe layout complexity are positively correlated with the specified duration. That is, the higher the pipe inner wall roughness and the greater the pipe layout complexity, the longer the corresponding specified duration.

[0058] This positive correlation is based on a large amount of experimental data, practical engineering experience, and relevant fluid mechanics theories. For example, in experiments simulating pipe flow with different roughness and layout complexity, the time required for the water flow to stabilize from the start of discharge was recorded. Through multiple experiments and data analysis, a quantitative relationship between the two and the time was obtained.

[0059] The acquired pipe inner wall roughness data and the evaluated pipe layout complexity data are simultaneously substituted into the established matching relationship for matching operations, ultimately yielding a specified duration corresponding to the pipe characteristics. The purified water flow rate and purified water pressure datasets obtained based on this specified duration can accurately reflect the true fluctuation characteristics of the discharged purified water under the given pipe conditions, overcoming external disturbance factors.

[0060] It should be noted that although the water flow rate and water pressure datasets within the specified time period also include front-end data affected by pipe inner wall roughness and pipe layout complexity, they already contain sufficient back-end data that can reflect the true fluctuation characteristics of purified water. Therefore, overall, they can meet the needs of subsequent analysis.

[0061] In some embodiments, predicting the water discharge stability level based on the water flow rate dataset and the water pressure dataset includes:

[0062] Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge.

[0063] Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

[0064] In this embodiment of the invention, the aforementioned embodiments directly input the purified water flow rate dataset and purified water pressure dataset into the flow control model. Although this includes sufficient back-end data that can reflect the true fluctuation characteristics of the purified water, the poor quality of the front-end data can still cause interference, hindering precise valve control. To address this issue, this embodiment sets a cutoff period to be determined from a specified time period. The cutoff period is a time period traced back from the start of purified water discharge, with the end of the specified time period as the starting point. This setting allows for the acquisition of recent water flow data that largely eliminates the influence of pipe wall roughness and pipe layout complexity. The fluctuation characteristics of this water flow data are more correlated with the fluctuation of the purified water discharge itself (i.e., changes in the amount and pressure of discharged purified water). The cutoff period can be a fixed percentage of the specified time period, such as 50% or 40%.

[0065] Convolutional neural networks (CNNs) were used to extract features from the target purified water flow rate dataset and the target purified water pressure dataset, respectively. For the target purified water flow rate dataset, the CNN extracted features such as the flow rate trend and the frequency of peak occurrences; for the target purified water pressure dataset, features such as the pressure fluctuation amplitude and the period of pressure change were extracted. Then, the two sets of extracted features were concatenated to form a comprehensive purified water fluctuation feature.

[0066] The spliced ​​purified water fluctuation features are input into a pre-trained prediction model, which is the aforementioned model built based on algorithms such as LSTM and CNN. This prediction model predicts and outputs the corresponding purified water discharge stability level. For example, the prediction model may divide the stability level into three levels: high, medium, and low (or number the levels with numbers, where a larger number corresponds to a higher stability level), and determine which stability level the current water flow is in based on the input purified water fluctuation features.

[0067] In some embodiments, the second flow control model is trained in the following manner:

[0068] Collect and construct a training dataset, wherein each piece of training data in the training dataset includes: real-time flow rate and real-time water pressure of the original pipeline, real-time flow rate of the bypass pipeline, opening degree of the electric valve of the main pipeline and opening degree of the electric valve of the bypass pipeline outlet, preset head, and label data; wherein the label data is the stability evaluation value of the preset head.

[0069] The training data in the training dataset is divided into a training set and a test set. The second traffic control model is trained using the training set. After the training reaches a preset level, the second traffic control model is tested using the test set. If the test is satisfactory, the training ends; if the test is unsatisfactory, the training continues using other training data in the training set until the test is satisfactory.

[0070] In this embodiment of the invention, the second flow control model needs to be pre-trained. When the preset head is in a stable state, the real-time flow rate and real-time water pressure of the original pipeline are collected, as well as the real-time flow rate of the bypass pipeline, the opening degree of the electric valve of the main pipeline and the opening degree of the electric valve at the outlet of the bypass pipeline, and the preset head at this time are also collected. The labeled data is the stability evaluation value of the preset head, such as the fluctuation range of the preset head. The lower the fluctuation range, the higher the stability evaluation value. This evaluation value can be manually evaluated or obtained by an automatic evaluator based on the actual head fluctuation.

[0071] The training objective of the second flow control model is to determine the appropriate valve opening under specific purified water flow conditions (i.e., the real-time flow and real-time water pressure of the original pipeline and the real-time flow of the bypass pipeline) and a preset head, so as to maintain the stability of the preset head (i.e., the fluctuation amplitude is less than the preset amplitude).

[0072] To ensure the effectiveness and generalization ability of the model training, the constructed training dataset needs to be reasonably divided. A portion of the data is used as the training set to train the second traffic control model. For example, the training set should comprise 70%-80% of the dataset. This allows the model to continuously learn the patterns and features in the data and adjust its parameters to achieve initial traffic control capabilities. The other portion of the data serves as the test set for evaluating the performance of the trained model.

[0073] The second flow control model is trained using a training set. Once the model reaches a predetermined level of training, such as a certain number of training rounds or a reduction in error to a certain threshold, it is tested using a test set. After the test set data is input into the model, the model outputs a control strategy, which is then compared with the corresponding preset head stability evaluation value in the test set. If the model's test results meet the standard—meaning the model's output control strategy keeps the error between the actual head and the preset head within an acceptable range—the model demonstrates good performance and generalization ability, and the training process ends. Conversely, if the test results fail, the model has shortcomings and needs further training using data not used in the training set to optimize model parameters and improve performance until the test results meet the standard. Through this iterative training and testing process, the second flow control model is ensured to accurately and stably control the flow based on various real-time water flow data, meeting the stable head requirements of the water purification residual pressure power generation system.

[0074] In some embodiments, the training data used for training the prediction model is derived from multiple sets of historical discharge data from the original water discharge pipeline of the wastewater treatment plant. The historical discharge data includes water flow rate data and water pressure data corresponding to the specified duration and a period of time after the specified duration.

[0075] In this embodiment of the invention, the prediction model is based on the historical data of the wastewater treatment plant’s purified water discharge. That is, a large number of sets of historical discharge data from the original purified water discharge pipelines of the wastewater treatment plant are collected. The historical discharge data includes water flow data from the start of purified water discharge to near the end of discharge, that is, multiple sets of purified water flow data and purified water pressure data obtained by sampling for a preset time period.

[0076] Based on these purified water flow and pressure data, training data is constructed to prepare the prediction model. The trained model can then predict the stability level of purified water discharge over a subsequent time period (e.g., 3 minutes, 5 minutes) based on the fluctuation characteristics of purified water discharge in the early stages. This information is used to select the appropriate first or second flow control model. The prediction of the purified water discharge stability level can be initiated gradually, allowing for the progressive determination of the appropriate flow control model for each subsequent time period.

[0077] like Figure 3 As shown in the figure, the present invention also discloses a flow control system for power generation from residual pressure in water plants. The system includes a water discharge stability level prediction unit, a flow control model screening unit, and a flow control unit.

[0078] The water discharge stability level prediction unit acquires the water flow rate dataset and water pressure dataset of the original water discharge pipeline of the sewage treatment plant within a specified time period, and predicts the water discharge stability level based on the water flow rate dataset and water pressure dataset.

[0079] The flow control model screening unit matches the flow control model according to the water discharge stability level. Specifically, if the water discharge stability level is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model.

[0080] The system receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It then inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

[0081] In some embodiments, the water discharge stability level prediction unit is used for:

[0082] Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge.

[0083] Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

[0084] This invention also discloses an electronic device comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program, when executed, implements the method as described in any of the preceding embodiments.

[0085] This invention also discloses a computer storage medium storing a computer program that, when executed, implements the method described in any of the preceding claims.

[0086] This invention also discloses a computer program product containing computer program code, which, when executed, implements the method described in any of the preceding embodiments.

[0087] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means two or more. Any process or method description in the flowcharts or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as should be understood by those skilled in the art to which the embodiments of the invention pertain.

[0088] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0089] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A flow control method for residual pressure power generation in a water plant, characterized in that, The method includes the following steps: Obtain the clean water flow rate and clean water pressure datasets from the original clean water discharge pipeline of the wastewater treatment plant within a specified time period, and predict the clean water discharge stability level based on the clean water flow rate dataset and the clean water pressure dataset. A flow control model is derived based on the stability level of the purified water discharge. Specifically, if the stability level of the purified water discharge is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model. The system receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It then inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

2. The flow control method for residual pressure power generation in a water plant according to claim 1, characterized in that: The specified duration is determined in the following manner: Obtain the inner wall roughness of the original pipeline and the layout information of the original pipeline, and evaluate the pipeline layout complexity based on the layout information. The pipe inner wall roughness and the pipe layout complexity are simultaneously matched with a matching relationship to obtain the specified duration; wherein, in the matching relationship, both the pipe inner wall roughness and the pipe layout complexity are positively correlated with the specified duration.

3. The flow control method for residual pressure power generation in a water plant according to claim 2, characterized in that: The prediction of the water discharge stability level based on the water flow rate dataset and the water pressure dataset includes: Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge. Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

4. The flow control method for residual pressure power generation in a water plant according to claim 1, characterized in that: The second flow control model is trained in the following manner: Collect and construct a training dataset. Each piece of training data in the training dataset includes: real-time flow rate and real-time water pressure of the original pipeline, real-time flow rate of the bypass pipeline, opening degree of the electric valve of the main pipeline and opening degree of the electric valve of the bypass pipeline outlet, and label data; wherein, the label data is the stability evaluation value of the preset head. The training data in the training dataset is divided into a training set and a test set. The second traffic control model is trained using the training set. After the training reaches a preset level, the second traffic control model is tested using the test set. If the test is satisfactory, the training ends; if the test is unsatisfactory, the training continues using other training data in the training set until the test is satisfactory.

5. The flow control method for residual pressure power generation in a water plant according to claim 1, characterized in that: The training data used for the prediction model training is derived from multiple sets of historical discharge data from the original water discharge pipeline of the wastewater treatment plant. The historical discharge data includes water flow rate data and water pressure data corresponding to the specified time period and a certain time period after the specified time period.

6. A flow control system for residual pressure power generation in a water plant, characterized in that, The system includes a water discharge stability level prediction unit, a flow control model screening unit, and a flow control unit. The water discharge stability level prediction unit acquires the water flow rate dataset and water pressure dataset of the original water discharge pipeline of the sewage treatment plant within a specified time period, and predicts the water discharge stability level based on the water flow rate dataset and water pressure dataset. The flow control model screening unit matches the flow control model according to the water discharge stability level. Specifically, if the water discharge stability level is higher than the preset level, the first flow control model is selected as the flow control model; otherwise, the second flow control model is selected as the flow control model. The control accuracy of the first flow control model is lower than that of the second flow control model. The flow control unit receives the first real-time flow rate from the main pipeline flow meter and the real-time water pressure data from the pressure gauge, as well as the second real-time flow rate from the bypass pipeline flow meter. It inputs the first real-time flow rate, the real-time water pressure data, the second real-time flow rate, and the preset head into the flow control model. The flow control model outputs the first opening degree of the main pipeline electric valve and the second opening degree of the bypass pipeline outlet electric valve, and executes the first opening degree and the second opening degree.

7. A flow control system for residual pressure power generation in a water plant according to claim 6, characterized in that: The water discharge stability level prediction unit is used for: Based on the specified duration, the interception duration is obtained, and the target clean water flow dataset and the target clean water pressure dataset are respectively intercepted from the clean water flow dataset and the clean water pressure dataset according to the interception duration; wherein, the interception duration is the duration from the end of the specified duration as the starting point towards the start of the clean water discharge. Convolutional networks are used to extract features from the target clean water flow dataset and the target clean water pressure dataset. The extracted features are then concatenated to obtain clean water fluctuation features. These clean water fluctuation features are then input into a prediction model to obtain the clean water discharge stability level.

8. An electronic device, characterized in that: The electronic device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the computer program, when executed, implements the method as described in any one of claims 1-5.

9. A computer storage medium, characterized in that: The computer-readable storage medium stores a computer program that, when executed, implements the method as described in any one of claims 1-5.

10. A computer program product, characterized in that: The computer program product contains computer program code, which, when executed, implements the method as described in any one of claims 1-5.

Images