Circulating water system energy-saving optimization control method based on pipeline characteristic identification

Abstract

The invention relates to the technical field of circulating water system optimization control, in particular to a circulating water system energy-saving optimization control method based on pipeline characteristic identification, and the method comprises the steps: building a lumped parameter model of a pipeline according to a momentum equation of fluid in the pipeline; constructing an agent model of the thermodynamic system mechanism model; identifying at least one pipeline characteristic of the circulating water system; predicting the seawater temperature and the unit load in the target period; establishing a pipeline characteristic model for energy-saving optimization control of the circulating water system; and on the basis of a closed-loop adaptive online correction and model iteration mechanism, optimizing the weights of the long and short-term memory neural network model and the regression model of the circulating water system by using the new operation data. Based on mechanism model and data model hybrid modeling and measuring point data verification, high-precision pipeline characteristic identification is realized, and by predicting future working conditions such as seawater temperature and the like and combining real-time pipeline characteristics, prospective optimization is carried out, and closed-loop frequency conversion optimization control is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of circulating water system optimization control, and in particular to a circulating water system energy-saving optimization control method based on pipeline characteristic identification. Background Art

[0002] During power plant operation, the circulating water system plays a critical role in cooling the plant. By flowing cooling water through the condenser, it removes heat released from the steam, ensuring stable operation of the units. However, the water pumps must overcome the ever-changing flow resistance in the pipeline to maintain the required flow rate and pressure. The circulating water system, consisting of the circulating water pumps and the circulating water loop, consumes 20%-30% of the plant's electricity, reducing the system's energy efficiency.

[0003] The circulating water system plays a role in the thermal cycle of a power plant, and the following energy consumption relationship exists: the circulating water system consumes electricity, but increasing the circulating water flow rate can reduce condenser backpressure, thereby increasing the power generation capacity of the power plant. In theory, the power plant can use variable frequency control to control the circulating water pump power to maximize the net power generation capacity of the power plant, at which the backpressure reaches the optimal condenser vacuum. However, in practice, the optimal condenser vacuum of most power plants is determined based on slight power increases, focusing only on changes on the thermal side and ignoring the impact of changes in cold-end piping characteristics. Variable frequency pump control requires precise calculation of these piping characteristics.

[0004] Among the related technologies, the existing technologies of most power plants have low accuracy in predicting pipeline characteristics, resulting in the technology currently used by most power plants often failing to achieve optimal vacuum, thereby increasing the energy consumption of the power plants and increasing costs, which urgently needs to be improved. Summary of the Invention

[0005] The present invention provides an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification, in order to solve the problem in the related art that the existing technology of most power stations has low prediction accuracy of pipeline characteristics, resulting in the technology currently used by most power stations often failing to achieve optimal vacuum, thereby increasing the energy consumption of the power stations and high costs.

[0006] In a first aspect, an embodiment of the present invention provides a method for energy-saving optimization control of a circulating water system based on pipeline characteristic identification, which is applied to a model building stage and includes the following steps: constructing a lumped parameter model of the pipeline based on the momentum equation of the fluid in the pipeline, and obtaining at least one operating data of the circulating water system based on the lumped parameter model; constructing a proxy model of the thermal system mechanism model based on the at least one operating data, generating a simulation data set covering multiple operating conditions based on the proxy model, and inputting at least one boundary condition among predicted water temperature, circulating water flow rate, and predicted load scheduling into the proxy model to output at least one performance indicator of the circulating water system; identifying at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator; based on the at least one pipeline characteristic, constructing a long-short-term memory neural network model of the circulating water system based on historical multi-dimensional meteorological data and a preset time series to predict seawater temperature and unit load in a target period and generate predicted seawater temperature and unit load; and establishing a pipeline characteristic model for energy-saving optimization control of the circulating water system based on at least one sensor data, at least one key parameter, a design benchmark point, and the predicted seawater temperature and unit load of the circulating water system.

[0007] Optionally, in one embodiment of the present invention, the identifying of at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator includes: identifying the at least one operating data to generate an objective function of the circulating water system; calculating the flow field details and local resistance at any position in the pipeline according to a preset equation; establishing a prediction function containing time series characteristics using a long short-term memory network; and identifying at least one pipeline characteristic of the circulating water system based on the objective function, the flow field details, the local resistance and the prediction function.

[0008] Optionally, in one embodiment of the present invention, the calculation formula of the objective function is:

[0009] in, is the measured pressure difference, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic; The calculation formula of the prediction function is:

[0010] in, is the nonlinear mapping function of the long short-term memory network, is the input feature, is the hidden state at the previous moment, The weights and biases for the trained parameters of the LSTM network.

[0011] The second aspect of the present invention provides an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification, which is applied to the model application stage and includes the following steps: obtaining real-time monitoring data, predicted seawater temperature and unit load of the circulating water system; inputting the real-time monitoring data, the predicted seawater temperature and unit load into a pre-constructed pipeline characteristic model, so as to use the pre-constructed pipeline characteristic model to identify the pipeline characteristic curve of mechanism-data fusion, so as to match at least one pipeline characteristic with the pipeline characteristic curve to generate matching data, and generate a water pump operation combination that meets the preset optimal conditions and the operating frequency corresponding to the water pump operation combination based on the matching data, wherein the pre-constructed pipeline characteristic model is constructed by a mechanism model and a data model; based on the water pump operation combination and the operating frequency, a closed-loop adaptive online correction and model iteration mechanism of the circulating water system is established, and based on the closed-loop adaptive online correction and model iteration mechanism, the weights of the long and short-term memory neural network model and the regression model of the circulating water system are optimized using new operating data.

[0012] Optionally, in one embodiment of the present invention, generating a water pump operation combination that meets preset optimal conditions and an operating frequency corresponding to the water pump operation combination based on the matching data includes: based on the matching data, traversing all potential operating conditions under different water pump combinations and different variable frequency frequencies, so as to generate the water pump operation combination that meets the preset optimal conditions and an operating frequency corresponding to the water pump operation combination under the condition that the circulating water system meets the preset cooling requirements.

[0013] Optionally, in one embodiment of the present invention, the closed-loop adaptive online correction and model iteration mechanism is based on which new operating data is used to optimize the weights of the long short-term memory neural network model and the regression model of the circulating water system, including: based on the closed-loop adaptive online correction and model iteration mechanism, comparing the prediction result of the proxy model with the at least one pipeline characteristic to generate a prediction error; when the cumulative value or sliding average value of the prediction error exceeds a preset dynamic threshold, controlling the iterative training of the circulating water system to generate new operating data, and using the new operating data to optimize the weights of the long short-term memory neural network and the regression model of the circulating water system.

[0014] The third aspect of the present invention provides an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification, which is applied to the model construction stage and includes: an acquisition module for constructing a lumped parameter model of the pipeline according to the momentum equation of the fluid in the pipeline, and obtaining at least one operating data of the circulating water system based on the lumped parameter model; a construction module for constructing a proxy model of the thermal system mechanism model based on the at least one operating data, generating a simulation data set covering multiple working conditions according to the proxy model, and inputting at least one boundary condition of the predicted water temperature, circulating water flow and predicted load scheduling into the proxy model to output at least one performance of the circulating water system An indicator; an identification module for identifying at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator; a prediction module for constructing a long-short-term memory neural network model of the circulating water system based on the at least one pipeline characteristic, according to historical multi-dimensional meteorological data and a preset time series, so as to predict the seawater temperature and unit load in the target period and generate predicted seawater temperature and unit load; an establishment module for establishing a pipeline characteristic model for energy-saving optimization control of the circulating water system based on at least one sensor data, at least one key parameter, a design benchmark point and the predicted seawater temperature and unit load of the circulating water system.

[0015] Optionally, in one embodiment of the present invention, the identification module includes: a generation unit for identifying the at least one item of operating data to generate the objective function of the circulating water system; a calculation unit for calculating the flow field details and local resistance at any position in the pipeline according to a preset equation; an establishment unit for establishing a prediction function containing time series characteristics using a long short-term memory network; and an identification unit for identifying at least one pipeline characteristic of the circulating water system based on the objective function, the flow field details, the local resistance and the prediction function.

[0016] Optionally, in one embodiment of the present invention, the calculation formula of the objective function is:

[0017] in, is the measured pressure difference, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic; The calculation formula of the prediction function is:

[0018] in, is the nonlinear mapping function of the long short-term memory network, is the input feature, is the hidden state at the previous moment, The weights and biases for the trained parameters of the LSTM network.

[0019] The fourth aspect of the present invention provides an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification, which is applied to the model application stage and includes: a data acquisition module for acquiring real-time monitoring data, predicted seawater temperature and unit load of the circulating water system; a generation module for inputting the real-time monitoring data, the predicted seawater temperature and unit load into a pre-constructed pipeline characteristic model, so as to use the pre-constructed pipeline characteristic model to identify the pipeline characteristic curve of the mechanism-data fusion, so as to match at least one pipeline characteristic with the pipeline characteristic curve, generate matching data, and generate a water pump operation combination that meets the preset optimal conditions and an operating frequency corresponding to the water pump operation combination based on the matching data, wherein the pre-constructed pipeline characteristic model is constructed by a mechanism model and a data model; an optimization module for establishing a closed-loop adaptive online correction and model iteration mechanism of the circulating water system based on the water pump operation combination and the operating frequency, and based on the closed-loop adaptive online correction and model iteration mechanism, using new operating data to optimize the weights of the long and short-term memory neural network model and the regression model of the circulating water system.

[0020] Optionally, in one embodiment of the present invention, the generation module includes: a traversal unit, which is used to traverse all potential operating conditions under different water pump combinations and different variable frequency frequencies based on the matching data, so as to generate the water pump operation combination that meets the preset optimal conditions and the operating frequency corresponding to the water pump operation combination under the condition that the circulating water system meets the preset cooling requirements.

[0021] Optionally, in one embodiment of the present invention, the optimization module includes: a comparison unit, which is used to compare the prediction result of the agent model with the at least one pipeline characteristic based on the closed-loop adaptive online correction and model iteration mechanism to generate a prediction error; an optimization unit, which is used to control the iterative training of the circulating water system to generate new operating data when the cumulative value or sliding average value of the prediction error exceeds a preset dynamic threshold, and use the new operating data to optimize the weights of the long short-term memory neural network and regression model of the circulating water system.

[0022] A fifth aspect of the present invention provides an electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for energy-saving optimization control of a circulating water system based on pipeline characteristic identification as described in the above embodiment.

[0023] A sixth aspect of the present invention provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above-mentioned energy-saving optimization control method for a circulating water system based on pipeline characteristic identification.

[0024] The embodiments of the present invention can be based on hybrid modeling of mechanism models and data models, verified by measurement point data, taking into account the multiple influences of thermal cycle data and environmental data, and fully considering factors such as pipeline resistance and equipment resistance, thereby achieving high-precision pipeline characteristic identification. By predicting future operating conditions such as seawater temperature and combining them with real-time pipeline characteristics, forward-looking optimization is carried out to maximize the net power output of the power plant and achieve closed-loop variable frequency optimization control. This solves the problem in the related art that the existing technology of most power plants has low accuracy in predicting pipeline characteristics, resulting in the technology currently used by most power plants often failing to achieve optimal vacuum, thereby increasing the energy consumption of the power plants and increasing costs.

[0025] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A flow chart of a circulating water system energy-saving optimization control method based on pipeline characteristic identification provided by an embodiment of the present invention applied to model construction; Figure 2 This is an overall technical framework diagram of a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to an embodiment of the present invention; Figure 3 A schematic diagram of derivation of a momentum equation for unsteady flow in a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to an embodiment of the present invention; Figure 4 Schematic diagram of multi-dimensional boundary parameter prediction of a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to one embodiment of the present invention; Figure 5 Schematic diagram of pipeline characteristic feedforward identification of a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to one embodiment of the present invention; Figure 6 A flow chart of a circulating water system energy-saving optimization control method based on pipeline characteristic identification provided by an embodiment of the present invention applied to a model application stage; Figure 7Schematic diagram of an adaptive online correction and model iteration mechanism of a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to an embodiment of the present invention; Figure 8 A schematic diagram of a circulating water pump flow-efficiency curve of a circulating water system energy-saving optimization control method based on pipeline characteristic identification according to an embodiment of the present invention; Figure 9 A schematic structural diagram of a circulating water system energy-saving optimization control device based on pipeline characteristic identification provided by an embodiment of the present invention applied to a model building stage; Figure 10 A schematic structural diagram of a circulating water system energy-saving optimization control device based on pipeline characteristic identification according to an embodiment of the present invention applied to a model application stage; Figure 11 A schematic structural diagram of an electronic device provided according to an embodiment of the present invention.

[0027] Among them, 10-an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification; 100-acquisition module, 200-construction module, 300-identification module, 400-prediction module, 500-establishment module; 20-an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification, 600-data acquisition module, 700-generation module, 800-optimization module; 1101-memory, 1102-processor, 1103-communication interface. DETAILED DESCRIPTION

[0028] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0029] The following describes, with reference to the accompanying drawings, an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification according to an embodiment of the present invention. In response to the problem mentioned in the background art above, that the existing technologies used in most power plants have low accuracy in predicting pipeline characteristics, resulting in the technologies currently used in most power plants often failing to achieve optimal vacuum, thereby increasing energy consumption and high costs for the power plants, the present invention provides an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification. In this method, based on hybrid modeling of mechanism models and data models, and verification of measurement point data, the method takes into account the multiple influences of thermal cycle data and environmental data, and fully considers factors such as pipeline resistance and equipment resistance, thereby achieving high-precision pipeline characteristic identification. By predicting future operating conditions such as seawater temperature and combining them with real-time pipeline characteristics, a forward-looking optimization search is performed to achieve closed-loop variable frequency optimization control with the goal of maximizing the net power of the power plant. This solves the problem in the related art that the existing technologies used in most power plants have low accuracy in predicting pipeline characteristics, resulting in the technologies currently used in most power plants often failing to achieve optimal vacuum, thereby increasing energy consumption and high costs for the power plants.

[0030] Specifically, Figure 1 A schematic flow chart of a circulating water system energy-saving optimization control method based on pipeline characteristic identification provided by an embodiment of the present invention applied to the model building stage.

[0031] like Figure 1 As shown, the energy-saving optimization control method for the circulating water system based on pipeline characteristic identification includes the following steps: In step S101 , a lumped parameter model of the pipeline is constructed according to the momentum equation of the fluid in the pipeline, and at least one item of operating data of the circulating water system is obtained based on the lumped parameter model.

[0032] It can be understood that at least one operating data in the embodiment of the present invention includes exhaust steam volume, exhaust steam temperature, condensate temperature, circulating water inlet and outlet water temperature, back pressure, vacuum, circulating pump current and liquid level before and after the rotary filter.

[0033] In the actual implementation process, the overall technical framework of the embodiment of the present invention is as follows Figure 2 As shown. The embodiment of the present invention can perform mechanism modeling: Based on the momentum equation of the fluid in the pipeline, since the pressure wave velocity is much greater than the flow velocity of the fluid in the pipeline, the propagation process of the pressure wave can be ignored. At the same time, the compressibility of the fluid in the pipe and the elastic effect of the pipe wall are not considered. Therefore, the flow velocity (or flow rate) of the fluid in the pipe does not change with the spatial position, and a lumped parameter model describing the dynamic hydraulic process of the pipeline can be obtained. Figure 3The figure below is a schematic diagram of the derivation of the momentum equation for unsteady flow. The cross-sectional area of ​​the pipe is A, the thickness of the control body is δx, the positive direction of the coordinate axis x is along the centerline of the pipe, and the elevation of the control body is z. According to Newton's second law, omitting the higher-order terms yields the following equation:

[0034] in, is the pressure distribution of the fluid along the coordinate axis x, is the cross-sectional area of ​​the pipe, is the shear stress of the wall on the fluid, is the pipe diameter, is the sine of the angle between the center of the pipe and the horizontal plane, is the fluid momentum density, γ is the specific gravity, that is, .

[0035] Shear stress can be calculated using Darcy's formula under steady-state conditions. Under steady-state conditions, for a horizontal pipe with a length of L and a diameter of D, the resistance caused by the wall shear stress should be equal to the resistance calculated using Darcy's formula and satisfy the following formula:

[0036] in, is the pressure difference, is the shear stress of the wall on the fluid, is the pipe diameter, is the length.

[0037] Without considering the propagation and attenuation of pressure waves in the pipeline, and given that the pressure propagation speed in the pipeline is much faster than the dynamic change of flow rate, the hot water in the pipeline is approximately an incompressible fluid, and the pressure wave propagates very quickly along the pipeline fluid, so it can be approximated to 0. For a horizontal pipeline, the result is integrated along the pipe length from x = 0 to x = L, and the flow rate is introduced instead of the flow velocity, resulting in the following form of the pipeline unsteady hydraulic model:

[0038] in, is the pipe diameter, is the square of multiple groups of flow, is the length, is the density, is the pressure difference, is the Darcy friction factor.

[0039] In order to get closer to the dynamic calculation of the water pump head, after further arrangement, a formula was obtained, which reflects the dynamic relationship between the pressure change and flow in the pipeline, and the following formula was obtained:

[0040] in, is the pipe resistance coefficient, is the length, is the pipe diameter, is the acceleration due to gravity, is the volume flow rate, is the instantaneous change in flow rate, is the time element.

[0041] For water pumps, the pump head curve formula is:

[0042] in, is the secondary resistance coefficient, is the linear resistance coefficient, is the volume flow rate, is the static pressure head.

[0043] According to the first law of thermodynamics, the resistance loss along the pipeline should be equal to the head of the pump, so the following equation can be established to characterize the dynamic relationship between pressure change and flow rate:

[0044] in, is the secondary resistance coefficient, is the linear resistance coefficient, is the volume flow rate, is the static pressure head, is the Darcy friction factor, is the length, is the pipe diameter, is the acceleration due to gravity, is the instantaneous change in flow rate, is the time element.

[0045] To simplify the problem, the pump head curve is replaced by the pipeline characteristic curve:

[0046] in, is the total head of the pump, For static lift, is the dynamic resistance coefficient of the water pump, is the volume flow rate.

[0047] Sorting out the above, the dynamic relationship between pressure change and flow rate is:

[0048] in, is the dynamic resistance coefficient of the water pump, is the volume flow rate, For static lift, is the Darcy friction factor, is the length, is the acceleration due to gravity is the instantaneous change in flow rate, is the time element.

[0049] The embodiment of the present invention can collect data, including exhaust steam volume, exhaust steam temperature, condensate temperature, circulating water inlet and outlet water temperatures, back pressure, vacuum, circulating pump current, and liquid levels before and after the rotary filter, to obtain at least one item of operating data of the circulating water system, and preprocess the data by removing noise through the moving average method and filling missing values ​​through a time series prediction model.

[0050] In step S102, based on at least one item of operating data, a proxy model of the thermal system mechanism model is constructed, a simulation data set covering multiple working conditions is generated according to the proxy model, and at least one boundary condition of the predicted water temperature, circulating water flow and predicted load scheduling is input into the proxy model to output at least one performance indicator of the circulating water system.

[0051] It is understandable that system optimization relies on an accurate model of the thermal system mechanism. Commonly used methods include the NTU-ε heat exchange model and the iterative heat balance model. Although these two methods can ensure physical accuracy through energy conservation calculations, their iterative process is complex and it is difficult to meet the speed requirements of online real-time optimization.

[0052] The embodiment of the present invention can perform thermal system modeling: To meet the needs of real-time computing, a neural network algorithm is innovatively used to establish a surrogate model (Surrogate Model) of the complex mechanism model, that is, a surrogate model of the thermal system mechanism model. The surrogate model first generates a large number of simulation data sets covering various operating conditions by running a high-precision thermal system mechanism model offline, and then uses this as a training sample to learn and reproduce the complex nonlinear relationships inherent in the mechanism model. When applied online, the neural network surrogate model can bypass time-consuming iterative calculations, using boundary conditions such as predicted water temperature, circulating water flow, and predicted load scheduling as input values, and instantaneously output key performance indicators such as back pressure, unit output, and other parameters. This approach not only retains the physical completeness of the mechanism model, but also obtains real-time computing capabilities through the surrogate model, thereby providing a fast and accurate thermal side state assessment for subsequent dynamic optimization.

[0053] In step S103 , at least one pipeline characteristic of the circulating water system is identified based on the simulation data set and at least one performance indicator.

[0054] It is understandable that pipeline characteristics identification is a key step in the modeling and optimization of circulating water systems.

[0055] During the actual implementation process, the embodiment of the present invention can identify at least one pipeline characteristic of the circulating water system based on a simulation data set and at least one performance indicator, thereby comprehensively considering multi-dimensional disturbance data and providing support for subsequent energy-saving optimization control of the circulating water system based on pipeline characteristic identification.

[0056] The present invention is based on hybrid modeling of mechanism model and data model, verification of measurement point data, taking into account the multiple influences of thermal cycle side data and environmental data, and fully considering factors such as pipeline resistance and equipment resistance, thereby achieving high-precision identification of pipeline characteristics.

[0057] Optionally, in one embodiment of the present invention, at least one pipeline characteristic of the circulating water system is identified based on a simulation data set and at least one performance indicator, including: identifying at least one operating data to generate an objective function of the circulating water system; calculating the flow field details and local resistance at any position in the pipeline according to a preset equation; establishing a prediction function containing time series characteristics using a long short-term memory network; and identifying at least one pipeline characteristic of the circulating water system based on the objective function, flow field details, local resistance and prediction function.

[0058] It can be understood that the hybrid modeling of the present invention, based on a mechanistic and data model, ensures that pipeline characteristics conform to physical processes while also refining the model's practicality. The resistance characteristics of a pipeline are determined by a variety of factors, which vary with operating conditions and can be calculated using the Darcy-Weisbach equation.

[0059] Specifically, the embodiment of the present invention can perform pipeline characteristic identification: accurately identify the pipeline characteristics of the circulating water system to accurately determine the comprehensive resistance coefficient in the relationship between the total head and flow rate of the system S , the relationship between head and flow is:

[0060] in, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic.

[0061] (1) The embodiment of the present invention can identify the parameters of the operating data, treat the system as a "gray box", and use the multiple flow groups collected by the DCS (Distributed Control System) to Q and pressure drop Data, solve the objective function through optimization algorithms such as least squares method.

[0062] In one embodiment of the present invention, the calculation formula of the objective function is:

[0063] in, is the measured pressure difference, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic; (2) The embodiment of the present invention can be based on the "white box" simulation method of computational CFD (Computational Fluid Dynamics). By establishing an accurate three-dimensional model, the Navier-Stokes equations can be solved in software such as ANSYS Fluent, thereby forward calculating the flow field details at any position in the pipeline, and the friction coefficient can be solved according to the Darcy-Weisbach formula. and the local resistance of each component, the Darcy-Weisbach formula is:

[0064] in, is the Darcy friction factor, is the length, is the pipe diameter, is the density, is the flow rate.

[0065] (3) The embodiment of the present invention is based on a data-driven machine learning method and utilizes models such as LSTM (Long Short-Term Memory) to establish a prediction function that includes time series features. The calculation formula of the prediction function is:

[0066] in, is the nonlinear mapping function of the long short-term memory network, is the input feature, is the hidden state at the previous moment, Weights and biases for the trained parameters of the LSTM network.

[0067] Input features X t It includes not only operating parameters, but also time-varying factors such as "last cleaning days", so as to achieve the comprehensive resistance coefficient. S Dynamic prediction of drift due to scaling, aging, etc. These three methods provide a complete technical tool chain for pipeline characteristic analysis from different dimensions, from real-time evaluation to physical diagnosis to trend prediction.

[0068] The fusion process of the above three methods uses the "grey box" model as the core part of the system. It continuously uses the flow and pressure drop data collected in real time by DCS to calculate the comprehensive resistance coefficient through algorithms such as the least squares method. SThe current value of the resistance can be monitored in real time. S When there is a significant deviation from the long-term trend predicted by the "black box" model or the dynamically set warning threshold is exceeded, the system will automatically trigger the "white box" model with higher computational costs for in-depth diagnosis.

[0069] This fusion can form a complete closed loop of "monitoring-diagnosis-prediction-recalibration": real-time monitoring identifies problems, in-depth diagnosis identifies the root cause, and predictive models predict the future. Diagnostic results and historical data continuously optimize the accuracy of monitoring and prediction. This mechanism ensures the long-term and efficient operation of the entire circulating water system control strategy. The present invention combines the prediction results with high-precision pipeline characteristics, and optimizes with the goal of maximizing the net power of the power station. It realizes the transformation of the control mode from "passive response" to "active prediction", and finally outputs precise variable frequency control instructions, forming a complete technical route of "data verification-characteristic identification-prediction optimization-closed-loop control", ensuring that the unit always operates in the optimal economic range.

[0070] In step S104, based on at least one pipeline characteristic, a long-short-term memory neural network model of the circulating water system is constructed according to historical multi-dimensional meteorological data and a preset time series to predict the seawater temperature and unit load in the target period and generate predicted seawater temperature and unit load.

[0071] It can be understood that the seawater temperature and unit load in the target period in the embodiment of the present invention may be the future seawater temperature and the future unit load.

[0072] In the actual implementation process, Figure 4 As shown, embodiments of the present invention enable multi-dimensional boundary parameter prediction based on deep learning. To accurately predict the future state of complex systems, the present invention first constructs a deep time series prediction model for key boundary parameters. Considering the strong temporal nature and high inertia of seawater temperature, coupled with multiple physical factors such as solar radiation, atmospheric conditions, and ocean currents, the present invention abandons traditional linear models and instead employs a long-short-term memory neural network that captures long-term dependencies. This network integrates historical multi-dimensional meteorological data (such as air temperature, wind speed, wind direction, and visibility) with the seawater temperature time series as input. By constructing a complex nonlinear mapping, it achieves rolling, high-precision predictions of future seawater temperature, generating predicted seawater temperature and unit loads. Furthermore, by deeply analyzing historical unit heat balance data under different operating conditions, a unit load prediction model capable of accurately responding to grid dispatch commands is established. These two highly accurate prediction results together constitute the core thermal and hydraulic boundary inputs required for subsequent pipeline characteristic identification.

[0073] The embodiments of this invention are not only based on historical and current conditions but also proactive. By integrating long-term time series prediction models based on algorithms such as LSTM, they can accurately predict changes in key environmental boundary conditions, such as seawater temperature. By predicting future operating conditions such as seawater temperature and combining them with real-time pipeline characteristics, they conduct a forward-looking optimization search, achieving closed-loop variable frequency optimization control with the goal of maximizing the power plant's net power.

[0074] In step S105 , a pipeline characteristic model for energy-saving optimization control of the circulating water system is established based on at least one sensor data, at least one key parameter, a design reference point, and predicted seawater temperature and unit load of the circulating water system.

[0075] Specifically, if Figure 5 As shown, the embodiment of the present invention can perform feedforward identification of pipeline characteristics with mechanism-data fusion: after obtaining accurate boundary condition predictions, the present invention innovatively combines future seawater temperature and future unit load with real-time monitored tide data (tide level directly affects the suction pressure of the circulating water pump and is a key variable in the hydraulic dynamic process), and feeds them forward to a pipeline characteristic model with deep fusion of mechanism and data. The pipeline characteristic model uses the lumped parameter model derived from the fluid momentum equation as its core mechanism, which ensures the correctness and interpretability of the model deduction from a physical level. At the same time, in order to overcome the deviation between the theoretical model and the actual working conditions, data-driven methods such as polynomial linear regression are used to dynamically fit and identify complex terms such as the comprehensive resistance coefficient in the mechanism equation, so that it can accurately reflect the actual operating characteristics of the actual pipeline.

[0076] The embodiment of the present invention can use the predicted boundary conditions to directly solve the pipeline characteristic curve of the next time period, achieving a transition from post-diagnosis to pre-diagnosis.

[0077] Figure 6 The present invention is a flowchart of a circulating water system energy-saving optimization control method based on pipeline characteristic identification provided in an embodiment of the present invention applied to the model application stage.

[0078] In step S601, real-time monitoring data of the circulating water system, predicted seawater temperature and unit load are obtained.

[0079] Among them, the embodiment of the present invention can obtain real-time monitoring data of the circulating water system, predicted seawater temperature and unit load, thereby providing support for the subsequent construction of a pipeline characteristic model.

[0080] In step S602, real-time monitoring data, predicted seawater temperature and unit load are input into a pre-built pipeline characteristic model, so as to use the pre-built pipeline characteristic model to identify the pipeline characteristic curve of mechanism-data fusion, so as to match at least one pipeline characteristic with the pipeline characteristic curve, generate matching data, and generate a water pump operation combination that meets preset optimal conditions and an operating frequency corresponding to the water pump operation combination based on the matching data, wherein the pre-built pipeline characteristic model is constructed by the mechanism model and the data model.

[0081] It is understandable that in the embodiment of the present invention, the pump operation combination that meets the preset optimal conditions and the operating frequency corresponding to the pump operation combination can be the optimal pump operation combination and the corresponding optimal operating frequency. The resistance of the equipment, such as the valve opening, will significantly change the local resistance, thereby affecting the pipeline characteristic curve and affecting the optimal working point. Therefore, accurate monitoring is required to accurately solve the pipeline characteristics. The influence of the thermal cycle side data and environmental data on the cold end is taken into account, and the measurement point data is verified to ensure that the measured value is close to the true value to the greatest extent. In addition, the pipeline characteristics realize the adaptive update of the model and set safety constraints for the circulating water pump. The high-precision pipeline characteristics significantly improve the frequency conversion adjustment accuracy of the circulating water pump, so that the power station operation is significantly close to the optimal vacuum, saving the power plant's electricity consumption.

[0082] In this embodiment, real-time monitoring data, predicted seawater temperature, and unit load can be input into a pre-built pipeline characteristic model. After accurately feedforward identifying the pipeline characteristic curve and thermal load demand for the next period, the system enters the core energy-saving optimization decision-making stage. The core of this stage is an optimization solver aimed at minimizing system energy consumption. It intelligently matches the predicted pipeline characteristics with the performance characteristic curve of the circulating water pump itself, generates matching data, and calculates the unique solution that minimizes the total system power consumption, namely the optimal pump operation combination and the corresponding optimal operating frequency. The pipeline characteristic model is constructed from the mechanism model and the data model.

[0083] This embodiment of the present invention builds a comprehensive framework encompassing multi-level data validation and adaptive model updating. By integrating a fluid dynamics model with a data-driven regression model and establishing an error-driven adaptive update mechanism, it enables real-time and accurate identification of dynamic pipeline characteristics affected by scaling, wear, and environmental changes.

[0084] It should be noted that the preset optimal conditions can be set by those skilled in the art according to actual conditions and are not specifically limited here.

[0085] Optionally, in one embodiment of the present invention, a water pump operation combination that meets preset optimal conditions and an operating frequency corresponding to the water pump operation combination are generated based on matching data, including: based on the matching data, traversing all potential operating conditions under different water pump combinations and different variable frequency frequencies, so as to generate a water pump operation combination that meets preset optimal conditions and an operating frequency corresponding to the water pump operation combination under the condition that the circulating water system meets the preset cooling requirements.

[0086] Based on a genetic algorithm (GA), the solver in this embodiment of the present invention traverses all potential operating conditions under different pump combinations and variable frequency drives, calculating a unique solution that minimizes total system power consumption while meeting the predicted cooling demand. This solution, in other words, the optimal pump operating combination and its corresponding optimal operating frequency, is used to determine the optimal operating frequency. The output of this optimization decision is no longer a rough gear position or fixed speed, but rather a set of variable frequency instructions with decimal precision. This can be directly executed by the pump control system, effectively transforming "pre-emptive predictions" into "real-time energy-saving" control actions.

[0087] In step S603, based on the water pump operation combination and operation frequency, a closed-loop adaptive online correction and model iteration mechanism of the circulating water system is established, and based on the closed-loop adaptive online correction and model iteration mechanism, the weights of the long-short-term memory neural network model and the regression model of the circulating water system are optimized using new operation data.

[0088] It is understandable that the closed-loop adaptive online correction and model iteration mechanism in the embodiment of the present invention can use the pump operation combination and operation frequency as input, combined with real-time sensor data, to continuously update the system model.

[0089] Among them, such as Figure 7 As shown, the embodiment of the present invention is based on the water pump operation combination and operating frequency. In order to address the problem of model performance drift caused by factors such as equipment aging, pipeline scaling or marine organism adhesion, a closed-loop adaptive online correction and model iteration mechanism is established to utilize new operating data to optimize the weights of the long-short-term memory neural network model and regression model of the circulating water system.

[0090] The embodiments of the present invention can establish an error-driven adaptive update mechanism, thereby being able to accurately identify dynamic pipeline characteristics affected by scaling, wear, and environmental changes in real time.

[0091] Optionally, in one embodiment of the present invention, based on a closed-loop adaptive online correction and model iteration mechanism, new operating data is used to optimize the weights of the long-short-term memory neural network model and the regression model of the circulating water system, including: based on a closed-loop adaptive online correction and model iteration mechanism, comparing the prediction results of the agent model with at least one pipeline characteristic to generate a prediction error; when the cumulative value or sliding average value of the prediction error exceeds a preset dynamic threshold, controlling the iterative training of the circulating water system to generate new operating data, and using the new operating data to optimize the weights of the long-short-term memory neural network and the regression model of the circulating water system.

[0092] Among them, embodiments of the present invention can utilize accumulated rich historical data to complete the "cold start" training of the model and establish an initial baseline. During online operation, the system will continuously compare and analyze the model's prediction results with the actual pipeline characteristics identified in real time through multi-source data fusion to generate a prediction error. Once the cumulative value or sliding average of the prediction error exceeds the preset dynamic threshold, the system will automatically trigger the retraining process, controlling the iterative training of the circulating water system to generate new operating data. This process will use the latest, verified operating data to optimize and adjust the weights of the LSTM network and regression model.

[0093] The embodiment of the present invention ensures the long-term effectiveness of the model prediction accuracy through continuous learning capability, and enables the system to have strong robustness in coping with changing working conditions.

[0094] The present invention takes a coastal power plant as an example to introduce a specific technical solution embodiment.

[0095] This paper takes a 2x300MW direct cooling unit as the background and introduces in detail the specific implementation process and results of the energy-saving optimization technology of its circulating water system.

[0096] 1. Current status of circulating water system operation and optimization basis Before implementing optimization, the primary task was to accurately understand the actual operating characteristics of the circulating water system. Analysis revealed that the plant's circulating water pumps suffered from low pump efficiency and high power consumption. This was primarily due to significant discrepancies between the unit's actual piping characteristics and the circulating water pump's design and selection criteria, resulting in the pumps operating in an inefficient range of low head and high flow for extended periods. Furthermore, the system lacked effective flow control measures during low and medium load periods.

[0097] To address this issue, the present invention combines online monitoring with system modeling to provide a foundation for optimizing control strategies. By online monitoring of key parameters such as condenser heat load, circulating water flow, temperature, and pressure, a system model is established that reflects equipment status in real time, thereby determining the optimal circulating water pump operating mode.

[0098] This analysis enabled the power plant to obtain system operating data that more closely matched actual operating conditions than the design values, providing a solid basis for the subsequent development of variable frequency drive modifications and optimized control strategies. Traditional system design often relies on a theoretically designed pipeline characteristic curve. However, in actual operation, various factors such as valve opening, pipeline scaling, and water temperature fluctuations can cause dynamic changes in pipeline resistance.

[0099] The pipeline characteristic identification method proposed in this invention can accurately capture these actual changes. Figure 8 As shown in Figure 1, by identifying operating data under different working conditions, a series of realistic pipeline characteristic curves can be obtained. Therefore, formulating optimization strategies based on these high-precision pipeline characteristic identification results is a key prerequisite for achieving system energy conservation.

[0100] 2. Development and implementation of optimized control strategies Based on a precise analysis of the current operating status, an optimized operating strategy for the circulating water pump frequency conversion transformation was developed. Its goal is to flexibly adjust the circulating water flow rate under different loads and circulating water temperatures to find a balance between achieving the optimal vacuum of the unit and minimizing the pump energy consumption, thereby improving the net output of the unit. Table 1 shows the combined operating mode of the circulating water pump after optimization, as shown in Table 1: Table 1

[0101] 3. Project effectiveness and economic benefit analysis After the implementation of this optimization plan, it is expected to achieve significant comprehensive benefits, including reduced system energy consumption and enhanced operational flexibility, ultimately achieving the goal of energy conservation and consumption reduction and improved economic benefits. Table 2 is an estimated table of energy savings for each unit under different operating conditions, as shown in Table 2: Table 2

[0102] Table 3 is a summary (estimate) of the project's annual comprehensive economic benefits, as shown in Table 3: Table 3

[0103] In summary, based on an accurate analysis of circulating water system operational issues, the implementation of a "one-to-one" high-voltage variable-frequency speed regulation system, combined with an optimized control strategy focused on maximizing unit net output, effectively addresses the inefficiency of circulating water pumps caused by operating conditions deviating from their design points. This invention not only enhances system operational flexibility and affordability, but also provides a significant return on investment, representing an effective approach to energy conservation and emission reduction in power plant circulating water systems.

[0104] According to an embodiment of the present invention, a circulating water system energy-saving optimization control method based on pipeline characteristic identification is proposed. The method first models the pump pipeline characteristics, using a hybrid model that combines mechanism modeling with data modeling. Since pipeline resistance characteristics vary with operating conditions, accurate data such as water temperature, flow rate, and velocity are required. The method selects required power plant sensor data, preprocesses the data to increase the data volume, and then compares and selects appropriate data processing methods to calibrate key parameters and design benchmarks, reducing the impact of disturbances and reflecting real data. The preprocessed sensor data, calibrated key parameters and design benchmarks, circulating water flow rate, and predicted seawater temperature are input into the pipeline characteristic model to obtain pipeline characteristics that account for both thermal cycle and environmental influences. Based on the precise identification of pipeline characteristics, the circulating water pump operates at a variable frequency, reducing power plant energy consumption. This solves the problem of low pipeline characteristic prediction accuracy in most power plants, resulting in the inability of existing technologies to achieve optimal vacuum, increasing energy consumption and increasing costs.

[0105] Next, an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0106] Figure 9 It is a structural diagram of an energy-saving optimization control device for a circulating water system based on pipeline characteristic identification according to an embodiment of the present invention applied to a model building stage.

[0107] like Figure 9 As shown, the energy-saving optimization control device 10 for a circulating water system based on pipeline characteristic identification includes: an acquisition module 100 , a construction module 200 , an identification module 300 , a prediction module 400 and an establishment module 500 .

[0108] Specifically, the acquisition module 100 is used to construct a lumped parameter model of the pipeline according to the momentum equation of the fluid in the pipeline, and acquire at least one operating data of the circulating water system based on the lumped parameter model.

[0109] Construction module 200 is used to construct a proxy model of the thermal system mechanism model based on at least one operating data, generate a simulation data set covering multiple operating conditions according to the proxy model, and input at least one boundary condition of the predicted water temperature, circulating water flow and predicted load scheduling into the proxy model to output at least one performance indicator of the circulating water system.

[0110] The identification module 300 is configured to identify at least one pipeline characteristic of the circulating water system based on a simulation data set and at least one performance indicator.

[0111] Prediction module 400 is used to construct a long-short-term memory neural network model of the circulating water system based on at least one pipeline characteristic, historical multi-dimensional meteorological data and a preset time series to predict the seawater temperature and unit load during the target period and generate predicted seawater temperature and unit load.

[0112] Establishing module 500, for establishing a pipeline characteristic model for energy-saving optimization control of the circulating water system based on at least one sensor data, at least one key parameter, design reference point and predicted seawater temperature and unit load of the circulating water system.

[0113] Optionally, in one embodiment of the present invention, the identification module 300 includes: a generation unit, a calculation unit, a creation unit and an identification unit.

[0114] The generating unit is used to identify at least one item of operating data to generate an objective function of the circulating water system.

[0115] The calculation unit is used to calculate the flow field details and local resistance at any position in the pipeline according to the preset equations.

[0116] Establish a unit for establishing a prediction function containing time series features using long short-term memory networks.

[0117] The identification unit is used to identify at least one pipeline characteristic of the circulating water system according to the objective function, flow field details, local resistance and prediction function.

[0118] Optionally, in one embodiment of the present invention, the calculation formula of the objective function is:

[0119] in, is the measured pressure difference, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic; The calculation formula of the prediction function is:

[0120] in, is the nonlinear mapping function of LSTM, is the input feature, is the hidden state at the previous moment, The weights and biases for the trained parameters of the LSTM.

[0121] like Figure 10 As shown, the circulating water system energy-saving optimization control device 20 based on pipeline characteristic identification includes: a data acquisition module 600, a generation module 700 and an optimization module 800.

[0122] Specifically, the data acquisition module 600 is used to obtain real-time monitoring data of the circulating water system, predicted seawater temperature and unit load.

[0123] Generation module 700 is used to input real-time monitoring data, predicted seawater temperature and unit load into a pre-built pipeline characteristic model, so as to use the pre-built pipeline characteristic model to identify the pipeline characteristic curve of mechanism-data fusion, match at least one pipeline characteristic with the pipeline characteristic curve, generate matching data, and generate a water pump operation combination that meets preset optimal conditions and the operating frequency corresponding to the water pump operation combination based on the matching data, wherein the pre-built pipeline characteristic model is constructed by the mechanism model and the data model.

[0124] The optimization module 800 is used to establish a closed-loop adaptive online correction and model iteration mechanism for the circulating water system based on the water pump operation combination and operation frequency, and based on the closed-loop adaptive online correction and model iteration mechanism, use new operation data to optimize the weights of the long-short-term memory neural network model and regression model of the circulating water system.

[0125] Optionally, in one embodiment of the present invention, the generation module 700 includes: a traversal unit.

[0126] Among them, the traversal unit is used to traverse all potential operating conditions under different water pump combinations and different variable frequency based on matching data, so as to generate a water pump operation combination that meets the preset optimal conditions and the operating frequency corresponding to the water pump operation combination under the condition that the circulating water system meets the preset cooling requirements.

[0127] Optionally, in one embodiment of the present invention, the optimization module 800 includes: a comparison unit and an optimization unit.

[0128] The comparison unit is used to compare the prediction result of the proxy model with at least one pipeline characteristic based on a closed-loop adaptive online correction and model iteration mechanism to generate a prediction error.

[0129] The optimization unit is used to control the iterative training of the circulating water system to generate new operating data when the cumulative value or sliding average value of the prediction error exceeds a preset dynamic threshold, and use the new operating data to optimize the weights of the long and short-term memory neural network and regression model of the circulating water system.

[0130] It should be noted that the above-mentioned explanation of the embodiment of an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification applied to the model construction stage is also applicable to the energy-saving optimization control device 10 for a circulating water system based on pipeline characteristic identification in this embodiment. The above-mentioned explanation of the embodiment of an energy-saving optimization control method for a circulating water system based on pipeline characteristic identification applied to the model application stage is also applicable to the energy-saving optimization control device 20 for a circulating water system based on pipeline characteristic identification in this embodiment, which will not be repeated here.

[0131] According to an embodiment of the present invention, a circulating water system energy-saving optimization control device based on pipeline characteristic identification can be based on a hybrid modeling of mechanism models and data models, verified by measurement point data, taking into account the multiple influences of thermal cycle side data and environmental data, and fully considering factors such as pipeline resistance and equipment resistance, thereby achieving high-precision pipeline characteristic identification. By predicting future operating conditions such as seawater temperature and combining real-time pipeline characteristics, forward-looking optimization is carried out to maximize the net power of the power station and achieve closed-loop variable frequency optimization control. This solves the problem in the related art that the existing technology of most power stations has low accuracy in predicting pipeline characteristics, resulting in the technology currently used by most power stations often failing to achieve optimal vacuum, thereby increasing the energy consumption of the power stations and increasing costs.

[0132] Figure 11 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device may include: A memory 1101 , a processor 1102 , and a computer program stored in the memory 1101 and executable on the processor 1102 .

[0133] When the processor 1102 executes the program, the energy-saving optimization control method for a circulating water system based on pipeline characteristic identification provided in the above embodiment is implemented.

[0134] Furthermore, the electronic device further includes: The communication interface 1103 is used for communication between the memory 1101 and the processor 1102 .

[0135] The memory 1101 is used to store computer programs that can be run on the processor 1102 .

[0136] The memory 1101 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0137] If the memory 1101, processor 1102, and communication interface 1103 are implemented independently, the communication interface 1103, memory 1101, and processor 1102 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0138] Optionally, in a specific implementation, if the memory 1101, the processor 1102 and the communication interface 1103 are integrated on a chip, the memory 1101, the processor 1102 and the communication interface 1103 can communicate with each other through an internal interface.

[0139] The processor 1102 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0140] This embodiment also provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the above-mentioned energy-saving optimization control method for a circulating water system based on pipeline characteristic identification is implemented.

[0141] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction.

[0142] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "N" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0143] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or N executable instructions for implementing a custom logical function or step of a process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0144] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" is any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (not exhaustive) of computer-readable media include: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing it in other suitable ways as necessary, and then storing it in a computer memory.

[0145] It should be understood that various components of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0146] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0147] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0148] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and are not to be construed as limiting the present invention. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A circulating water system energy-saving optimization control method based on pipeline characteristic identification, characterized in that: Applied to the model building phase, it includes the following steps: Constructing a lumped parameter model of the pipeline according to a momentum equation of the fluid in the pipeline, and obtaining at least one operating data of the circulating water system based on the lumped parameter model; Based on the at least one item of operating data, a proxy model of the thermal system mechanism model is constructed, a simulation data set covering multiple operating conditions is generated according to the proxy model, and at least one boundary condition among the predicted water temperature, the circulating water flow rate, and the predicted load scheduling is input into the proxy model to output at least one performance indicator of the circulating water system; identifying at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator; Based on the at least one pipeline characteristic, a long short-term memory neural network model of the circulating water system is constructed according to historical multi-dimensional meteorological data and a preset time series to predict the seawater temperature and unit load during a target period, thereby generating a predicted seawater temperature and unit load; Based on at least one sensor data, at least one key parameter, a design reference point, and the predicted seawater temperature and unit load of the circulating water system, a pipeline characteristic model for energy-saving optimization control of the circulating water system is established.

2. The energy-saving optimization control method for a circulating water system based on pipeline characteristic identification according to claim 1 is characterized in that: The identifying at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator includes: Identifying the at least one item of operating data to generate an objective function of the circulating water system; Calculating flow field details and local resistance at any position in the pipeline according to preset equations; Use long short-term memory networks to establish prediction functions that include time series features; At least one pipeline characteristic of the circulating water system is identified based on the objective function, the flow field details, the local resistance, and the prediction function.

3. The energy-saving optimization control method for a circulating water system based on pipeline characteristic identification according to claim 2 is characterized in that: The calculation formula of the objective function is: in, is the measured pressure difference, For static lift, is the comprehensive resistance coefficient, For multiple groups of traffic; The calculation formula of the prediction function is: in, is the nonlinear mapping function of the long short-term memory neural network, is the input feature, is the hidden state at the previous moment, The weights and biases of the trained parameters of the long short-term memory neural network.

4. A circulating water system energy-saving optimization control method based on pipeline characteristic identification, characterized in that: Applied to the model application phase, it includes the following steps: Obtain real-time monitoring data of the circulating water system, predicted seawater temperature and unit load; Inputting the real-time monitoring data, the predicted seawater temperature, and the unit load into a pre-constructed pipeline characteristic model, using the pre-constructed pipeline characteristic model to identify a mechanism-data fusion pipeline characteristic curve, matching at least one pipeline characteristic with the pipeline characteristic curve to generate matching data, and generating a water pump operation combination that meets a preset optimal condition and an operating frequency corresponding to the water pump operation combination based on the matching data, wherein the pre-constructed pipeline characteristic model is constructed by a mechanism model and a data model; Based on the water pump operation combination and the operation frequency, a closed-loop adaptive online correction and model iteration mechanism of the circulating water system is established, and based on the closed-loop adaptive online correction and model iteration mechanism, the weights of the long-short-term memory neural network model and the regression model of the circulating water system are optimized using new operation data.

5. The energy-saving optimization control method for a circulating water system based on pipeline characteristic identification according to claim 4 is characterized in that: Generating a water pump operation combination that meets a preset optimal condition and an operation frequency corresponding to the water pump operation combination according to the matching data includes: Based on the matching data, all potential operating conditions under different water pump combinations and different variable frequency are traversed to generate the water pump operating combination that meets the preset optimal conditions and the operating frequency corresponding to the water pump operating combination under the condition that the circulating water system meets the preset cooling requirements.

6. The energy-saving optimization control method for a circulating water system based on pipeline characteristic identification according to claim 4 is characterized in that: The method of optimizing the weights of the long-short-term memory neural network model and the regression model of the circulating water system using new operating data based on the closed-loop adaptive online correction and model iteration mechanism includes: Based on the closed-loop adaptive online correction and model iteration mechanism, comparing the prediction result of the proxy model with the at least one pipeline characteristic to generate a prediction error; When the cumulative value or sliding average value of the prediction error exceeds a preset dynamic threshold, the circulating water system is controlled to iteratively train to generate new operating data, and the new operating data is used to optimize the weights of the long short-term memory neural network and regression model of the circulating water system.

7. A circulating water system energy-saving optimization control device based on pipeline characteristic identification, characterized in that: A circulating water system energy-saving optimization control method based on pipeline characteristic identification according to any one of claims 1 to 3 is applied in the model building stage, comprising: an acquisition module, configured to construct a lumped parameter model of the pipeline according to a momentum equation of the fluid in the pipeline, and acquire at least one item of operating data of the circulating water system based on the lumped parameter model; a construction module, configured to construct a proxy model of the thermal system mechanism model based on the at least one item of operating data, generate a simulation data set covering multiple operating conditions according to the proxy model, and input at least one boundary condition among the predicted water temperature, circulating water flow rate, and predicted load scheduling into the proxy model to output at least one performance indicator of the circulating water system; an identification module, configured to identify at least one pipeline characteristic of the circulating water system based on the simulation data set and the at least one performance indicator; a prediction module configured to construct a long-short-term memory neural network model of the circulating water system based on the at least one pipeline characteristic, historical multi-dimensional meteorological data, and a preset time series, so as to predict the seawater temperature and unit load during a target period and generate predicted seawater temperature and unit load; A module is established for establishing a pipeline characteristic model for energy-saving optimization control of the circulating water system based on at least one sensor data, at least one key parameter, a design reference point, and the predicted seawater temperature and unit load of the circulating water system.

8. A circulating water system energy-saving optimization control device based on pipeline characteristic identification, characterized in that: A circulating water system energy-saving optimization control method based on pipeline characteristic identification as described in any one of claims 4 to 6 is applied in the model application stage, comprising: Data acquisition module, used to obtain real-time monitoring data of the circulating water system, predicted seawater temperature and unit load; a generation module, configured to input the real-time monitoring data, the predicted seawater temperature, and the unit load into a pre-constructed pipeline characteristic model, identify a mechanism-data fusion pipeline characteristic curve using the pre-constructed pipeline characteristic model, match at least one pipeline characteristic with the pipeline characteristic curve to generate matching data, and generate, based on the matching data, a water pump operation combination that satisfies a preset optimal condition and an operating frequency corresponding to the water pump operation combination, wherein the pre-constructed pipeline characteristic model is constructed from a mechanism model and a data model; An optimization module is used to establish a closed-loop adaptive online correction and model iteration mechanism for the circulating water system based on the water pump operation combination and the operation frequency, and based on the closed-loop adaptive online correction and model iteration mechanism, use new operation data to optimize the weights of the long short-term memory neural network model and the regression model of the circulating water system.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a circulating water system energy-saving optimization control method based on pipeline characteristic identification as described in any one of claims 1 to 3 or a circulating water system energy-saving optimization control method based on pipeline characteristic identification as described in any one of claims 4 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement a circulating water system energy-saving optimization control method based on pipeline characteristic identification as described in any one of claims 1-3 or a circulating water system energy-saving optimization control method based on pipeline characteristic identification as described in any one of claims 4-6.

Images