Real-time optimization automatic control method, system, equipment and medium for desuperheating water system of power plant

Abstract

The invention discloses a real-time optimization automatic control method, system, equipment and medium for a power plant desuperheating water system, and belongs to the technical field of real-time optimization of the power plant desuperheating water system.The method comprises the steps that a physical model of the desuperheating water system is built based on the law of conservation of mass and the law of conservation of energy; mapping the physical model to a digital twin model, fusing historical and real-time data, constructing a three-dimensional thermal field model, and generating an octree data structure to dynamically divide physical space control blocks; setting abnormal judgment conditions, calculating temperature and pressure variation, and detecting whether the attemperation water system is abnormal or not; recursive search is carried out based on an octree structure, and an anomaly occurrence area is positioned; and a PID control algorithm is adopted to generate a control vector, and running parameters of the attemperation water system are adjusted to realize real-time optimization control. According to the method, abnormal node information is recorded in time, accurate control over the running state of the attemperation water system is achieved, and faults are effectively prevented.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time optimization of a desuperheated water system of a power plant, and in particular to a real-time optimization automatic control method, system, equipment and medium for a desuperheated water system of a power plant. Background Art

[0002] With the rapid development of the power industry, thermal power generation, as one of the main sources of electricity, has become particularly important in terms of efficiency, stability and safety of its production process. In thermal power plants, the attemperation water system is a key link in ensuring stable steam temperature and preventing equipment from overheating and damage. The attemperation water system can timely reduce the steam temperature, effectively protect the boiler superheater, prevent equipment damage caused by overheating, extend equipment service life, and reduce maintenance costs. By precisely controlling the attemperation water flow rate, the main steam temperature can be maintained within the set range, ensuring steam quality, meeting the operating requirements of subsequent equipment such as turbines, and improving the efficiency of the entire power generation system. Stable steam temperature helps maintain pressure balance in the boiler, prevents safety accidents caused by excessive pressure fluctuations, and ensures the safe and stable operation of the power plant.

[0003] Traditional control methods often rely on manual experience or simple PID control, and have a slow response speed to changes in steam temperature. It is difficult to adjust the cooling water flow in time, resulting in large fluctuations in steam temperature. Due to the lack of precise control algorithms and real-time monitoring methods, traditional control cannot achieve high-precision regulation of the cooling water flow, affecting the control effect of steam temperature. Faced with different operating conditions, such as load fluctuations, changes in fuel types, etc., traditional control systems often show poor adaptability and cannot effectively cope with complex operating environments. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: how to solve the problem that traditional control methods often rely on manual experience or simple PID control, have a slow response speed to changes in steam temperature, and are difficult to adjust the cooling water flow in time, resulting in large fluctuations in steam temperature. Due to the lack of precise control algorithms and real-time monitoring means, traditional control is difficult to achieve high-precision regulation of the cooling water flow, affecting the control effect of steam temperature. Faced with different operating conditions, such as load fluctuations, changes in fuel types, etc., traditional control systems often show poor adaptability and cannot effectively cope with complex operating environments.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a real-time optimization and automatic control method for a cooling water system in a power plant, which includes the following steps: based on the law of conservation of mass and the law of conservation of energy, constructing a physical model of the cooling water system, confirming the heat exchange process between steam and cooling water and the boundary of the cooling water system; mapping the physical model to a digital twin model, integrating historical and real-time data, constructing a three-dimensional thermal field model, and generating an octree data structure to dynamically divide the physical space control blocks; setting abnormality judgment conditions, calculating temperature and pressure changes, and detecting whether there are abnormalities in the cooling water system; performing recursive search based on the octree structure to locate the area where the abnormality occurs; defining an error function based on the set temperature and pressure targets, using the PID control algorithm to generate a control vector, and adjusting the operating parameters of the cooling water system to achieve real-time optimization control.

[0007] As a preferred solution of the real-time optimization automatic control method of the power plant cooling water system described in the present invention, the physical model of the cooling water system is constructed, including establishing a thermodynamic process model of steam, cooling water and mixing area based on the mass flow conservation relationship and energy flow conservation relationship between the steam side and the cooling water side, and setting the energy exchange equations of the steam inlet, outlet and mixing point.

[0008] As a preferred solution of the real-time optimization and automatic control method of the power plant cooling water system described in the present invention, the mapping of the physical model to the digital twin model includes dynamically correcting the pressure field matrix and temperature gradient tensor in the three-dimensional thermal field model based on real-time collected data, and synchronously updating the thermodynamic parameters in the octree nodes.

[0009] The present invention ensures that the digital twin model reflects the actual operating status of the current cooling water system by continuously correcting the thermal field model parameters and node data, thereby improving the timeliness of anomaly detection and the accuracy of anomaly positioning.

[0010] As a preferred solution of the real-time optimization automatic control method of the power plant cooling water system described in the present invention, the positioning of the abnormality occurrence area includes calculating the temperature change rate and pressure change rate in each control block respectively. When any change rate exceeds the preset threshold, a depth-first search method is used to recursively locate the abnormal node in the octree structure.

[0011] As a preferred solution of the real-time optimization automatic control method of the power plant cooling water system described in the present invention, the detection of whether there is an abnormality in the cooling water system includes calculating the change rate of the temperature data and pressure data collected in two consecutive sampling periods, setting a corresponding change rate threshold in each control block divided by the thermal field, and when it is detected that the temperature change rate or the pressure change rate exceeds the set threshold, further combining the trend of the steam flow change and the cooling water flow change to comprehensively judge the abnormal state.

[0012] As a preferred solution of the real-time optimization automatic control method of the power plant cooling water system described in the present invention, the recursive search based on the octree structure to locate the area where the abnormality occurs includes using a depth-first search method to recursively traverse the octree structure starting from the root node, and recording the temperature change rate, pressure change rate and flow change trend corresponding to each node during the traversal process. When there are multiple child nodes that meet the abnormal conditions in succession, the boundary of the area where the abnormality occurs is determined, and an abnormal trend record is generated.

[0013] By dynamically recording node states and performing trend analysis during recursive traversal, the present invention can predict potential abnormal trends of expansion earlier, implement pre-emptive intervention in abnormal handling, and improve the security and stability of system operation.

[0014] As a preferred solution of the real-time optimization automatic control method of a power plant cooling water system described in the present invention, the PID control algorithm is used to generate the control vector, including calculating the adjustment priority of the temperature control vector and the pressure control vector respectively according to the abnormal area information obtained by abnormal location, setting the distribution ratio of the control amount increment according to the node weight, and generating local dynamic adjustment instructions for the abnormal area to adjust the cooling water injection amount and steam delivery amount.

[0015] The present invention sets control priorities and node weight distribution mechanisms, so that control adjustments can be processed in layers according to the severity of the abnormality, achieving a balance between local response and overall optimization, and improving the flexibility and accuracy of the cooling water system regulation.

[0016] Another object of the present invention is to provide a real-time optimization automatic control system for a power plant desuperheating water system.

[0017] To solve the above technical problems, the present invention provides the following technical solutions: a real-time optimization automatic control system for a power plant's cooling water system, comprising: a physical model construction module, a digital twin model establishment module, an anomaly detection module, an anomaly location module, and a control vector generation module; the physical model construction module is used to construct a physical model of the cooling water system based on the law of conservation of mass and the law of conservation of energy, and to confirm the heat exchange process between steam and cooling water and the boundary of the cooling water system; the digital twin model establishment module is used to map the physical model to the digital twin model, integrate historical and real-time data, construct a three-dimensional thermal field model, and generate an octree data structure to dynamically divide the physical space control blocks; the anomaly detection module is used to set anomaly judgment conditions, calculate temperature and pressure changes, and detect whether there is an anomaly in the cooling water system; the anomaly location module is used to perform recursive search based on the octree structure to locate the area where the anomaly occurs; the control vector generation module is used to define an error function based on set temperature and pressure targets, generate a control vector using a PID control algorithm, and adjust the operating parameters of the cooling water system to achieve real-time optimization control.

[0018] The present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program and is characterized in that when the processor executes the computer program, the steps of the real-time optimization automatic control method of a power plant desuperheating water system are implemented.

[0019] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the real-time optimization automatic control method of a power plant desuperheating water system are implemented.

[0020] The beneficial effects of the present invention are as follows: the present invention can realize comprehensive and accurate monitoring of the system, grasp key parameters such as steam temperature, pressure, flow, cooling water flow, temperature and unit load in real time, and detect anomalies in time. By optimizing the control algorithm, the cooling water flow can be accurately adjusted to ensure the stability of steam temperature, improve energy utilization efficiency, and reduce energy consumption. The digital twin model helps to quickly locate the fault point and shorten the maintenance time. The use of the gossip tree data structure can quickly locate system anomalies. By calculating the average node temperature, pressure and change rate in real time and comparing them with abnormal conditions, the abnormal node information can be recorded in time to achieve accurate control of the operating status of the cooling water system and effectively prevent the occurrence of faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 The present invention provides an overall flow chart of a real-time optimization automatic control method for a power plant desuperheating water system according to an embodiment of the present invention.

[0023] Figure 2 A system solution module diagram of a real-time optimization automatic control system for a power plant desuperheating water system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0024] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0025] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a real-time optimization automatic control method for a desuperheating water system in a power plant, comprising:

[0026] S1. Based on the laws of conservation of mass and energy, construct a physical model of the desuperheated water system to confirm the heat exchange process between steam and desuperheated water and the boundaries of the desuperheated water system.

[0027] S2. Map the physical model to the digital twin model, integrate historical and real-time data, build a three-dimensional thermal field model, and generate an octree data structure to dynamically divide the physical space control blocks.

[0028] S3. Set abnormality judgment conditions, calculate temperature and pressure changes, and detect whether there is any abnormality in the cooling water system.

[0029] S4. Perform recursive search based on the octree structure to locate the area where the abnormality occurs.

[0030] S5. Based on the set temperature and pressure targets, define the error function, use the PID control algorithm to generate the control vector, and adjust the operating parameters of the desuperheating water system to achieve real-time optimization control.

[0031] When executing step S1, the mass flow conservation relationship and energy flow conservation relationship on the steam side and the desuperheating water side are expressed by establishing a mathematical model, and the energy balance relationship of the steam inlet, outlet and mixing point is defined respectively to accurately describe the heat exchange process.

[0032] When executing step S2, the digital twin model collects historical operating data and real-time monitoring data of the power plant, dynamically corrects the temperature gradient tensor and pressure field matrix in the three-dimensional thermal field, and uses an octree structure to recursively divide the physical space. Each child node corresponds to a specific thermal parameter area to achieve fine-grained management.

[0033] When executing step S3, the temperature change and the pressure change are calculated based on the difference values of the data of two consecutive monitoring cycles, and combined with the set temperature error threshold and pressure error threshold to determine whether the abnormal trigger condition is met.

[0034] When executing step S4, a depth-first search algorithm is used to recursively traverse the octree structure. By judging the temperature change rate and pressure change rate of each node, the specific location of the anomaly is determined, and the information of the abnormal node is recorded for subsequent analysis.

[0035] When executing step S5, the temperature error and pressure error are defined respectively according to the node information of the abnormal location, and the corresponding PID control quantity increments are calculated respectively. By setting different node weights, the control quantity is distributed to each execution unit of the cooling water regulation system to achieve priority response of the abnormal area and overall load balance adjustment.

[0036] Through the above steps, this embodiment can achieve real-time modeling of the power plant's cooling water system, accurate detection and positioning of anomalies, and adjust operating parameters through dynamic control vectors, thereby effectively improving the operating stability and control accuracy of the cooling water system.

[0037] Example 2, reference Figure 1 , which is the second embodiment of the present invention, provides a real-time optimization automatic control method for a power plant cooling water system based on the above embodiment.

[0038] In the embodiment of the present application, the physical model of the desuperheated water system is constructed in step S1. Based on the law of conservation of mass and energy, a thermodynamic process model of steam and desuperheated water in the desuperheated water system is established, the boundary range of the desuperheated water system is clarified, and the steam pipe, desuperheated water pipe and mixing area involved in heat exchange are defined. According to the law of conservation of mass, for the steam side, under stable conditions, the steam mass flow rate m entering the desuperheated water system per unit time is s,in Equal to the steam mass flow rate out of the desuperheating water system m s,out , that is, m s,in =m s,outFor the cooling water side, the cooling water mass flow rate m entering the cooling water system per unit time is w,in Equal to the mass change of the desuperheated water in the desuperheated water system. When the desuperheated water is completely mixed and there is no leakage, m w,in All of them will be mixed with steam. The total mass flow rate of steam and cooling water after mixing is m total =m s,in +m w,in .

[0039] According to the law of conservation of energy, the energy brought by steam into the desuperheating water system can be expressed as:

[0040] E s,in =m s,in h s,in

[0041] Among them, E s,in Indicates the energy brought by steam into the desuperheating water system, h s,in The specific enthalpy of steam at the steam inlet is a function of temperature and pressure. The energy brought into the desuperheating water system by desuperheating water can be expressed as:

[0042] E w,in =m w,ih h w,in

[0043] Among them, E w,in It represents the energy brought into the cooling water system by cooling water, h w,in represents the specific enthalpy of the desuperheating water inlet, h w,in =c p,w T w,in , where c p,w represents the constant pressure specific heat capacity of desuperheated water, T w,in Indicates the inlet temperature of the desuperheated water and the total energy E of the mixed steam and desuperheated water total =m total +h total , where h total Represents the specific enthalpy after mixing. According to the law of conservation of energy, the total energy of steam and desuperheated water after mixing is E total =E s,in +E w,in When the desuperheated water and steam undergo sufficient heat exchange during the mixing process, the heat released by the steam is equal to the heat absorbed by the desuperheated water. The heat released by the steam can be expressed as:

[0044] Q s =m s,in c p,s (T s,in -T mix )

[0045] Among them, Q sIndicates the heat released by steam, c p,s represents the constant pressure specific heat capacity of steam, T s,in represents the steam inlet temperature, T mix The temperature after mixing is expressed as follows:

[0046] Q w =m w,in c p,w (T mix -T w,in )

[0047] Among them, Q w Indicates the heat absorbed by the cooling water, Q s =Q w The temperature after mixing is T mix The specific calculation formula is as follows:

[0048]

[0049] Among them, T mix It represents the mixed temperature. The mixed temperature calculation formula is used as the output parameter of the physical model construction module, which is used to set the initial temperature gradient of the three-dimensional thermal field model in the digital twin model and provide the target temperature value for the error function of the control vector generation module.

[0050] Basic thermodynamic parameters are calculated using the mass and energy conservation formulas, and the above parameters are mapped into a three-dimensional thermal field model as a benchmark for real-time monitoring. The changes in actual monitoring data are compared with the theoretical parameters (based on energy conservation) to trigger abnormality judgment. In the octree structure, the abnormal position is located by calculating the average temperature / pressure and change rate of the node area. Taking the theoretical parameters as the target, the control quantity is adjusted in combination with the PID algorithm to ensure that the system returns to an energy balance state.

[0051] In an optional embodiment, the physical model of the attemperating water system can be constructed by: relying solely on the law of conservation of energy, assuming that there is no obvious mass leakage in the mixing process and ignoring minor losses, the energy balance formula of the simple accumulation of the steam inlet energy and the attemperating water inlet energy is used to approximately deduce the mixing temperature, without separately considering the mass conservation processes on the steam side and the attemperating water side.

[0052] This method simplifies the modeling process and is suitable for application scenarios with high requirements for real-time computing and energy changes as the dominant characteristic, but there may be certain errors under complex operating conditions.

[0053] In another optional embodiment, the physical model of the cooling water system can be constructed by using a regression analysis method based on historical data training. By fitting a large amount of historical steam, cooling water and mixing area temperature and flow data, an empirical heat exchange prediction model is obtained to quickly estimate the temperature change trend after mixing without explicitly establishing conservation equations.

[0054] This method can reduce modeling dependence and improve prediction speed. It is suitable for application scenarios with extremely high real-time requirements but acceptable accuracy loss.

[0055] This implementation strictly adheres to the laws of conservation of mass flow and energy for both steam and attemperated water during physical modeling, incorporating the heat exchange process modeling in the actual mixing region. This ensures a true reflection of the thermodynamic dynamics of the attemperated water system under various operating conditions. This ensures initial model accuracy during the digital twin modeling phase, further improving the accuracy of anomaly detection and location, and providing highly reliable basic data for control vector generation.

[0056] This implementation effectively improves the system's real-time response capability, dynamic control accuracy, and overall operational reliability.

[0057] In the implementation mode of the present application, in step S2, the physical model is mapped to the digital twin model, and the historical operating data of the power plant is collected, covering the steam temperature, cooling water flow and load data under various operating conditions of different seasons and different power generation loads. The current status information of the cooling water system is collected in real time through sensors, including steam temperature, pressure, flow, cooling water flow, temperature, and unit load parameters. The collected data is formatted and pre-processed. The physical building of the power plant cooling water system is topographically mapped by 3D scanning technology, and the outline of the physical building is scanned for digital modeling to establish a digital twin model. After being scaled down in proportion, it is displayed in the established digital twin model to obtain a visual building simulation module with the same number as the physical building. The preprocessed data is integrated into the digital twin model, and a three-dimensional thermal field model corresponding to the actual cooling water system is constructed in the digital twin model. On the basis of the three-dimensional thermal field model, the physical space of the cooling water system is dynamically divided into L levels according to thermodynamic parameters, and a control block with a pressure-temperature gradient is generated. A three-dimensional thermal field octree T = (P, T, F) is constructed, where P represents the pressure field matrix, T represents the temperature gradient tensor, and the initial value of the temperature gradient tensor T is determined by the mixed temperature T output by the physical model construction module. mix The calculation generates, its tensor element T ijkIt represents the temperature gradient component at the three-dimensional coordinate (i, j, k), F represents the fluid dynamic parameter set, the root node of the three-dimensional thermal field octree corresponds to the main water supply pipeline of the entire plant, and each root node stores the corresponding thermodynamic parameters, status information, and the connection relationship with other nodes. According to the real-time monitoring data in each control block, the digital twin model parameters are updated in real time. A data monitoring unit is set in the digital twin model. The data monitoring unit continuously tracks the changes of the model parameters by obtaining the updated digital twin model parameter data in real time.

[0058] In an optional embodiment, the physical model can be mapped to the digital twin model by: adopting a simplified spatial modeling method based on rule division, directly dividing the physical space of the cooling water system into fixed grid areas, uniformly setting thermal parameters in each grid unit, ignoring local minor differences, and updating state parameters only based on the grid center point data without the need for dynamic recursive subdivision.

[0059] This method has a fast modeling speed and small computational complexity. It is suitable for scenarios that do not require high modeling accuracy but have high requirements for real-time response. However, it cannot accurately reflect fine-grained abnormal changes.

[0060] In another optional embodiment, the physical model can be mapped to the digital twin model by using a simplified one-dimensional thermal field modeling method, setting the temperature gradient change only along the direction of the steam main pipeline, treating the cooling water injection point as the disturbance input point, ignoring the complex lateral and longitudinal flows, establishing a one-dimensional dynamic thermal change model, and mapping it to the twin system by simplifying the data.

[0061] This method can greatly simplify the complexity of the model and is suitable for systems with simple pipeline structures and single control strategies, but its ability to reflect complex mixing processes and local anomalies is limited.

[0062] This implementation integrates historical data covering a wide range of operating conditions with real-time data during digital twin modeling. This approach, combined with precise three-dimensional physical modeling, employs an octree-based dynamic recursive subdivision of control blocks. This enables a spatially fine-grained and temporally continuous representation of the desuperheated water system's operational state. The introduction of the octree data structure enables rapid response to local thermal changes, supporting efficient anomaly detection and location.

[0063] Therefore, the present invention can significantly improve the digital twin model's fit to the actual system, anomaly detection sensitivity and positioning accuracy, and provide high-reliability support for subsequent real-time control optimization.

[0064] In the embodiment of the present application, the abnormality determination condition is set in step S3, and the changes in temperature and pressure are calculated in each update cycle of the data monitoring unit. The current time is t n , the previous moment tn-1 The specific calculation formula for temperature change is as follows:

[0065]

[0066] Where ΔT represents the temperature change, T n Indicates the temperature value at the current moment, T n-1 Indicates the temperature value obtained at the last moment, ∈ T The specific calculation formula for the temperature measurement error and the error term caused by the uncertainty of time interval measurement is as follows:

[0067]

[0068] Where ΔP represents the pressure change, ∈ P It represents the error term caused by the pressure measurement error and the uncertainty of the time interval at a uniform speed. It compares the calculated temperature change and pressure change to see if they meet the abnormal conditions. If they do, it enters the abnormal positioning process. If they do not, it continues to monitor the abnormality.

[0069] In an optional embodiment, the abnormality judgment conditions can be set and abnormality detection can be performed by directly using the set absolute value change threshold method to compare the temperature difference and pressure difference between the current moment and the previous moment in each data update cycle. If the absolute value of the difference exceeds a fixed set value, it is judged as an abnormality without the need to normalize the time interval or correct the measurement error.

[0070] This method is simple to calculate and is suitable for application scenarios with high real-time requirements but smooth system changes and small noise impact. However, the probability of misjudgment is high when there are changes in sampling frequency or large measurement noise.

[0071] In another optional embodiment, the abnormality judgment conditions can be set and abnormality detection can be performed by: performing statistical analysis of the mean and standard deviation of the temperature change and pressure change within a certain time window based on the sliding window statistics method. When it is detected that the mean deviates from the normal range or the standard deviation increases, it is determined to be a system abnormality.

[0072] This method can effectively suppress misjudgments caused by short-term abnormal fluctuations and is suitable for systems with slow change rates or periodic disturbances, but the response speed is relatively slow.

[0073] During anomaly detection, this implementation sets temperature and pressure change thresholds for each control block, calculates the rate of change based on measurement data from two consecutive moments, and comprehensively corrects for measurement errors. This allows for accurate perception and determination of changes in the desuperheated water system's operating status. By introducing a rate-of-change determination mechanism based on time interval normalization, it effectively adapts to data monitoring under varying sampling frequencies, improving the adaptability and reliability of anomaly detection.

[0074] Therefore, this embodiment can significantly improve the sensitivity and accuracy of anomaly detection, and provide an accurate trigger basis for subsequent anomaly positioning and real-time control.

[0075] In the embodiment of the present application, in step S4, a recursive search is performed based on the octree structure to locate the abnormality occurrence area, and an octree data structure is constructed in the three-dimensional thermal field model of the digital twin model. In the octree data structure, the root node is located, which represents the main water supply pipeline of the entire plant. The root node is used as the starting point of the search, the search queue is initialized, and a child node is selected from the root node, marked as visited, and pushed into the search stack to obtain the pressure-temperature gradient range of the control block of the node and the connected child node information. For the currently visited node, the average temperature and average pressure of the area represented by the node are calculated based on the data collected in its control block. If there are m temperature measurement points T1, T2, ..., T in the node area m , then the average temperature The specific calculation formula is as follows:

[0076]

[0077] Among them, T i It represents the temperature value measured at the i-th temperature measurement point in the node area. If there are n pressure measurement points P1, P2, ..., P in the node area, m , then the average pressure The specific calculation formula is as follows:

[0078]

[0079] Among them, P i Represents the pressure value measured at the i-th measurement point in the node area, and calculates the temperature change rate of the area represented by the current node and pressure change rate The temperature change rate The specific calculation formula is as follows:

[0080]

[0081] in, Indicates the rate of change of temperature and pressure The specific calculation formula is as follows:

[0082]

[0083] in, Represents the pressure change rate. The calculated temperature change rate and pressure change rate are compared with the preset abnormal conditions. When the abnormal conditions are met, the node is considered to be the location where the abnormality occurs, and the node information is recorded. When the abnormal conditions are not met, the child nodes are recursively searched. When the current node is not a leaf node, the abnormal condition judgment is repeated for the child nodes connected to it. A child node is popped from the search stack to determine whether it has been visited. If not, the search is continued. If all child nodes have been visited, the parent node of the current node is traced back to continue exploring other branches. The above process is repeated until the search stack is empty. All nodes that meet the abnormal conditions are recorded as the location where the abnormality occurs.

[0084] In an optional embodiment, the abnormal area is located based on the spatial grid structure. The following methods can be used: directly divide the entire physical space of the cooling water system according to a fixed cubic grid, use each fixed grid unit as an independent control block, traverse all grid units, and detect whether there are abnormal changes. If multiple consecutive adjacent grid units are found to be abnormal, they are marked as abnormal areas.

[0085] This method does not require tree structure management and has a simple structure, but when the spatial division is not fine-grained enough and the node density is uneven, the positioning accuracy and efficiency are low.

[0086] In another optional embodiment, the abnormal area positioning can be carried out by: based on the regional clustering algorithm, such as DBSCAN (density-based clustering method), clustering analysis is performed on the monitoring points with large abnormal changes in the thermal field, identifying dense abnormal point areas, and then calibrating the position of the abnormal area.

[0087] This method is suitable for scenarios where abnormal areas are densely distributed and have complex shapes. It can adaptively adjust the range of the abnormal area, but the computational complexity is high and the real-time performance is relatively poor.

[0088] During the anomaly location process, this embodiment recursively subdivides the physical space based on the octree structure and adopts a recursive traversal mechanism combining depth-first search with node state backtracking, which can quickly determine the abnormal node and its spatial boundary when a local thermal anomaly occurs.

[0089] By dynamically recording the node status and combining it with the abnormal condition judgment of continuous sub-nodes, abnormal trend records can be formed in time at the early stage of abnormal occurrence.

[0090] This implementation can significantly improve the real-time and accuracy of anomaly positioning, reduce positioning errors, and provide precise spatial targets for subsequent dynamic control.

[0091] In the embodiment of the present application, in step S5, a PID control algorithm is used to generate a control vector, and the node of the abnormal location recorded by the abnormality positioning module is obtained. Based on the temperature control and pressure control targets of the desuperheated water system, error functions are defined, including temperature error and pressure error. The specific calculation formula of the temperature error is as follows:

[0092] e T (t) = T est -T(t)

[0093] Among them, e T (t) represents the temperature error, T est represents the set temperature, T(t) represents the actual temperature, and the specific calculation formula for the pressure error is as follows:

[0094] e P (t) = P est -P(t)

[0095] Among them, e P (t) represents the pressure error, P est represents the set pressure, P(t) represents the actual pressure, and the increment of the control quantity is calculated according to the PID controller. The specific calculation formula of the increment of the control quantity Δu(t) is as follows:

[0096]

[0097] Among them, K p represents the proportional coefficient, e(t) represents the error function, which can be temperature error and pressure error, K i Indicates the integral coefficient, K d represents the differential coefficient, dt represents the time interval, de(t) represents the change in the error function, and the control quantity increment for temperature control is:

[0098]

[0099] Where Δu T (t) represents the control amount increment of temperature control, Indicates the proportional coefficient of temperature control, K iT Indicates the integral coefficient of temperature control, K dt represents the differential coefficient of temperature control, represents the cumulative error from the initial moment to the current moment t, Δt represents the sampling interval, e T (kΔt) represents the temperature error Δt at sampling time k. The control increment for pressure control is:

[0100]

[0101] Where Δu P (t) represents the control amount increment of pressure control, Indicates the proportional coefficient of pressure control, K iP Indicates the integral coefficient of pressure control, K dP Denotes the differential coefficient of pressure control, e T (kΔt) represents the pressure error of Δt at sampling time k. The increment of the control vector is mapped to the increment of each component of the control vector, and the increment of the PID controller output is reasonably distributed to achieve automatic control.

[0102] In an optional embodiment, the control vector can be generated according to the set target by using a simple proportional control method (P control), that is, the control quantity increment is generated only according to the current error multiplied by the proportional coefficient $K_p$, and the integral and differential terms are ignored to reduce the control calculation amount.

[0103] This method is suitable for scenarios with slow changes and low requirements for control accuracy, but it cannot effectively eliminate steady-state deviations and the system adjustment response is relatively slow.

[0104] In another optional embodiment, the control vector may be generated according to the set target by using a fuzzy control method, which uses a preset fuzzy rule table to directly infer the adjustment instruction based on the current error and the error change rate without the need for specific mathematical modeling.

[0105] This method is suitable for scenarios with complex system models or large uncertainty in changes. It has good adaptability, but the design and debugging complexity is high, and its stability depends on the quality of the rule settings.

[0106] During the control vector generation process, this embodiment dynamically calculates the temperature error and pressure error based on real-time monitoring data, adopts a complete PID control algorithm, combines abnormal location information, and dynamically allocates the control amount increment through node weights, which can achieve fast and accurate local control optimization for abnormal areas.

[0107] Through the proportional-integral-differential comprehensive adjustment, it can not only quickly respond to the system change trend, but also eliminate long-term steady-state errors and improve the overall stability of the system.

[0108] Therefore, this embodiment can significantly improve the dynamic response performance, abnormality handling efficiency and long-term operation reliability of the desuperheated water system.

[0109] Example 3 is the third embodiment of the present invention, which differs from the first two embodiments in that:

[0110] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0111] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered 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" can be 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.

[0112] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, 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 may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0113] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple 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, it can be implemented using a combination of any of the following technologies known in the art: a discrete logic circuit having logic gate circuits for implementing logic 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.

[0114] Example 4, reference Figure 2 , which is the fourth embodiment of the present invention, provides a real-time optimization automatic control system for a desuperheated water system in a power plant, comprising a physical model building module, a digital twin model building module, an anomaly detection module, an anomaly location module, and a control vector generation module;

[0115] The physical model construction module is used to construct a physical model of the desuperheated water system based on the laws of conservation of mass and energy, confirm the heat exchange process between steam and desuperheated water, and the boundaries of the desuperheated water system;

[0116] A digital twin model building module is used to map the physical model to the digital twin model, integrate historical and real-time data, build a three-dimensional thermal field model, and generate an octree data structure to dynamically divide the physical space control blocks;

[0117] The abnormality detection module is used to set abnormality judgment conditions, calculate temperature and pressure changes, and detect whether there are abnormalities in the desuperheating water system;

[0118] The anomaly location module is used to perform recursive search based on the octree structure to locate the area where the anomaly occurs;

[0119] The control vector generation module is used to define the error function according to the set temperature and pressure targets, generate the control vector using the PID control algorithm, and adjust the operating parameters of the desuperheated water system to achieve real-time optimization control.

[0120] Physical model construction module: Based on the law of conservation of mass and energy, a thermodynamic process model of steam and attemperated water in the attemperated water system is established, and the system boundaries and the steam pipes, attemperated water pipes, and mixing areas involved in heat exchange are clearly defined to fully present the thermodynamic relationship of the system.

[0121] In this embodiment, it is necessary to specifically explain the physical model construction module. The physical model construction module establishes a thermodynamic process model of steam and desuperheated water in the desuperheated water system based on the law of conservation of mass and energy, clarifies the boundary range of the desuperheated water system, and defines the steam pipes, desuperheated water pipes, and mixing areas involved in heat exchange. According to the law of conservation of mass, for the steam side, under a stable state, the steam mass flow rate m entering the desuperheated water system per unit time is s,in Equal to the steam mass flow rate out of the desuperheating water system m s,out , that is, m s,in =m s,out For the cooling water side, the cooling water mass flow rate m entering the cooling water system per unit time is w,in Equal to the mass change of the desuperheated water in the desuperheated water system. When the desuperheated water is completely mixed and there is no leakage, m w,in All of them will be mixed with steam. The total mass flow rate of steam and cooling water after mixing is m total =m s,in +m w,in According to the law of conservation of energy, the energy brought by steam into the desuperheating water system can be expressed as:

[0122] E s,in =m s,in h s,in

[0123] Among them, E s,in Indicates the energy brought by steam into the desuperheating water system, h s,in The specific enthalpy of steam at the steam inlet is a function of temperature and pressure. The energy brought into the desuperheating water system by desuperheating water can be expressed as:

[0124] E w,in =m w,in h w,in

[0125] Among them, E w,in It represents the energy brought into the cooling water system by cooling water, h w,in represents the specific enthalpy of the desuperheating water inlet, h w,in =c p,w T w,in , where c p,w represents the constant pressure specific heat capacity of desuperheated water, T w,in Indicates the inlet temperature of the desuperheated water and the total energy E of the mixed steam and desuperheated water total =m total +h total , where h total Represents the specific enthalpy after mixing. According to the law of conservation of energy, the total energy of steam and desuperheated water after mixing is E total =E s,in +E w,inWhen the desuperheated water and steam undergo sufficient heat exchange during the mixing process, the heat released by the steam is equal to the heat absorbed by the desuperheated water. The heat released by the steam can be expressed as:

[0126] Q s =m s,in c p,s (T s,in -T mix )

[0127] Among them, Q s Indicates the heat released by steam, c p,s represents the constant pressure specific heat capacity of steam, T s,in represents the steam inlet temperature, T mix The temperature after mixing is expressed as follows:

[0128] Q w =m w,in c p,w (T mix -T w,in )

[0129] Among them, Q w Indicates the heat absorbed by the cooling water, Q s =Q w The temperature after mixing can be obtained as T mix The specific calculation formula is as follows:

[0130]

[0131] Among them, T mix It represents the mixed temperature. The mixed temperature calculation formula is used as the output parameter of the physical model construction module, which is used to set the initial temperature gradient of the three-dimensional thermal field model in the digital twin model and provide the target temperature value for the error function of the control vector generation module.

[0132] It should be noted that E total =E s,in +E w,in It expresses that in the desuperheated water system, the sum of the energy brought in by steam and the energy brought in by desuperheated water is equal to the total energy of the steam and desuperheated water after mixing. In the process of mixing steam and desuperheated water, energy is transferred and redistributed between them, but the total energy of the entire system remains unchanged. Through the calculation of total energy, we can calculate the specific enthalpy of the mixed substances when the inlet parameters of steam and desuperheated water are known, providing an important basis for analyzing and controlling the operation of the desuperheated water system.

[0133] Based on the laws of conservation of mass and energy, a thermodynamic model of the steam and cooling water mixing process is constructed to clarify the system boundaries and dynamic parameters to provide theoretical support: for example, the energy conservation equation provides initial parameters for the digital twin model and control algorithm. Through the constraints of physical laws, the deviation of purely data-driven models is avoided and the control complexity is reduced: the input-output relationship of the system is clarified to provide quantifiable control targets for subsequent modules.

[0134] Digital twin model establishment module: Map the physical model to the digital twin model, fuse data to build a three-dimensional thermal field model, divide the physical space control blocks of the cooling water system according to thermodynamic parameters, generate a gossip tree to store parameters, update the model parameters based on real-time monitoring data, and set up a data monitoring unit to track changes.

[0135] In this embodiment, what needs to be specifically explained is the digital twin model establishment module. The digital twin model establishment module maps the physical model in the physical model construction module to the digital twin model, collects the historical operation data of the power plant, covers the steam temperature, cooling water flow and load data under various working conditions in different seasons and different power generation loads, and collects the current status information of the cooling water system in real time through sensors, including steam temperature, pressure, flow, cooling water flow, temperature, and unit load parameters. The collected data is formatted and pre-processed, and the physical building of the power plant cooling water system is topographically mapped by 3D scanning technology for the entire power plant, and the physical building outline scan is completed for digital mapping. Digital modeling is performed to establish a digital twin model. After being scaled down in proportion, it is displayed in the established digital twin model to obtain a visual building simulation module with the same number as the physical building. The preprocessed data is integrated into the digital twin model. A three-dimensional thermal field model corresponding to the actual cooling water system is constructed in the digital twin model. On the basis of the three-dimensional thermal field model, the physical space of the cooling water system is dynamically divided into L levels according to thermodynamic parameters to generate control blocks with pressure-temperature gradients, and a three-dimensional thermal field octree T = (P, T, F) is constructed, where P represents the pressure field matrix, T represents the temperature gradient tensor, and the initial value of the temperature gradient tensor T is determined by the mixed temperature T output by the physical model construction module. mix The calculation generates, its tensor element T ijk It represents the temperature gradient component at the three-dimensional coordinate (i, j, k), F represents the fluid dynamic parameter set, the root node of the three-dimensional thermal field octree corresponds to the main water supply pipeline of the entire plant, and each root node stores the corresponding thermodynamic parameters, status information, and the connection relationship with other nodes. According to the real-time monitoring data in each control block, the digital twin model parameters are updated in real time. A data monitoring unit is set in the digital twin model. The data monitoring unit continuously tracks the changes of the model parameters by obtaining the updated digital twin model parameter data in real time.

[0136] It should be noted that the mixing temperature T mix As the output parameter of the three-dimensional thermal field model, the initial temperature gradient tensor T is generated through the spatial interpolation algorithm.

[0137] The physical model is mapped into a three-dimensional thermal field model, and the control blocks are dynamically divided using the octree data structure. The pressure field matrix P and temperature gradient tensor T are updated in real time. Historical data, real-time sensor data and 3D scanning models are integrated to generate a visual thermodynamic state distribution. Through octree hierarchical division, local abnormal areas can be quickly located, and the control block granularity can be dynamically adjusted according to thermodynamic parameters to balance computational efficiency and accuracy.

[0138] Anomaly detection module: In the digital twin model control algorithm, the anomaly judgment conditions are set by comparing whether the calculated temperature and pressure changes meet the anomaly conditions.

[0139] In this embodiment, the anomaly detection module needs to be specifically explained. The anomaly detection module sets the anomaly judgment condition in the control algorithm of the digital twin model, calculates the change in temperature and pressure in each update cycle of the data monitoring unit, and assumes that the current time is t n , the previous moment t n-1 The specific calculation formula for temperature change is as follows:

[0140]

[0141] Where ΔT represents the temperature change, T n Indicates the temperature value at the current moment, T n-1 Indicates the temperature value obtained at the last moment, ∈ T The specific calculation formula for the temperature measurement error and the error term caused by the uncertainty of time interval measurement is as follows:

[0142]

[0143] Where ΔP represents the pressure change, ∈ P It represents the error term caused by the pressure measurement error and the uncertainty of the time interval at a uniform speed. It compares the calculated temperature change and pressure change to see if they meet the abnormal conditions. If they do, it enters the abnormal positioning process. If they do not, it continues to monitor the abnormality.

[0144] In each data update cycle, the temperature change rate and pressure change rate are calculated and compared with the preset thresholds to trigger abnormal alarms. The measurement noise is filtered through the error term to avoid false alarms. Millisecond-level anomaly detection prevents system status deterioration. The deviation between the physical model prediction value and the measured value is combined to distinguish between real anomalies and transient interference.

[0145] Anomaly location module: The anomaly detection module and the anomaly location module are integrated into the three-dimensional thermal field model of the digital twin model, and anomaly judgment and location are realized based on the Bagua tree data structure.

[0146] In this embodiment, the abnormality positioning module needs to be specifically explained. The abnormality positioning module constructs an octree data structure in the digital twin model. In the octree data structure, the root node is located. The node represents the main water supply pipeline of the entire plant. The root node is used as the starting point of the search, the search queue is initialized, and a child node is selected from the root node, marked as visited, and pushed into the search stack. The pressure-temperature gradient range of the control block of the node and the connected child node information are obtained. For the currently visited node, the average temperature and average pressure of the area represented by the node are calculated based on the data collected in its control block. If there are m temperature measurement points T1, T2, ..., T in the node area m , then the average temperature The specific calculation formula is as follows:

[0147]

[0148] Among them, T i It represents the temperature value measured at the i-th temperature measurement point in the node area. If there are n pressure measurement points P1, P2, ..., P in the node area, m , then the average pressure The specific calculation formula is as follows:

[0149]

[0150] Among them, P i Represents the pressure value measured at the i-th measurement point in the node area, and calculates the temperature change rate of the area represented by the current node and pressure change rate The temperature change rate The specific calculation formula is as follows:

[0151]

[0152] in, Indicates the rate of change of temperature and pressure The specific calculation formula is as follows:

[0153]

[0154] in, Represents the pressure change rate. The calculated temperature change rate and pressure change rate are compared with the preset abnormal conditions. When the abnormal conditions are met, the node is considered to be the location where the abnormality occurs, and the node information is recorded. When the abnormal conditions are not met, the child nodes are recursively searched. When the current node is not a leaf node, the abnormal condition judgment is repeated for the child nodes connected to it. A child node is popped from the search stack to determine whether it has been visited. If not, the search is continued. If all child nodes have been visited, the parent node of the current node is traced back to continue exploring other branches. The above process is repeated until the search stack is empty. All nodes that meet the abnormal conditions are recorded as the location where the abnormality occurs.

[0155] Based on the octree structure, the system recursively searches child nodes from the root node, calculates the average temperature and pressure of each control block, compares abnormal conditions, records the coordinates of nodes that meet the abnormal conditions, narrows the scope of the abnormality from the entire system to specific pipelines or equipment, provides spatial location information, and guides operation and maintenance personnel to intervene quickly.

[0156] Control vector generation module: Based on the temperature and pressure control targets of the desuperheated water system, the temperature error and pressure error are defined. By reasonably allocating the increments of the PID controller output, they are mapped into increments of each component of the control vector to achieve automatic control of the desuperheated water system.

[0157] In this embodiment, the control vector generation module needs to be specifically explained. The control vector generation module obtains the node of the abnormality location recorded by the abnormality positioning module, and defines error functions based on the temperature control and pressure control targets of the desuperheated water system, including temperature error and pressure error. The specific calculation formula of the temperature error is as follows:

[0158] e T (t) = T est -T(t)

[0159] Among them, e T (t) represents the temperature error, T est Indicates the set temperature, which is determined by the mixed temperature calculation formula. T(t) indicates the actual temperature. The specific calculation formula for the pressure error is as follows:

[0160] e P (t) = P est -P(t)

[0161] Among them, e P (t) represents the pressure error, P est represents the set pressure, P(t) represents the actual pressure, and the increment of the control quantity is calculated according to the PID controller. The specific calculation formula of the increment of the control quantity Δu(t) is as follows:

[0162]

[0163] Among them, K p represents the proportional coefficient, e(t) represents the error function, which can be temperature error and pressure error, K i Indicates the integral coefficient, K d represents the differential coefficient, dt represents the time interval, de(t) represents the change in the error function, and the control quantity increment for temperature control is:

[0164]

[0165] Where Δu T (t) represents the control amount increment of temperature control, Indicates the proportional coefficient of temperature control, K iT Indicates the integral coefficient of temperature control, K dt represents the differential coefficient of temperature control, represents the cumulative error from the initial moment to the current moment t, Δt represents the sampling interval, e T (kΔt) represents the temperature error Δt at sampling time k. The control increment for pressure control is:

[0166]

[0167] Where Δu P (t) represents the control amount increment of pressure control, Indicates the proportional coefficient of pressure control, K iP Indicates the integral coefficient of pressure control, K dP Denotes the differential coefficient of pressure control, e T (kΔt) represents the pressure error of Δt at sampling time k. The increment of the control vector is mapped to the increment of each component of the control vector, and the increment of the PID controller output is reasonably distributed to achieve automatic control.

[0168] Based on the deviation between the set value and the actual value, the PID control increment is calculated, and the temperature and pressure control increments are mapped to actuator instructions such as valve opening and pump speed to achieve multi-parameter coordinated regulation. The system stability is maintained through adaptive adjustment of PID parameters, the cooling water flow rate is optimized, and energy waste caused by excessive spraying is avoided.

[0169] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A real-time optimization automatic control method for a desuperheating water system in a power plant, characterized by: include, Based on the laws of conservation of mass and energy, a physical model of the desuperheated water system was constructed to confirm the heat exchange process between steam and desuperheated water and the boundaries of the desuperheated water system. Map the physical model to the digital twin model, integrate historical and real-time data, build a three-dimensional thermal field model, and generate an octree data structure to dynamically divide the physical space control blocks; Set abnormality judgment conditions, calculate temperature and pressure changes, and detect whether there are abnormalities in the cooling water system; Perform recursive search based on the octree structure to locate the abnormality area; According to the set temperature and pressure targets, the error function is defined, the PID control algorithm is used to generate the control vector, and the operating parameters of the desuperheating water system are adjusted to achieve real-time optimization control.

2. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 1, characterized in that: The physical model of the desuperheating water system is constructed by establishing a thermodynamic process model of steam, desuperheating water and mixing area based on the mass flow conservation relationship and energy flow conservation relationship between the steam side and the desuperheating water side, and setting energy exchange equations for the steam inlet, outlet and mixing point.

3. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 2, characterized in that: The mapping of the physical model to the digital twin model includes dynamically correcting the pressure field matrix and temperature gradient tensor in the three-dimensional thermal field model based on real-time collected data, and synchronously updating the thermodynamic parameters in the octree node.

4. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 3, characterized in that: The method of locating the abnormality occurrence area includes calculating the temperature change rate and the pressure change rate in each control block respectively. When any change rate exceeds a preset threshold, a depth-first search method is used to recursively locate the abnormal node in the octree structure.

5. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 4, characterized in that: The detection of whether there is an abnormality in the cooling water system includes calculating the change rate of the temperature data and pressure data collected in two consecutive sampling periods, setting a corresponding change rate threshold in each control block divided by the thermal field, and when it is detected that the temperature change rate or the pressure change rate exceeds the set threshold, further combining the trend of the steam flow change and the cooling water flow change to comprehensively judge the abnormal state.

6. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 4, characterized in that: The recursive search based on the octree structure is performed to locate the abnormality area. include, Using a depth-first search method, the octree structure is recursively traversed starting from the root node. During the traversal process, the temperature change rate, pressure change rate and flow change trend corresponding to each node are recorded. When there are multiple consecutive child nodes that meet the abnormal conditions, the boundary of the area where the abnormality occurs is determined, and an abnormal trend record is generated.

7. The real-time optimization automatic control method for a power plant desuperheating water system according to claim 4, characterized in that: The PID control algorithm is used to generate the control vector, including: Based on the abnormal area information obtained from abnormality positioning, the adjustment priorities of the temperature control vector and the pressure control vector are calculated respectively, the distribution ratio of the control quantity increment is set according to the node weight, and local dynamic adjustment instructions are generated for the abnormal area to adjust the cooling water injection volume and steam delivery volume.

8. A real-time optimization automatic control system for a desuperheated water system in a power plant, using a real-time optimization automatic control method for a desuperheated water system in a power plant according to any one of claims 1 to 7, characterized in that: include: Physical model construction module, digital twin model establishment module, anomaly detection module, anomaly location module and control vector generation module; The physical model construction module is used to construct a physical model of the desuperheated water system based on the law of conservation of mass and the law of conservation of energy, and to confirm the heat exchange process between steam and desuperheated water and the boundary of the desuperheated water system; The digital twin model building module is used to map the physical model to the digital twin model, integrate historical and real-time data, build a three-dimensional thermal field model, and generate an octree data structure to dynamically divide the physical space control blocks; The abnormality detection module is used to set abnormality judgment conditions, calculate temperature and pressure changes, and detect whether there is an abnormality in the desuperheated water system; The anomaly locating module is used to perform recursive search based on the octree structure to locate the area where the anomaly occurs; The control vector generation module is used to define an error function based on set temperature and pressure targets, generate a control vector using a PID control algorithm, and adjust the operating parameters of the desuperheated water system to achieve real-time optimization control.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the real-time optimization automatic control method of the power plant desuperheating water system according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a real-time optimization automatic control method for a power plant desuperheating water system according to any one of claims 1 to 7 are implemented.

Images