Intelligent monitoring method for cold pump starting and cold pump closing scheduling of DCS thermal power plant

Abstract

The invention provides a DCS thermal power plant cold pump starting and cold pump closing scheduling intelligent monitoring method, and aims to solve the problems that a traditional DCS constant value alarm is high in false alarm rate, a static light indicator cannot reflect a fault evolution process, an operator is difficult to timely and accurately deal with the fault, and the like. Key parameters such as cold pump starting current, valve state and closed cold water temperature are monitored in real time, and when the parameters deviate from a model predicted value or meet a preset logic condition, graded alarm is triggered automatically, and operation guidance of starting / cutting off a cold pump or adding / cutting off a closed cooler is provided. Economic dispatching is achieved by setting the temperature threshold value. According to the invention, load and environment temperature fluctuation can be effectively coped with, power consumption of equipment is reduced, operation safety, economy and automation level are improved, and monitoring burden of operators is reduced.

Description

Technical Field

[0001] This invention relates to the field of industrial intelligent monitoring and fault early warning technology, and in particular to a smart monitoring method for DCS scheduling of cooling pumps and coolers in thermal power plants. Background Technology

[0002] The open cooling water system (open cooling system) of a thermal power plant is a core support system for ensuring the safe operation of main and auxiliary equipment. Its operating efficiency directly affects the plant's power consumption and equipment reliability. With the development of industrial process automation and intelligent operation and maintenance technologies, traditional DCS systems have evolved from single-parameter monitoring to multi-device collaborative control. However, existing technologies still rely mainly on setpoint alarms and manual experience-based judgment. Specifically, this system covers the entire process from cooling water supply and pump start-up and shutdown control to closed-loop cooler heat exchange efficiency monitoring. The open cooling pump operating status is determined by a single current threshold, while the closed-loop cooler scheduling is based on static judgment of the absolute value of the outlet header temperature, lacking dynamic modeling of the coupling relationship between equipment status and operating conditions.

[0003] However, existing monitoring methods directly employ fixed thresholds and isolated parameter analysis, failing to establish a dynamic decision-making system for multi-device collaborative optimization. This can lead to redundant equipment operation, alarm lag, and frequent false alarms. Specifically, the logic for determining the operating status of the cooling pump and the cooler does not consider the coupling characteristics between devices. When the system is in a single cooler-closed operation state (such as the outlet valve of cooler A), the logic may fail to consider this. And shut off the outlet valve of cooler B When heat exchange capacity is insufficient, it is difficult to accurately identify the root cause. Traditional scheduling strategies rely on human experience, leading to insufficient consistency in decision-making. Furthermore, they cannot achieve early warning of equipment performance degradation through multi-parameter correlation analysis, ultimately causing systemic problems such as wasted plant power and reduced equipment lifespan. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a smart monitoring method for the scheduling of cooling pumps and coolers in a DCS thermal power plant.

[0006] Another objective of this invention is to propose a smart monitoring device for the scheduling of cooling pumps and coolers in a DCS thermal power plant.

[0007] The third objective of this invention is to provide a computer device.

[0008] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0009] To achieve the above objectives, a first aspect of the present invention proposes a smart monitoring method for the scheduling of cooling pumps and coolers in a DCS thermal power plant, comprising: S1 acquires real-time operating data of the cooling pump current, outlet electric valve status, cooler outlet header temperature, and valve opening. S2, based on the real-time operating data, determine the operating status of the cooling pump and the cooling unit by combining electrical parameters and mechanical position parameters; S3 generates scheduling suggestions for starting and stopping the cooling pump or turning off the cooler based on the preset temperature threshold range and system operating status. S4 uses a multivariate state estimation model to dynamically verify the scheduling suggestions. It judges whether there are hidden anomalies by comparing the residuals of real-time parameters with the expected values ​​of the model, and outputs the optimized scheduling decision.

[0010] In one embodiment of the present invention, S2 includes: S21, based on the starting current of the cooling water pump With outlet electric valve The combinational logic determines that the cooling pump is in operation. S22, based on the starting current of the cooling water pump With outlet electric valve The combinational logic determines that the cooling pump is in a stopped state.

[0011] In one embodiment of the present invention, S3 includes: S31, when the outlet header temperature of the cooler is... Furthermore, when the system is in single-cooler operation mode, an alarm for adding a cooler is triggered and a high cooler water temperature is output, suggesting the addition of another cooler.

[0012] In one embodiment of the present invention, S4 includes: S41, Calculate the residuals between the real-time parameters and the model's expected values. If the absolute value of any residual exceeds the upper limit of the preset statistical confidence interval, or below the lower limit If so, then a latent anomaly is determined to exist.

[0013] To achieve the above objectives, a second aspect of the present invention provides a smart monitoring device for the scheduling of cooling pumps and coolers in a DCS thermal power plant, comprising: The real-time data acquisition module is used to acquire real-time operating data such as the current of the cooling pump, the status of the outlet electric valve, the temperature of the outlet main pipe of the cooler, and the valve opening. The operation status determination module is used to determine the operation status of the cooling pump and the cooling unit based on the real-time operation data by combining electrical parameters and mechanical position parameters. The scheduling suggestion generation module is used to generate scheduling suggestions for starting and stopping the cooling pump or turning off the cooler based on the preset temperature threshold range and system operating status. The dynamic verification module is used to dynamically verify the scheduling suggestions using a multivariate state estimation model. It determines whether there are hidden anomalies by comparing the residuals of real-time parameters with the expected values ​​of the model, and outputs the optimized scheduling decision.

[0014] The present invention discloses a smart monitoring method and device for scheduling open cooling pumps and closed cooling units in a DCS thermal power plant, which can realize economic scheduling and early fault warning of open cooling water systems in thermal power plants, effectively reduce plant power consumption, and improve operational safety and equipment reliability.

[0015] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a smart monitoring method for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant as described in the first aspect embodiment.

[0016] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements a smart monitoring method for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant as described in the first aspect embodiment.

[0017] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0018] 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 taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a smart monitoring method for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant according to an embodiment of the present invention. Figure 2 This is a schematic diagram of energy-saving scheduling operation of the cooling pump and the cooler according to an embodiment of the present invention; Figure 3 This is a structural diagram of a smart monitoring device for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant according to an embodiment of the present invention. Figure 4 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] The following describes, with reference to the accompanying drawings, a smart monitoring method and apparatus for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant according to an embodiment of the present invention.

[0022] Example 1 Figure 1 This is a flowchart of a smart monitoring method for scheduling the start-up and shut-down of cooling pumps and coolers in a DCS thermal power plant according to an embodiment of the present invention. Figure 1 As shown, it includes: S1 acquires real-time operating data of the cooling pump current, outlet electric valve status, outlet header temperature of the cooler, and valve opening.

[0023] Specifically, this step involves real-time data acquisition of key operating parameters in the open cooling water system of a thermal power plant, including the current of the start-up cooling pump, the status of the outlet electric valve, the temperature of the outlet header pipe of the shut-off cooler, and the valve opening. These parameters are the basic input data for constructing the system's two-layer intelligent architecture of "economic scheduling layer" and "health diagnosis layer," and their acquisition accuracy and real-time performance directly affect the accuracy of the system's judgment and response speed.

[0024] Furthermore, this step utilizes the standard measurement point interface in the power plant's DCS system, employing OPCServer or direct database connections (such as SQL Server, Oracle, etc.), to achieve millisecond-level real-time reading of the aforementioned parameters. The cooling pump current is acquired via a current transformer (CT), with a sampling frequency typically set to 1Hz to 10Hz to ensure it reflects the trend of equipment load changes. The status of the outlet electric valve is obtained from the digital input (DI) signal in the DCS, with a status value of 0 (closed) or 1 (open), conforming to the Boolean variable definition in the IEC 61131-3 standard. The outlet header temperature of the shut-off cooler is acquired by a PT100 platinum resistance temperature sensor installed on the header, with a sampling frequency of 1Hz and a temperature accuracy of ±0.3℃, conforming to the ASME B40.7 standard. The valve opening degree is converted into a percentage opening value through the feedback signal (4-20mA analog input) of the electric actuator, with an accuracy of ±1%.

[0025] Furthermore, this step is deployed within the DCS network, acquiring data through periodic polling or event-driven methods via an independent application server, and caching the data in an in-memory database (such as Redis) or a real-time database (such as PISystem) for subsequent rule judgment and MSET model analysis. During cold seasons or under low-load conditions, the system can use the temperature and equipment status data acquired through this step to determine whether the conditions for "pump switching" or "refrigerant switching" are met, thereby achieving optimized control of plant power consumption.

[0026] Furthermore, the system provides high-quality, real-time input data streams, which is a prerequisite for realizing intelligent monitoring and economic scheduling. Through the synchronous acquisition and standardized processing of multi-source heterogeneous data, the system can accurately identify the operating status of equipment, providing a reliable basis for subsequent rule triggering and model analysis, thereby improving the overall intelligence level and operational economy of the system.

[0027] S2, based on the real-time operating data, determine the operating status of the cooling pump and the cooling unit by combining electrical parameters and mechanical position parameters.

[0028] Specifically, based on the real-time operating data, the operating status of the cooling pump and the cooling unit is determined by combining electrical parameters and mechanical position parameters. This step constructs a multi-dimensional state judgment logic by fusing electrical signals (such as current) and mechanical position signals, thereby effectively improving the system's ability to identify the true operating status of the equipment.

[0029] Furthermore, this step uses AND gates to combine multiple key parameters to achieve a comprehensive determination of the equipment status. For example, for the "running" status of the chiller pump, the system determines the status by judging the "chiller pump current". "and "open the cold water pump outlet electric valve If both conditions (indicating activation) are met, the system is considered operational; otherwise, if the "cooling water pump current" condition is not met, the system is considered operational. "and the outlet electric valve" If the value "(indicates shutdown)" is displayed, the system is determined to be in a shutdown state. This combined judgment method effectively avoids the influence of interference or false alarms on a single parameter, thus improving the robustness of state recognition.

[0030] Furthermore, the electrical parameters involved in this step include, but are not limited to, the operating current of the cooling pump, the threshold of which is set as follows: This value is determined based on the no-load and full-load current characteristics of a typical start-up cooling pump, ensuring accurate differentiation between operating and shutdown states under different working conditions. Mechanical position parameters include the opening and closing status of electric valves, using binary signals. (Close) and The (on) indication conforms to the common status signal definition standards in DCS systems.

[0031] Furthermore, this step is widely applied in the real-time monitoring and scheduling decisions of open cooling water systems in thermal power plants. In the DCS system, by acquiring the electrical and mechanical signals of the open cooling pump and the closed cooling unit in real time, the system can immediately determine whether the equipment is in operation or shutdown state, providing accurate equipment status input for subsequent economic scheduling rules (such as rule 1 and rule 2), thereby supporting the start-up and shutdown suggestions of the open cooling pump and the activation and deactivation strategies of the closed cooling unit.

[0032] Furthermore, this step, through multi-parameter fusion judgment, significantly improves the accuracy and stability of equipment status identification, providing a reliable data foundation for the intelligent monitoring system. At the same time, its logic is clear and easy to implement, meeting the requirements of DCS system integration, and providing a standardized input interface for subsequent rule judgment and MSET model analysis, which is an important prerequisite for realizing intelligent and refined operation control of the system.

[0033] Furthermore, S2 includes: S21, based on the starting current of the cooling water pump With outlet electric valve The combinational logic determines that the cooling pump is in operation.

[0034] Specifically, the determination of the "operating" state of the chiller pump is based on a combination of electrical parameters and mechanical status logic, specifically: when the chiller pump is turned on, the current... And the outlet electric valve When the pump is turned on, it is determined that the cooling pump is in the "running" state. This logic design fully considers the typical characteristics of the cooling pump in actual operation. That is, during normal operation, the pump motor current is usually higher than the value in its no-load or shutdown state, and the open state of the outlet electric valve (i.e., the state value is 1) further confirms that the flow path of the cooling water has been established, thereby ensuring that the operation of the pump is practically meaningful.

[0035] Furthermore, this decision-making logic acquires the current signal of the cooling pump and the on / off status signal of the outlet electric valve through the real-time data acquisition module in the DCS system. The current signal is typically acquired by a current transformer (CT) installed in the pump motor circuit, with an accuracy class of 0.5 or higher, and a sampling frequency of 1~5 seconds / time to ensure real-time response to the pump's operating status. The status signal of the outlet electric valve is acquired through the digital input (DI) module of the PLC or DCS, and its status value is Boolean (0 indicates closed, 1 indicates open), which has high reliability and response speed.

[0036] Furthermore, the current threshold This setting is based on the typical current difference between the cooling pump and its operating conditions under no-load and load. Typically, the current value of the cooling pump when it is stopped or under no-load is lower than... During normal operation, the current value increases significantly as the motor drives the water pump. The setting of this threshold needs to be calibrated based on the rated current, starting current curve, and actual operating data of the specific pump model to ensure the accuracy of the logical judgment.

[0037] Furthermore, this combinational logic is embedded into the expert rule base of the intelligent monitoring system, serving as the basis for making decisions regarding the start / stop of the cooling pump. For example, it is used when the outlet header temperature of the shut-off cooler... When the system is activated, it will first check if a cooling pump is currently in "running" mode. If not, it will trigger a "start cooling pump" suggestion. This logic is applicable to different seasons and load conditions, and has good adaptability and robustness.

[0038] Furthermore, by fusing and judging multi-source signals, the system effectively avoids misjudgments that may arise from relying on a single signal (such as depending solely on current or valve status), thus improving the accuracy of equipment status identification. Simultaneously, this logic serves as the input condition for economic scheduling rules, providing reliable data support for subsequent energy-saving decisions and forming a key foundation for achieving intelligent and refined system operation.

[0039] S22, based on the starting current of the cooling water pump With outlet electric valve The combinational logic determines that the cooling pump is in a stopped state.

[0040] Specifically, this invention achieves accurate determination of the operating status of a turn-on cooling water pump by combining electrical parameters and mechanical position parameters. Specifically, the determination logic for "the turn-on cooling pump is in a 'stopped' state" is as follows: when the turn-on cooling water pump current is less than 50A and the outlet electric valve is closed (i.e., the value is 0), the system determines that the pump is in a stopped state. This logic is based on in-depth analysis of the operating characteristics of the turn-on cooling pump, combined with dual verification of current threshold and valve status, effectively avoiding misjudgments that may be caused by a single signal.

[0041] Furthermore, this decision-making logic is typically embedded in the logic configuration module or script engine of the DCS system. It performs Boolean logic operations by real-time acquisition of the cooling pump's current signal (in amperes, A) and the on / off status of the outlet electric valve (0 for closed, 1 for open). In practical engineering, the current signal usually comes from a frequency converter or current transformer, with a sampling frequency of 1 second / time and a data accuracy of ±0.5A, conforming to industrial communication protocol standards such as IEC60870-5-101 or Modbus. The valve status signal is acquired through the DI module of the PLC or DCS, exhibiting clear binarization characteristics.

[0042] Furthermore, the current threshold of 50A is set based on statistical analysis of measured data from typical start-up pumps under no-load or shutdown conditions. According to actual operating experience, when the pump is completely stopped, its current is typically less than 10% of the rated current. 50A, as an empirical threshold, can effectively distinguish between shutdown and low-load operating states. Additionally, a 0% status on the outlet electric valve indicates that no flow path has been established at the pump outlet, further confirming that the pump is not participating in system circulation.

[0043] Furthermore, this decision-making logic is widely applied in the status monitoring and scheduling decisions of cooling systems in thermal power plants. For example, during low-load operation in winter, the system uses this logic to determine whether there are too many pumps currently in operation, thereby triggering a "shut down cooling pumps" suggestion to achieve optimized control of plant power. Simultaneously, this status determination serves as the basic input to the expert rule base, providing accurate equipment status information for subsequent economic scheduling and anomaly diagnosis.

[0044] Furthermore, the technical effect of this step is to improve the accuracy and robustness of equipment status identification. By logically combining current and valve status, the system can effectively filter out misjudgments caused by current fluctuations or valve malfunctions, ensuring the reliability of scheduling recommendations. In addition, this judgment logic provides accurate equipment status labels for the MSET model, which helps improve the model training quality and anomaly detection accuracy, thereby achieving a technical upgrade from "absolute value alarm" to "behavioral anomaly early warning".

[0045] S3 generates scheduling suggestions for starting and stopping the cooling pump or turning off the cooler based on the preset temperature threshold range and system operating status.

[0046] Specifically, in some implementations, scheduling suggestions for starting and stopping the cooling pump or turning on and off the cooler are generated based on the preset temperature threshold range and system operating status. This step is based on a deep understanding of the operating characteristics of the open cooling water system of thermal power plants, combined with expert experience and real-time operating condition judgment, to achieve intelligent start-stop control of cooling equipment, thereby optimizing plant power consumption and improving the economic efficiency of system operation.

[0047] Furthermore, this step first relies on a combination of a preset temperature threshold and equipment status logic for judgment. For example, when the outlet header temperature of the shut-off cooler... And the chiller pump is currently in a "shutdown" state (i.e., the chiller pump current is low). And the outlet electric valve When the temperature of the cooling water is high, the system will trigger a "start cooling pump" suggestion, prompting operators to "start a cooling pump if the cooling water temperature is high." Conversely, when the outlet header temperature is low... And the cooling pump is in "running" mode (i.e., current is flowing). And the outlet electric valve When the system is in operation, it will suggest "disabling the cold pump" to avoid the phenomenon of "overpowering a small vehicle" and achieve energy-saving operation.

[0048] Furthermore, the system determines whether to add or remove coolers by analyzing the relationship between the number of currently operating shut-off coolers and the outlet header temperature. For example, when the outlet header temperature... Furthermore, when the system is in "single closed-circuit operation" mode, the system will recommend "adding a closed-circuit cooler"; when the temperature Furthermore, if the system is in "dual closed-circuit operation" mode, it is recommended to "disconnect one closed-circuit cooler" to reduce system resistance and reduce plant power consumption.

[0049] Furthermore, the system employs multi-dimensional signal fusion to determine equipment status, including key parameters such as current and valve opening. For example, determining the "operating" status of the cooling pump requires simultaneously satisfying the current... With outlet electric valve The "out of service" state requires current. With outlet electric valve These parameters are set based on the equipment's typical operating curves and actual working conditions to ensure the accuracy and robustness of the judgment.

[0050] Furthermore, this step is embedded in the power plant's DCS system, acquiring and processing analog and digital signals from the PLC or DCS in real time, and automatically triggering rules through configuration or script logic. System operators can intuitively obtain scheduling suggestions on the HMI interface to assist them in starting and stopping equipment, thereby achieving standardization and automation of operating strategies.

[0051] Furthermore, this step effectively avoids the inconsistencies and uneconomicalities caused by traditional manual experience-based scheduling by introducing precise temperature thresholds and equipment status logic. Its innovation lies in combining equipment operating status with system thermodynamic performance indicators to achieve dynamic scheduling decisions based on operating conditions. This provides a quantifiable and replicable intelligent control method for thermal power plant cooling systems, with significant energy-saving, consumption-reducing, and operational optimization value.

[0052] Furthermore, S3 includes: S31, when the outlet header temperature of the cooler is... Furthermore, when the system is in single-cooler operation mode, an alarm for adding a cooler is triggered and an operation instruction of "cooler water temperature is high, it is recommended to add a cooler" is output.

[0053] Specifically, in some implementations, when the outlet header temperature of the shut-off cooler... Furthermore, when the system is in "single closed-cooler operation" mode, the system will trigger a "add closed-cooler" alarm and output the operation instruction "closed-cooler water temperature is high, it is recommended to add one closed-cooler." This step is an important component of the economic dispatch layer in the intelligent monitoring and economic dispatch system of this invention. Its technical implementation is based on the joint judgment logic of the closed-cooler operating status and the outlet water temperature. It aims to promptly propose equipment addition suggestions when the system cooling capacity is insufficient, so as to ensure the cooling needs of the main unit and auxiliary equipment, while avoiding the risk of equipment overheating due to insufficient cooling capacity.

[0054] Furthermore, this step first involves acquiring the outlet header temperature signal of the cooler in real time through the DCS system and comparing it with a preset temperature threshold. The system compares the valve status with the electric valve position signal to determine whether it is currently in "single closed cooler operation" mode. Specifically, the system status determination logic includes the combination of the status of the outlet electric valve of closed cooler A or B and the open water inlet electric valve. For example: if the outlet electric valve of closed cooler A... And imported electric valves The outlet and inlet electric valves of the closed cooler B If the condition is met, the system is determined to be in the "single closed cooler A in operation" state. This logic is implemented through Boolean operations and state combination judgment to ensure that the system accurately identifies the operating status of the equipment.

[0055] Furthermore, the threshold temperature of the outlet header of the closed cooler is set as follows: This value is determined based on the design specifications and statistical analysis of historical operating data for closed-loop cooling systems in thermal power plants. It aims to reflect that a single closed-loop cooler is approaching its maximum heat exchange capacity under current load and ambient temperature. Furthermore, the determination of the electric valve status signal is based on... Indicates "closed". It indicates "on" and, in conjunction with the current signal, helps determine whether the device is truly in operation, avoiding incorrect decisions due to signal drift or false alarms.

[0056] Furthermore, this procedure applies to thermal power plants operating at low loads or in low ambient temperatures, where only a single closed-circuit cooler needs to operate. When the system detects an abnormal rise in water temperature, indicating that the heat exchange capacity of the current closed-circuit cooler is insufficient to meet the system's cooling requirements, the system will automatically trigger an alarm and provide operational suggestions to assist operators in making quick decisions and avoiding the risk of equipment overheating or shutdown due to insufficient cooling capacity.

[0057] Furthermore, the technical advantage of this step lies in achieving dynamic evaluation of the cooler's operational capacity by combining temperature thresholds with equipment status logic. This allows for the early suggestion of increasing cooling capacity, improving system response speed and operational economy. Simultaneously, this rule, as solidification of expert knowledge, reduces reliance on operator experience and standardizes and automates scheduling strategies.

[0058] S4 uses a multivariate state estimation model to dynamically verify the scheduling suggestions. It judges whether there are hidden anomalies by comparing the residuals of real-time parameters with the expected values ​​of the model, and outputs the optimized scheduling decision.

[0059] Specifically, this invention uses a multivariate state estimation model (MSET) to dynamically verify scheduling recommendations, thereby enabling real-time monitoring and optimization decision-making for the operating status of open cooling water systems in thermal power plants. This step constructs a normal operating condition model based on historical system operating data and identifies potential hidden anomalies through residual analysis between real-time data and the model's expected values, ensuring the accuracy of scheduling recommendations and the economic efficiency of system operation.

[0060] Furthermore, the MSET model is trained based on historical data of the system under various healthy operating conditions, covering operating states under different seasons and load levels. Model inputs include key operating parameters such as chiller pump start-up current, outlet pressure, chiller inlet and outlet temperatures, and valve opening. The output is the expected value of each parameter under the current operating conditions. During system operation, these parameters are collected in real time and input into the MSET model. The model calculates the theoretical expected value of each parameter based on the current operating conditions, and then calculates the residual between the real-time value and the expected value. ,in For real-time measurements, This is the model's predicted value. If the residual exceeds the preset statistical confidence interval (such as the 95% confidence interval), the parameter is considered to have abnormal behavior, which may indicate a decline in equipment performance or an imbalance in system coupling.

[0061] Furthermore, the setting of residual thresholds is based on the statistical characteristics of historical data, typically employing the 3σ principle or quantile methods based on the historical residual distribution. For example, if the standard deviation of the residuals of a certain parameter is... Then the confidence interval can be set as In addition, the training data for the model must meet the requirements of continuity and representativeness. It is recommended to select at least 6 months of stable running data and perform outlier removal and data normalization to improve the model's generalization ability.

[0062] Furthermore, this step is embedded in the power plant's DCS system, working in conjunction with the expert rule base of the economic dispatch layer. When the rule layer proposes dispatch suggestions such as "start the cooling pump" or "cut off the shut-off cooler," the MSET model dynamically verifies the current system state to determine if there are any hidden anomalies not identified by the rules. For example, under low-load conditions, if the MSET model detects an abnormally high cooling pump current and a low outlet pressure, even if the shut-off coolant temperature does not trigger a rule alarm, the system can still issue a warning of "pump efficiency decline," thereby avoiding blind operation.

[0063] Furthermore, this step significantly improves the reliability of scheduling recommendations and the economy of system operation. Through residual analysis, the system can identify the slow degradation trend of equipment performance, enabling predictive maintenance and avoiding sudden failures. Simultaneously, this method elevates scheduling decisions from single-parameter threshold judgments to dynamic verification at the system behavior level, enhancing adaptability to complex operating conditions and providing operators with more scientific and precise operational guidance.

[0064] Furthermore, S4 includes: S41, Calculate the residuals between the real-time parameters and the model's expected values. If the absolute value of any residual exceeds the upper limit of the preset statistical confidence interval, or below the lower limit If so, then a latent anomaly is determined to exist.

[0065] Specifically, in the health diagnosis layer of this invention, the residual between real-time parameters and model expected values ​​is calculated. This step is based on the Multivariate State Estimation (MSET) model, which compares the current system operating parameters. The expected value predicted by the model under the same working conditions This allows for the identification of early, slowly evolving anomalies that may exist in the system.

[0066] Furthermore, the residual calculation process employs a point-by-point differencing method between real-time data acquisition and model predictions, wherein... Indicates the first Real-time measured values ​​of each monitoring parameter This represents the expected value of the parameter predicted by the MSET model under current load, ambient temperature, equipment status, and other conditions. Residual This reflects the deviation between the actual operating state and the "normal behavior pattern" learned by the model. Furthermore, to enhance the robustness of anomaly detection, the system introduces a statistical confidence interval mechanism, which sets an upper limit on the residuals. and lower limit ,when or When this occurs, the system determines that the parameter has a hidden anomaly.

[0067] Furthermore, parameters and The system's design is based on statistical analysis of historical data. During the model training phase, the system models the residual distribution of several key parameters (such as the cooling pump current, outlet pressure, inlet and outlet temperatures of the closed cooler, and valve opening) under normal operating conditions, typically using the 3σ principle or a 95% confidence interval as the threshold standard. For example, if the standard deviation of the residual of a certain parameter is... ,but , This method conforms to the Statistical Process Control (SPC) standard commonly used in industrial process monitoring and has good engineering applicability.

[0068] Furthermore, in practical applications, this step is deployed within the DCS system. It involves real-time acquisition of cooling system operating data, inputting it into the MSET model, obtaining the expected value, and then calculating the residual. The system operating environment requires high sampling frequency (typically 1 second / time) and low latency data processing capabilities to ensure timely warnings. During low-temperature seasons or under low-load conditions, this step can effectively identify hidden faults that are difficult to detect through traditional setpoint alarms, such as decreased pump efficiency, valve internal leakage, and heat exchanger scaling.

[0069] Furthermore, this residual calculation and confidence interval judgment mechanism realizes the transformation from "absolute value alarm" to "behavioral anomaly early warning", enabling the system to detect abnormal changes in equipment performance in advance before parameters exceed limits, thereby providing predictive maintenance suggestions for operators and improving the safety and economy of system operation.

[0070] The economic dispatching method for open cooling water systems in thermal power plants according to embodiments of the present invention can realize economic start-up and shutdown dispatching and early fault warning of open cooling water systems in thermal power plants, reduce plant power consumption rate and improve operational safety and automation level.

[0071] Example 2 This invention proposes a smart monitoring system for the scheduling of cooling pumps and coolers in DCS thermal power plants. The system aims to: establish an expert rule base based on temperature thresholds to provide economic start-up and shutdown recommendations for cooling pumps and coolers, thereby reducing plant power consumption; construct a normal operating condition model based on Multivariate State Estimation (MSET) to provide early warning of latent anomalies in system parameters; establish a hybrid intelligent diagnostic mechanism that verifies rules and models to improve the accuracy and reliability of alarms; and provide clear operational guidance for operators, reducing reliance on personal experience and achieving standardization and optimization of operational procedures.

[0072] Furthermore, the system described in this invention comprises constructing a two-layer intelligent architecture that combines an economic scheduling layer and a health diagnosis layer: In one embodiment of the present invention, the economic scheduling layer includes: the layer implementing energy-saving and consumption-reducing scheduling suggestions through preset precise temperature thresholds and equipment status logic, such as... Figure 2 As shown. The economic scheduling rules for starting the cooling pump include the following rule steps: Rule 1: Start the cooling pump Triggering logic: Condition 1 AND Condition 2 Condition 1: The outlet header temperature of the closed cooler is >37℃ (the system's cooling capacity is insufficient). Condition 2: The cooling pump is in a "shutdown" state (see the judgment logic below). Output result: Triggered "Start Cooling Pump" alarm, guidance: "The cooling water temperature is high when closed, it is recommended to start one cooling pump."

[0073] Rule 2: Disconnect the cold pump Triggering logic: Condition 1 AND Condition 2 Condition 1: The outlet header temperature of the closed cooler is <34℃ (the system has excess cooling capacity). Condition 2: The cooling pump is in "running" mode (see below for the judgment logic). Output result: Triggered "Disconnect cooling pump" alarm, guidance: "The cooling water temperature is low, one cooling pump can be shut down to save power."

[0074] In one embodiment of the present invention, the device status determination includes: Cold pump "Running" status: (Cold water pump current > 50A) AND (Cold water pump outlet electric valve = 1 "Open") Cold pump "Shutdown" status: (Cold water pump current < 50A) AND (Cold water pump outlet electric valve = 0 "Closed") 1.2 Economic Dispatch Rules for Closed-Cold Units: Rule 3: Add a closed-loop cooler Triggering logic: Condition 1 AND Condition 2 Condition 1: The outlet header temperature of the closed-loop cooler is >35℃ (the heat exchange capacity of a single closed-loop cooler has reached its maximum). Condition 2: The system is in "single closed cooler operation" state (see the judgment logic below). Output result: Triggered the "Add a closed-loop cooler" alarm, with the instruction: "The closed-loop cooler water temperature is high, it is recommended to add a closed-loop cooler."

[0075] Rule 4: Disconnect one shut-off cooler Triggering logic: Condition 1 AND Condition 2 Condition 1: The outlet header temperature of the closed-circuit cooler is <31℃ (excess heat exchange capacity). Condition 2: The system is in "dual-closed cooler operation" state (see the judgment logic below). Output result: Triggered the "Disconnect one shut-off cooler" alarm, with the instruction: "The shut-off cooler water temperature is low. Disconnect one shut-off cooler to reduce resistance and save plant power."

[0076] In one embodiment of the present invention, the system state determination includes: Dual-closed-cooler operation: (Closing-cooler A outlet electric valve = 1) AND (Closing-cooler B outlet electric valve = 1) AND (Closing-cooler A open water inlet electric valve = 1) AND (Closing-cooler B open water inlet electric valve = 1) Single-closed-cooler operation (A operation): (Closing-cooler A outlet electric valve = 1) AND (Closing-cooler B outlet electric valve = 0) AND (Closing-cooler A open water inlet electric valve = 1) AND (Closing-cooler B open water inlet electric valve = 0) Single-closed-cooler operation (B operation): (Closing-cooler A outlet electric valve = 0) AND (Closing-cooler B outlet electric valve = 1) AND (Closing-cooler A open water inlet electric valve = 0) AND (Closing-cooler B open water inlet electric valve = 1) In one embodiment of the present invention, the health diagnosis layer includes: this layer is used to detect early, slowly changing anomalies that cannot be captured by rules. Model construction: A large sample of historical data of the cooling system under various healthy and stable operating conditions (covering different seasons and different loads) is collected, and key parameters (such as: cooling pump current, outlet pressure, inlet and outlet temperatures of the cooling coil, valve opening, etc.) are selected to train the MSET model. The model learns the inherent correlation between these parameters under normal conditions. Anomaly warning: When the system is running in real time, the current parameter set is input into the MSET model, and the model will output a set of "expected values" that should be present under the current operating conditions. The real-time value is compared with the expected value. Warning condition: When the residual (real-time value - expected value) of any parameter or combination of parameters exceeds the set statistical confidence interval, even if the absolute values ​​of all parameters are within the normal range, the system issues a "parameter XX behavior abnormal" warning. Example: If the MSET model finds that "under the same load and ambient temperature, the current of the cooling pump is higher than the model's expected value, while the outlet pressure is slightly lower", it can provide an early warning that "the cooling pump may experience reduced efficiency or impeller wear", much earlier than the "high water temperature" alarm.

[0077] The embodiments of this invention also have the following technical effects: After the system of this invention is put into operation, it is expected to generate the following significant benefits: Direct economic benefits: Through precise "pump / cooler switching" suggestions, unnecessary equipment operation is avoided, effectively reducing plant power consumption and achieving annual energy-saving benefits. Improved safety and reliability: Proactive early warning: The MSET model can detect the slow degradation trend of equipment performance, realize predictive maintenance, and avoid sudden failures. Precise operation: Expert rule guidance avoids misjudgment and blind operation by operators, improving the safety of system operation. Improved operation and maintenance level: Reduced workload: Freeing operators from tedious status judgments, allowing them to focus on anomaly handling and decision-making. Knowledge solidification: Solidifying the optimal scheduling strategy into system rules achieves the standardization and inheritance of operational experience. System collaborative optimization: Monitoring the on-state pump and off-state cooler as a whole achieves collaborative optimization between subsystems, improving the overall plant economy.

[0078] Example 3 To achieve the above embodiments, such as Figure 3 As shown, this embodiment also provides a smart monitoring device 10 for DCS thermal power plant cooling pump and cooler shut-off scheduling, including: The real-time data acquisition module 100 is used to acquire real-time operating data such as the current of the cooling pump, the status of the outlet electric valve, the temperature of the outlet main pipe of the cooler, and the valve opening. The operation status determination module 200 is used to determine the operation status of the cooling pump and the cooling unit based on the real-time operation data by combining electrical parameters and mechanical position parameters. The scheduling suggestion generation module 300 is used to generate scheduling suggestions for starting and stopping the cooling pump or turning off the cooler based on the preset temperature threshold range and system operating status. The dynamic verification module 400 is used to dynamically verify the scheduling suggestions using a multivariate state estimation model. It judges whether there are hidden anomalies by comparing the residuals of real-time parameters with the expected values ​​of the model, and outputs the optimized scheduling decision.

[0079] Furthermore, the operation status determination module 200 is also used for: Based on the starting current of the cold water pump With outlet electric valve The combinational logic determines that the cooling pump is in operation. Based on the starting current of the cold water pump With outlet electric valve The combinational logic determines that the cooling pump is in a stopped state.

[0080] Furthermore, the scheduling suggestion generation module 300 is also used for: When the outlet header temperature of the cooler Furthermore, when the system is in single-cooler operation mode, an alarm for adding a cooler is triggered and a high cooler water temperature is output, suggesting the addition of another cooler.

[0081] Furthermore, the dynamic verification module 400 is also used for: Calculate the residual between real-time parameters and model expected values. If the absolute value of any residual exceeds the upper limit of the preset statistical confidence interval, or below the lower limit If so, then a latent anomaly is determined to exist.

[0082] This invention discloses a smart monitoring device for the scheduling of open cooling pumps and closed cooling units in a DCS thermal power plant. This device enables economic scheduling and early fault warning of the open cooling water system in the thermal power plant, effectively reducing the plant's power consumption rate and improving operational safety and equipment reliability.

[0083] Example 4 The present invention also provides an electronic device such as Figure 4 As shown, it includes a processor and a memory. The memory stores executable instructions. When the processor executes the instructions, it implements the above-mentioned intelligent monitoring method for scheduling the start-up of cooling pumps and the shut-down of cooling units in a DCS thermal power plant.

[0084] Example 5 The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned intelligent monitoring method for scheduling the start-up of cooling pumps and the shut-down of coolers in a DCS thermal power plant.

[0085] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0086] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for intelligent monitoring and dispatching of open and closed coolers of a DCS thermal power plant, characterized in that, The method comprises the following steps: S1, acquiring real-time operation data of the open cold pump current, outlet electric valve state, closed cold trap outlet mother pipe temperature and valve opening degree; S2, determining the operation state of the open cold pump and the closed cold trap based on the real-time operation data by combining electrical parameters and mechanical position parameters; S3, generating a dispatching suggestion of starting or stopping the open cold pump or putting in or taking out the closed cold trap according to a preset temperature threshold range and system operation state; S4, dynamically verifying the dispatching suggestion by using a multi-element state estimation model, judging whether there is an implicit abnormality by comparing a residual error of real-time parameters and model expected values, and outputting an optimized dispatching decision.

2. The method of claim 1, wherein, The S2 comprises: S21, open cold water pump current in combination with the outlet motorized valve determines that the cold pump is in an operational state; S22, open the cold water pump current combined with the outlet electric valve The combination logic determines that the cold pump is in a shutdown state.

3. The method of claim 1, wherein, The S3 comprises: S31, when the closed cooler outlet mother pipe temperature And the system is in single closed cooler operation state, trigger the increase of closed cooler alarm and output the closed water temperature is high, the operation guidance of the suggestion of increasing a closed cooler.

4. The method of claim 1, wherein, The S4 comprises: S41, calculating the residual of the real-time parameter and the model expected value If the absolute value of any residual exceeds the upper limit of the preset statistical confidence interval or is lower than the lower limit , it is determined that there is a hidden anomaly.

5. A smart monitoring device for DCS thermal power plant open cooling pump and closed cooler scheduling, characterized in that, The method comprises the following steps: The real-time data acquisition module is used to acquire real-time operation data of the open cold pump current, outlet electric valve state, closed cold trap outlet mother pipe temperature and valve opening degree; The operation state determination module is used to determine the operation state of the open cold pump and the closed cold trap based on the real-time operation data by combining electrical parameters and mechanical position parameters; The dispatching suggestion generation module is used to generate a dispatching suggestion of starting or stopping the open cold pump or putting in or taking out the closed cold trap according to a preset temperature threshold range and system operation state; The dynamic verification module is used to dynamically verify the dispatching suggestion by using a multi-element state estimation model, judge whether there is an implicit abnormality by comparing a residual error of real-time parameters and model expected values, and output an optimized dispatching decision.

6. The apparatus of claim 5, wherein, The operation state determination module is further used to: The open cold water pump current is determined in accordance with the combination logic with the outlet motorized valve determines that the open cold pump is in an operating state; The open chiller pump current is determined in accordance with the combination logic with the outlet motorized valve determines that the open chiller pump is in a shutdown state.

7. The apparatus of claim 5, wherein, The dispatching suggestion generation module is further used to: When the temperature of the outlet main pipe of the closed cooler When the system is in single closed cooler operation state, the alarm of increasing the closed cooler is triggered, the high temperature of the closed cooler water is output, and the operation guidance of increasing one closed cooler is suggested.

8. The apparatus of claim 5, wherein, The dynamic verification module is further used to: calculating residuals of real-time parameters from model expectations if any residual exceeds a preset upper limit of statistical confidence interval or is lower than a lower limit then determining that there is a hidden anomaly.

9. A computer device, comprising: The method comprises the following steps: The method comprises the following steps:

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to realize the method.

Images