A high-pressure bypass intelligent temperature and pressure reduction dynamic regulation method and system

Abstract

The application relates to the technical field, in particular to a high-pressure bypass intelligent temperature and pressure reduction dynamic regulation method and system. The method comprises the following steps: constructing a perception array according to the structural parameters of a steam turbine unit, wherein the perception array comprises multiple monitoring points; obtaining operation monitoring data collected by the perception array, setting an operation control strategy according to the operation monitoring data and a preset regulation model; judging whether an update instruction of the regulation model is generated according to a preset update period; through a double mechanism of feedforward compensation and feedback adjustment, the hysteresis effect of a temperature and pressure reduction system of the high-pressure bypass is accurately hedged, the steam temperature fluctuation range at the high-pressure bypass outlet is reduced, through the self-adaptive feedback control of the working condition, the adaptability of the system to complex working conditions such as load variation and steam parameter change is improved, stable temperature control is realized, the risk of pipeline thermal stress fatigue is reduced, and the service life of the steam turbine unit is prolonged.

Description

Technical Field

[0001] This application relates to the technical field, and in particular to a method and system for intelligent de-cooling and de-pressure dynamic control of high-voltage bypass. Background Technology

[0002] The high-pressure bypass system is a crucial component of thermal power generating units. Its function is to de-cool and depressurize the high-temperature, high-pressure main steam before introducing it into the reheater or condenser during unit start-up, shutdown, significant load fluctuations, or fault conditions. This maintains stable boiler operation and ensures turbine safety. The rapid and precise control of the bypass system has a critical impact on the unit's operational flexibility, safety, and economy.

[0003] Currently, the desuperheating and pressure reduction control of high-pressure bypasses mostly adopts a proportional-integral-derivative (PID) feedback control strategy based on fixed parameters, relying on a few measuring points such as outlet steam temperature and pressure for closed-loop regulation. However, in actual operation, due to multiple factors such as physical delays in steam transmission, inertia in valve action, and lag in the response of temperature and pressure measuring elements, traditional feedback control often exhibits significant regulation lag and overshoot phenomena. Especially under drastic changes in operating conditions, this can easily lead to large temperature and pressure fluctuations, prolonged recovery times, affect the stable operation of the unit, and even endanger equipment safety. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for intelligent de-cooling and pressure reduction dynamic control of high-pressure bypass in order to solve the above-mentioned technical problems, thereby ensuring the de-cooling and pressure reduction efficiency of high-pressure bypass and improving the operating efficiency of steam turbine units.

[0005] In some embodiments of this application, by constructing a dual mechanism of feedforward compensation and feedback regulation, the hysteresis effect of the high-pressure bypass cooling and depressurization system is precisely offset, the temperature fluctuation range of the high-pressure bypass outlet steam is reduced, and the system's adaptability to complex operating conditions such as load changes and steam parameter changes is improved through operating condition adaptive feedback control, thereby achieving stable temperature control, reducing the risk of pipeline thermal stress fatigue, and improving the service life of the turbine unit.

[0006] In some embodiments of this application, by real-time evaluation of operating conditions, an online learning window is automatically selected to promptly compensate for the performance degradation of the control system caused by the slow drift of equipment characteristics (such as valve wear and heat exchanger scaling), thereby reducing manual debugging and maintenance.

[0007] In some embodiments of this application, a method for intelligent cooling and pressure reduction dynamic control of a high-voltage bypass is provided, including:

[0008] A sensing array is constructed based on the structural parameters of the steam turbine unit, and the sensing array includes multiple monitoring points; Obtain the operation monitoring data collected by the sensing array, and set the operation control strategy according to the operation monitoring data and the preset regulation model; Judge whether to generate an update instruction for the regulation model according to the preset update period; Among them, the regulation model includes: a feedforward compensation model and a feedback correction model.

[0009] In some embodiments of the present application, the preset regulation model includes: Construct multiple lag scenarios according to the operation record data of the steam turbine unit; Establish a lag scenario sequence A, A=(a1, a2…ai…an), where ai is the i-th lag scenario; n is the number of lag scenarios; Set the feedforward sub-strategies of each lag scenario in turn, and construct a compensation sub-model according to all the feedforward sub-strategies Set ai as the target scenario in turn according to the lag scenario sequence A; Construct a mapping parameter set of the target scenario based on the sensing array and the operation record data; Set the feedforward sub-strategy of the target scenario, and the feedforward sub-strategy includes: feedforward duration and instruction change amplitude; Generate the mapping parameter sets of each lag scenario in turn; Construct a prediction sub-model according to all the mapping parameter sets; Generate a feedforward compensation model, and the feedforward compensation model includes: a prediction sub-model and a compensation sub-model.

[0010] In some embodiments of the present application, the construction of the mapping parameter set of the target scenario includes: Establish a monitoring point sequence C according to the sensing array, C=(c1, c2…ci…cm), where ci is the i-th monitoring point; m is the number of monitoring points; Generate the lag correlation values of each monitoring point; Preset a correlation value threshold F1; If F1<fi (i=1, 2…m), set the i-th monitoring point as a lag mapping point; Among them, fi is the lag correlation value of the i-th monitoring point; Set the expected mapping parameters of each lag mapping point in the target scenario in turn according to the operation record data; Construct a mapping parameter set of the target scenario according to all the expected mapping parameters.

[0011] In some embodiments of the present application, the preset regulation model further includes: Set a basic feedback strategy; Construct multiple disturbance conditions according to the mapping parameter sets of all the lag scenarios; Establish a disturbance condition sequence B, B = (b1, b2... bi... bn), where bi is the disturbance condition corresponding to the i-th lag scenario; n is the number of disturbance conditions; Set the correction sub-strategies for each disturbance condition in sequence; Construct a feedback correction model based on the basic feedback strategy and all correction sub-strategies; Establish a regulation model, the regulation model includes: a feed-forward compensation model and a feedback correction model.

[0012] In some embodiments of the present application, the setting of the operation control strategy includes: Preset multiple regulation cycles; Set the start time node of each regulation cycle as the monitoring time node; Obtain the operation monitoring data collected by the sensing array at the current monitoring time node; Generate a first-level mapping set based on the preprocessing result of the operation monitoring data; The prediction sub-model generates the matching value of the first-level mapping set and each mapping parameter set; Set the lag scenario corresponding to the maximum value in the matching value as the anchor scenario; Set the execution feed-forward strategy for the current regulation cycle according to the anchor scenario; Generate an execution feedback strategy according to the correction sub-strategy of the disturbance condition corresponding to the anchor scenario and the basic feedback strategy; Set the operation control strategy, the operation control strategy includes: an execution feed-forward strategy and an execution feedback strategy.

[0013] In some embodiments of the present application, judging whether to generate a correction instruction for the regulation model includes: Establish multiple update cycles; Set multiple time intervals within the current update cycle; Generate the working condition fluctuation value of each time interval in sequence based on the time series; Preset a working condition fluctuation value threshold K1; Obtain the working condition fluctuation value k of the current time interval; If k > K1, generate a first-level update instruction in the current time interval; If k < K1, obtain the working condition fluctuation value k1 of the next time interval, and judge whether to generate a first-level update instruction according to the working condition fluctuation value k1.

[0014] In some embodiments of the present application, the first-level update instruction includes: Obtain the execution monitoring data; Set ai as the to-be-evaluated scenario in sequence according to the lag scenario sequence A; Generate the regulation efficiency value d of the to-be-evaluated scenario according to the execution monitoring data; Preset regulation efficiency value threshold D1; If d < D1, set the to-be-evaluated scenario as a to-be-corrected scenario; Set the update strategy of the regulation model according to all to-be-corrected scenarios.

[0015] In some embodiments of the present application, a high-pressure bypass intelligent desuperheating and pressure-reducing dynamic regulation system is provided, including: A central control unit, configured to construct a perception array according to the structural parameters of the steam turbine unit, and the perception array includes a plurality of monitoring points; A monitoring unit, including a plurality of monitoring sub-modules, and the monitoring sub-modules are arranged at each monitoring point; The central control unit includes: A first control module, configured to establish a regulation model, and the regulation model includes: a feed-forward compensation model and a feedback correction model; A second control module, configured to set an operation control strategy according to the operation monitoring data acquired by the perception array and according to the operation monitoring data and a preset regulation model; A third control module, configured to determine whether to generate an update instruction for the regulation model according to a preset update period.

[0016] In some embodiments of the present application, the first control module is further configured to: Construct a variety of lag scenarios according to the operation record data of the steam turbine unit; Establish a lag scenario sequence A, A = (a1, a2... ai... an), where ai is the i-th lag scenario; n is the number of lag scenarios; Sequentially set the feed-forward sub-strategies of each lag scenario, and construct a compensation sub-model according to all feed-forward sub-strategies Sequentially set ai as the target scenario according to the lag scenario sequence A; Construct a mapping parameter set of the target scenario based on the perception array and the operation record data; Set the feed-forward sub-strategy of the target scenario, and the feed-forward sub-strategy includes: feed-forward duration and instruction change amplitude; Sequentially generate the mapping parameter sets of each lag scenario; Construct a prediction sub-model according to all mapping parameter sets; Generate a feed-forward compensation model, and the feed-forward compensation model includes: a prediction sub-model and a compensation sub-model; Set a basic feedback strategy; Construct a variety of disturbance conditions according to the mapping parameter sets of all lag scenarios; Establish a disturbance condition sequence B, B = (b1, b2... bi... bn), where bi is the disturbance condition corresponding to the i-th lag scenario; n is the number of disturbance conditions; Sequentially set the correction sub-strategies of each disturbance condition; Construct a feedback correction model based on the basic feedback strategy and all correction sub-strategies; A control model is established, which includes a feedforward compensation model and a feedback correction model.

[0017] In some embodiments of this application, the second control module is further configured to: Multiple control cycles can be preset; The start time of each control cycle is set as the monitoring time point; Acquire the operational monitoring data collected by the sensing array at the current monitoring time point; A primary mapping set is generated based on the preprocessing results of the operational monitoring data; The prediction sub-model generates the fit values ​​of the first-level mapping set and each mapping parameter set; The lag scenario corresponding to the maximum value in the fit value is defined as the anchoring scenario; The execution feedforward strategy for the current control cycle is set according to the anchoring scenario; An execution feedback strategy is generated based on the correction sub-strategy and the basic feedback strategy corresponding to the disturbance condition in the anchoring scenario. Set an operation control strategy, which includes: an execution feedforward strategy and an execution feedback strategy.

[0018] Compared with the prior art, the beneficial effects of the intelligent de-cooling and de-pressure dynamic control method and system for high-voltage bypass provided in this application are as follows: By constructing a dual mechanism of feedforward compensation and feedback regulation, the hysteresis effect of the high-pressure bypass cooling and depressurization system is precisely offset, reducing the temperature fluctuation range of the high-pressure bypass outlet steam. Through operating condition adaptive feedback control, the system's adaptability to complex operating conditions such as load changes and steam parameter variations is improved, achieving stable temperature control, reducing the risk of pipeline thermal stress fatigue, and extending the service life of the turbine unit.

[0019] By assessing the operating status in real time and automatically selecting the online learning window, the system can promptly compensate for the performance degradation of the control system caused by the slow drift of equipment characteristics (such as valve wear and heat exchanger scaling), thereby reducing manual debugging and maintenance. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a preferred embodiment of a high-voltage bypass intelligent cooling and pressure reduction dynamic control method. Detailed Implementation

[0021] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0022] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0023] 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. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0024] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0025] like Figure 1 As shown in the preferred embodiment of this application, a high-voltage bypass intelligent de-cooling and pressure reduction dynamic control method includes: S101: Construct a sensing array based on the structural parameters of the steam turbine unit. The sensing array includes multiple monitoring points. S102: Acquire the operation monitoring data collected by the sensing array, and set the operation control strategy based on the operation monitoring data and the preset control model; S103: Determine whether to generate an update instruction for the control model based on the preset update cycle; The control models include: feedforward compensation model and feedback correction model.

[0026] Specifically, multiple key data points are selected based on the structural parameters of the steam turbine unit, and real-time data of each key data point is collected synchronously through a set data acquisition device. The real-time data includes, but is not limited to: the main steam pressure, temperature and flow rate at the boiler outlet, the steam flow rate change rate, the steam pressure after the high-pressure bypass pressure reducing valve, the instantaneous temperature and volumetric flow rate of the steam downstream of the desuperheater, pipeline pressure and other parameters related to the desuperheating and pressure reduction control of the high-pressure bypass.

[0027] Specifically, a monitoring array is constructed based on key data points, where each monitoring point in the array represents a key data point. Based on the data categories required for each key data point, corresponding data acquisition devices are configured to collect the data.

[0028] Specifically, the preset control model includes: Multiple lag scenarios were constructed based on the operating record data of the steam turbine unit; Establish a sequence A of delayed scenarios, A=(a1, a2, ..., ai, ..., an), where ai is the i-th delayed scenario and n is the number of delayed scenarios; Sequentially define the feedforward sub-strategies for each lag scenario, and construct a compensation sub-model based on all feedforward sub-strategies. Based on the sequence of lagging scenarios A, ai is sequentially set as the target scenario; A mapping parameter set for the target scene is constructed based on the sensing array and operation record data; Define the feedforward sub-strategy for the target scenario. The feedforward sub-strategy includes: feedforward duration and instruction change magnitude. Generate mapping parameter sets for each delayed scenario in sequence; Construct a predictive sub-model based on the complete set of mapping parameters; Generate a feedforward compensation model, which includes a prediction sub-model and a compensation sub-model.

[0029] Specifically, the operation record data refers to the historical operation data of the steam turbine unit (including: monitoring data collected by the monitoring array, operating condition data of the steam turbine unit, desuperheating control parameters, etc.).

[0030] Specifically, lag indicators are defined as follows: lag duration (the flow and transmission time of steam from the mixing point of the desuperheater to the downstream temperature sensor) and thermal inertia time constant (the response speed of the water spray cooling process. The larger the thermal inertia time constant, the longer the time interval between the downstream temperature and the expected temperature change). Based on the historical value range of the turbine unit's operating record data, the historical value range of each lag indicator is segmented to generate multiple value intervals for each lag indicator. Random combinations of the value intervals of two lag indicators construct multiple lag scenarios. Each lag scenario contains a value interval for lag duration and a value interval for thermal inertia time constant, and the value intervals of the lag indicators corresponding to any two lag scenarios are not exactly the same.

[0031] Specifically, the set of influencing parameters for each lag index is generated. The parameters influencing the lag time include, but are not limited to, multiple parameters such as steam flow rate, steam change rate, and pipeline length. The set of influencing parameters for the thermal inertia time constant includes, but is not limited to, parameters such as real-time steam conditions (flow rate, density), pipeline internal pressure, temperature, and flow rate.

[0032] Specifically, a corresponding feedforward duration is set according to the average lag duration corresponding to the target scenario (i.e., the lead time for the execution of the control instruction, and the feedforward duration is the same as the average lag duration). The corresponding instruction change amplitude is set according to the thermal inertia time coefficient corresponding to the target scenario. The smaller the thermal inertia time coefficient, the greater the corresponding instruction change amplitude. The mapping relationship between the two can be set according to historical parameters. By setting the feedforward duration and the instruction change amplitude, rapid, stable, and overshoot-free temperature control of the high-pressure bypass in the target scenario is achieved.

[0033] Specifically, a mapping parameter set for the target scenario is constructed, including: Based on the sensing array, a monitoring point sequence C is established, C = (c1, c2... ci... cm), where ci is the i-th monitoring point; m is the number of monitoring points; The lag correlation value of each monitoring point is generated; A correlation value threshold F1 is preset; If F1 < fi (i = 1, 2... m), the i-th monitoring point is set as a lag mapping point; where fi is the lag correlation value of the i-th monitoring point; According to the operation record data, the expected mapping parameters of each lag mapping point in the target scenario are set in sequence; Based on all the expected mapping parameters, a mapping parameter set for the target scenario is constructed.

[0034] Specifically, by screening the operation record data, associated data corresponding to the target scenario is generated (i.e., relevant monitoring data of the steam turbine unit operating in the target scenario). According to the associated data, the data fluctuation value of each monitoring point is generated (i.e., the volatility of the continuous monitoring data of the current monitoring point collected in the target scenario). The greater the data fluctuation value, the smaller the corresponding lag correlation value. The mapping relationship between the two can be set according to historical parameters.

[0035] Specifically, the lag correlation value threshold can be set according to historical parameters. If the lag correlation value of the current monitoring point is greater than the preset lag correlation value threshold, it indicates that the current monitoring point can map the lag state corresponding to the target scenario, and the mapping parameters of each lag mapping point in the target scenario are set (i.e., the data parameters corresponding to the lag monitoring point when in the target scenario), and a mapping parameter set for the target scenario is constructed.

[0036] Specifically, the preset regulation model further includes: Set a basic feedback strategy; Based on the mapping parameter sets of all lag scenarios, multiple disturbance conditions are constructed; A disturbance condition sequence B is established, B = (b1, b2... bi... bn), where bi is the disturbance condition corresponding to the i-th lag scenario; n is the number of disturbance conditions; Set the correction sub-strategies for each disturbance condition in sequence; Construct a feedback correction model based on the basic feedback strategy and all correction sub-strategies; Establish a control model, which includes a feedforward compensation model and a feedback correction model.

[0037] Specifically, the basic feedback strategy includes conventional PID parameters (i.e., proportional, integral, and derivative). The feedback control quantity for valve opening is generated based on the difference between the actual temperature downstream of the high-pressure bypass and the target temperature, and the final valve opening is set by combining the basic feedback strategy with the corresponding feedforward compensation strategy.

[0038] Specifically, by quantifying the set of influence parameters corresponding to each lag scenario, corresponding operating conditions are generated, and multiple disturbance operating conditions are generated based on all operating conditions.

[0039] Specifically, based on the historical control parameters corresponding to each disturbance condition, adjustment parameters (i.e., correction sub-strategies) are generated for the conventional PID parameters. For example, when the thermal inertia time coefficient is small, the integral and derivative times are shortened to make the integral and derivative actions faster, thereby improving control reliability and safety.

[0040] It is understandable that in the above embodiments, by constructing a dual mechanism of feedforward compensation and feedback regulation, the hysteresis effect of the high-pressure bypass cooling and depressurization system is precisely offset, the temperature fluctuation range of the high-pressure bypass outlet steam is reduced, and the system's adaptability to complex operating conditions such as load changes and steam parameter changes is improved through operating condition adaptive feedback control, thereby achieving stable temperature control, reducing the risk of pipeline thermal stress fatigue, and improving the service life of the turbine unit.

[0041] In a preferred embodiment of this application, an operation control strategy is set, including: Multiple control cycles can be preset; The start time of each control cycle is set as the monitoring time point; Acquire the operational monitoring data collected by the sensing array at the current monitoring time point; A primary mapping set is generated based on the preprocessing results of the operational monitoring data; The prediction sub-model generates the fit values ​​of the first-level mapping set and each mapping parameter set; The lag scenario corresponding to the maximum value in the fit value is defined as the anchoring scenario; The execution feedforward strategy for the current control cycle is set according to the anchoring scenario; An execution feedback strategy is generated based on the correction sub-strategy and the basic feedback strategy corresponding to the disturbance condition in the anchoring scenario. Set up the operation control strategy, which includes: execution feedforward strategy and execution feedback strategy.

[0042] Specifically, the duration of the regulation cycle can be set according to the operating condition fluctuation frequency of the steam turbine unit. The higher the operating condition fluctuation frequency, the shorter the corresponding duration of the regulation cycle. The mapping relationship between the two can be set according to historical parameters.

[0043] Specifically, the operation monitoring data are the real-time data parameters collected by all the monitoring points in the sensing array. By processing all the operation monitoring data, a first-level mapping set (i.e., the data status of each monitoring point) is generated.

[0044] Specifically, by comparing and analyzing the first-level mapping set and the mapping parameter set corresponding to the current lag scenario, a corresponding similarity value is set according to the difference degree. The smaller the difference degree, the larger the corresponding similarity value. The mapping relationship between the two can be set according to historical parameters. The larger the similarity value, the greater the possibility that the current high-pressure bypass is in the current lag scenario.

[0045] Specifically, the feedforward sub-strategy corresponding to the anchored scenario is set as the execution of the feedforward strategy. A feedback control quantity of the valve opening is generated by executing the feedback strategy, and the final valve opening is set in combination with the corresponding feedforward compensation strategy.

[0046] In the preferred embodiment of the present application, determining whether to generate a correction instruction for the regulation model includes: Establishing multiple update cycles; Setting multiple time intervals within the current update cycle; Generating the operating condition fluctuation values of each time interval in sequence based on the time series; Presetting an operating condition fluctuation value threshold K1; Obtaining the operating condition fluctuation value k of the current time interval; If k > K1, generating a first-level update instruction in the current time interval; If k < K1, obtaining the operating condition fluctuation value k1 of the next time interval, and determining whether to generate a first-level update instruction according to the operating condition fluctuation value k1.

[0047] Specifically, the update cycle can be set according to the operating condition fluctuation frequency. The greater the operating condition fluctuation frequency, the shorter the duration of the update cycle.

[0048] Specifically, the operating condition fluctuation value threshold can be set according to historical parameters. When the real-time operating condition fluctuation value is lower than the preset operating condition fluctuation value threshold, it indicates that the current steam turbine unit is in a stable operating state and belongs to a suitable learning window period.

[0049] Specifically, the first-level update instruction includes: Obtaining the execution monitoring data; Successively setting ai as the to-be-evaluated scenario according to the lag scenario sequence A; Generate a regulation efficiency value d for the to-be-evaluated scenario based on the execution monitoring data; Preset a regulation efficiency value threshold D1; If d < D1, set the to-be-evaluated scenario as a to-be-corrected scenario; Set an update strategy for the regulation model according to all the to-be-corrected scenarios.

[0050] Specifically, the regulation efficiency value threshold can be set according to historical parameters. If the regulation efficiency value of the current lag scenario is less than the preset regulation efficiency value threshold, it indicates that the control efficiency of the current operation control strategy for the desuperheating and pressure reduction of the high-pressure bypass has not reached the expectation, and the feedforward sub-strategy of the current lag scenario needs to be corrected in a timely manner.

[0051] Specifically, the regulation efficiency value can be set according to the average difference between the actual temperature and the expected temperature downstream of the high-pressure bypass in the current lag scenario. The smaller the average difference, the higher the corresponding regulation efficiency value.

[0052] It can be understood that in the above embodiments, by real-time evaluating the working condition state, automatically selecting the online learning window, and automatically compensating for the problem of the degradation of the regulation system performance caused by the slow drift of the equipment characteristics (such as valve wear and heat exchanger fouling) in a timely manner, manual debugging and maintenance are reduced.

[0053] Based on another preferred embodiment of a high-pressure bypass intelligent desuperheating and pressure reduction dynamic regulation method in any of the above preferred embodiments, in this preferred embodiment, a high-pressure bypass intelligent desuperheating and pressure reduction dynamic regulation system is provided, including: A central control unit, configured to construct a perception array according to the structural parameters of the steam turbine unit, and the perception array includes multiple monitoring points; A monitoring unit, including multiple monitoring sub-modules, and the monitoring sub-modules are arranged at each monitoring point; The central control unit includes: A first control module, configured to establish a regulation model, and the regulation model includes: a feedforward compensation model and a feedback correction model; A second control module, configured to set an operation control strategy according to the operation monitoring data acquired by the obtained perception array and according to the operation monitoring data and the preset regulation model; A third control module, configured to judge whether to generate an update instruction for the regulation model according to the preset update period.

[0054] Specifically, the monitoring sub-module is preferably various types of data acquisition devices for collecting Set corresponding data acquisition devices according to the data types required to be collected at each monitoring point.

[0055] In the preferred embodiment of the present application, the first control module is further configured to: Multiple lag scenarios were constructed based on the operating record data of the steam turbine unit; Establish a sequence A of delayed scenarios, A=(a1, a2, ..., ai, ..., an), where ai is the i-th delayed scenario and n is the number of delayed scenarios; Sequentially define the feedforward sub-strategies for each lag scenario, and construct a compensation sub-model based on all feedforward sub-strategies. Based on the sequence of lagging scenarios A, ai is sequentially set as the target scenario; A mapping parameter set for the target scene is constructed based on the sensing array and operation record data; Define the feedforward sub-strategy for the target scenario. The feedforward sub-strategy includes: feedforward duration and instruction change magnitude. Generate mapping parameter sets for each delayed scenario in sequence; Construct a predictive sub-model based on the complete set of mapping parameters; Generate a feedforward compensation model, which includes a prediction sub-model and a compensation sub-model. Set up a basic feedback strategy; Multiple disturbance conditions are constructed based on the mapping parameter set of all lagging scenarios; Establish a disturbance condition sequence B, B=(b1, b2, ..., bi, ..., bn), where bi is the disturbance condition corresponding to the i-th lag scenario; n is the number of disturbance conditions. Set the correction sub-strategies for each disturbance condition in sequence; Construct a feedback correction model based on the basic feedback strategy and all correction sub-strategies; Establish a control model, which includes a feedforward compensation model and a feedback correction model.

[0056] In a preferred embodiment of this application, the second control module is further configured to: Multiple control cycles can be preset; The start time of each control cycle is set as the monitoring time point; Acquire the operational monitoring data collected by the sensing array at the current monitoring time point; A primary mapping set is generated based on the preprocessing results of the operational monitoring data; The prediction sub-model generates the fit values ​​of the first-level mapping set and each mapping parameter set; The lag scenario corresponding to the maximum value in the fit value is defined as the anchoring scenario; The execution feedforward strategy for the current control cycle is set according to the anchoring scenario; An execution feedback strategy is generated based on the correction sub-strategy and the basic feedback strategy corresponding to the disturbance condition in the anchoring scenario. Set up the operation control strategy, which includes: execution feedforward strategy and execution feedback strategy.

[0057] According to the first concept of the application, by constructing a dual mechanism of feedforward compensation and feedback regulation, the hysteresis effect of the high-pressure bypass cooling and depressurization system is precisely offset, the fluctuation range of steam temperature at the high-pressure bypass outlet is reduced, and the system's adaptability to complex operating conditions such as load changes and steam parameter changes is improved through condition-adaptive feedback control, thereby achieving stable temperature control, reducing the risk of pipeline thermal stress fatigue, and improving the service life of the turbine unit.

[0058] According to the second concept of this application, by real-time evaluation of the operating status, an online learning window is automatically selected to promptly compensate for the problem of slow drift of equipment characteristics (such as valve wear and heat exchanger scaling) that causes a decline in the performance of the control system, thereby reducing manual debugging and maintenance.

[0059] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A method for intelligent de-cooling and pressure reduction dynamic control of a high-voltage bypass, characterized in that, Including: Construct a perception array according to the structural parameters of the steam turbine unit, and the perception array includes multiple monitoring points; Obtain the operation monitoring data collected by the perception array, and set the operation control strategy according to the operation monitoring data and the preset regulation model; Judge whether to generate an update instruction for the regulation model according to the preset update period; Among them, the regulation model includes: a feedforward compensation model and a feedback correction model.

2. The high-voltage bypass intelligent cooling and pressure reduction dynamic control method as described in claim 1, characterized in that, The preset regulation model includes: Construct multiple lag scenarios according to the operation record data of the steam turbine unit; Establish a lag scenario sequence A, A=(a1, a2…ai…an), where ai is the i-th lag scenario; n is the number of lag scenarios; Set the feedforward sub-strategies of each lag scenario in turn, and construct a compensation sub-model according to all the feedforward sub-strategies Set ai as the target scenario in turn according to the lag scenario sequence A; Construct a mapping parameter set of the target scenario based on the perception array and the operation record data; Set the feedforward sub-strategy of the target scenario, and the feedforward sub-strategy includes: feedforward duration and instruction change amplitude; Generate the mapping parameter sets of each lag scenario in turn; Construct a prediction sub-model according to all the mapping parameter sets; Generate a feedforward compensation model, and the feedforward compensation model includes: a prediction sub-model and a compensation sub-model.

3. The high-voltage bypass intelligent cooling and pressure reduction dynamic control method as described in claim 2, characterized in that, The construction of the mapping parameter set of the target scenario includes: Establish a monitoring point sequence C according to the perception array, C=(c1, c2…ci…cm), where ci is the i-th monitoring point; m is the number of monitoring points; Generate the lag correlation value of each monitoring point; Preset the correlation value threshold F1; If F1<fi (i=1,2…m), set the i-th monitoring point as the lag mapping point; Among them, fi is the lag correlation value of the i-th monitoring point; Set the expected mapping parameters of each lag mapping point in the target scenario in turn according to the operation record data; Construct the mapping parameter set of the target scenario according to all the expected mapping parameters.

4. The high-voltage bypass intelligent cooling and pressure reduction dynamic control method as described in claim 3, characterized in that, The preset regulation model also includes: Set the basic feedback strategy; Construct multiple disturbance conditions according to the mapping parameter sets of all the lag scenarios; Establish a disturbance condition sequence B, B=(b1, b2…bi…bn), where bi is the disturbance condition corresponding to the i-th lag scenario; n is the number of disturbance conditions; Set the correction sub-strategies of each disturbance condition in turn; Construct a feedback correction model according to the basic feedback strategy and all the correction sub-strategies; Establish a regulation model, and the regulation model includes: a feedforward compensation model and a feedback correction model.

5. The high-voltage bypass intelligent cooling and pressure reduction dynamic control method as described in claim 4, characterized in that, The setting of the operation control strategy includes: Preset multiple regulation cycles; Set the start time node of each regulation cycle as the monitoring time node; Obtain the operation monitoring data collected by the perception array at the current monitoring time node; Generate a first-level mapping set according to the preprocessing result of the operation monitoring data; The prediction sub-model generates the fitting value of the first-level mapping set and each mapping parameter set; Set the lag scenario corresponding to the maximum value in the fitting value as the anchor scenario; Set the execution feedforward strategy of the current regulation cycle according to the anchor scenario; Generate an execution feedback strategy according to the correction sub-strategy and the basic feedback strategy of the disturbance condition corresponding to the anchor scenario; Set the operation control strategy, and the operation control strategy includes: an execution feedforward strategy and an execution feedback strategy.

6. The high-voltage bypass intelligent cooling and pressure reduction dynamic control method as described in claim 5, characterized in that, Determine whether to generate a correction instruction for the regulation model, including: Establish multiple update cycles; Set multiple time intervals within the current update cycle; Generate the operating condition fluctuation values for each time interval in sequence based on the time series; Preset the operating condition fluctuation value threshold K1; Obtain the operating condition fluctuation value k of the current time interval; If k > K1, generate a first-level update instruction in the current time interval; If k < K1, obtain the operating condition fluctuation value k1 of the next time interval, and determine whether to generate a first-level update instruction according to the operating condition fluctuation value k1.

7. The high-voltage bypass intelligent de-cooling and pressure reduction dynamic control method as described in claim 6, characterized in that, The first-level update instruction includes: Obtain the execution monitoring data; Set ai as the to-be-evaluated scenario in sequence according to the lag scenario sequence A; Generate the regulation efficiency value d of the to-be-evaluated scenario according to the execution monitoring data; Preset the regulation efficiency value threshold D1; If d < D1, set the to-be-evaluated scenario as the to-be-corrected scenario; Set the update strategy of the regulation model according to all the to-be-corrected scenarios.

8. A high-voltage bypass intelligent de-cooling and de-pressure dynamic control system, employing the high-voltage bypass intelligent de-cooling and de-pressure dynamic control method as described in any one of claims 1-7, characterized in that, Including: The central control unit is used to construct a perception array according to the structural parameters of the steam turbine unit, and the perception array includes multiple monitoring points; The monitoring unit includes multiple monitoring sub-modules, and the monitoring sub-modules are set at each monitoring point; The central control unit includes: The first control module is used to establish a regulation model, and the regulation model includes: a feedforward compensation model and a feedback correction model; The second control module is used to obtain the operation monitoring data collected by the perception array, and set the operation control strategy according to the operation monitoring data and the preset regulation model; The third control module is used to determine whether to generate an update instruction for the regulation model according to the preset update cycle.

9. The high-voltage bypass intelligent cooling and pressure reduction dynamic control system as described in claim 8, characterized in that, The first control module is further used for: Construct multiple lag scenarios according to the operation record data of the steam turbine unit; Establish a lag scenario sequence A, A = (a1, a2... ai... an), where ai is the i-th lag scenario; n is the number of lag scenarios; Set the feedforward sub-strategies of each lag scenario in sequence, and construct a compensation sub-model according to all the feedforward sub-strategies; Set ai as the target scenario in sequence according to the lag scenario sequence A; Construct a mapping parameter set of the target scenario based on the perception array and the operation record data; Set the feedforward sub-strategy of the target scenario, and the feedforward sub-strategy includes: feedforward duration and instruction change amplitude; Generate the mapping parameter sets of each lag scenario in sequence; Construct a prediction sub-model according to all the mapping parameter sets; Generate a feedforward compensation model, and the feedforward compensation model includes: a prediction sub-model and a compensation sub-model; Set the basic feedback strategy; Construct multiple disturbance operating conditions according to the mapping parameter sets of all the lag scenarios; Establish a disturbance operating condition sequence B, B = (b1, b2... bi... bn), where bi is the disturbance operating condition corresponding to the i-th lag scenario; n is the number of disturbance operating conditions; Set the correction sub-strategies of each disturbance operating condition in sequence; Construct a feedback correction model according to the basic feedback strategy and all the correction sub-strategies; Establish a regulation model, and the regulation model includes: a feedforward compensation model and a feedback correction model.

10. The high-voltage bypass intelligent de-cooling and pressure reduction dynamic control system as described in claim 9, characterized in that, The second control module is further used for: Preset multiple regulation cycles; Set the start time node of each regulation cycle as the monitoring time node; Obtain the operation monitoring data collected by the perception array at the current monitoring time node; A primary mapping set is generated based on the preprocessing results of the operational monitoring data; The prediction sub-model generates the fit values ​​of the first-level mapping set and each mapping parameter set; The lag scenario corresponding to the maximum value in the fit value is defined as the anchoring scenario; The execution feedforward strategy for the current control cycle is set according to the anchoring scenario; An execution feedback strategy is generated based on the correction sub-strategy and the basic feedback strategy corresponding to the disturbance condition in the anchoring scenario. Set an operation control strategy, which includes: an execution feedforward strategy and an execution feedback strategy.

Images