Coal pulverizing system and coal feeder low voltage ride through collaborative control method and system

Abstract

The application relates to a coal pulverizing system and coal feeder low-voltage ride-through collaborative control method and system, and belongs to the technical field of automatic control of coal-fired units. The method comprises the following steps: collecting system data in real time, performing collaborative state estimation through a dual-core driving model combining an LSTM integrated with a Monte Carlo Dropout module and a physical model; based on the estimated value, generating start-stop control instructions through model predictive control (MPC) rolling optimization; judging the low-voltage working condition through a two-out-of-three logic, triggering the coal feeder protection process and adjusting the pulverizing system parameters, and dynamically assigning the coal feeder full-stop protection delay according to the pulverizing working condition; and constructing a four-level safety mechanism including a low-voltage special protection level to perform collaborative risk disposal. The application solves the system interruption problem caused by independent operation of the pulverizing system control and the coal feeder protection, realizes self-adaption and uninterrupted and smooth operation of the whole pulverizing start-stop process under voltage fluctuation, and significantly improves the operation reliability, safety and economy of the unit.

Description

Technical Field

[0001] This application relates to the field of automatic control technology for coal-fired power units, and more specifically, to a method and system for coordinated control of low voltage ride-through of the coal pulverizing system and the coal feeder. Background Technology

[0002] In coal-fired power generating units, the safe and economical start-up and shutdown control of the pulverizing system and the low-voltage ride-through (LVRT) protection of the coal feeder are two crucial technologies. Currently, these two systems are typically designed and operated independently. Start-up and shutdown optimization of the pulverizing system mainly focuses on energy consumption, wear, and environmental indicators, while the low-voltage protection of the coal feeder is implemented as a separate electrical protection logic, triggering tripping or ride-through reset only when a voltage drop is detected.

[0003] This fragmented operation mode reveals serious problems in practice, especially under deep peak-shaving conditions with frequent grid voltage fluctuations: When a brief voltage drop occurs in the grid, the coal feeder trips according to its independent protection logic. However, this directly interrupts the ongoing start-up and shutdown process of the pulverizing system, which has strict timing and parameter requirements. This leads to drastic fluctuations in key parameters such as the pulverizer outlet temperature and primary air volume, easily causing serious consequences such as coal blockage, pulverizer interruption, or even boiler main fuel tripping. This forces the unit to reduce load unplanned, threatening grid security and power plant economics. Conversely, the coal feeder's self-starting after a low-voltage ride-through also fails to coordinate with the real-time status of the pulverizing system, posing a risk of asynchronous restart. Therefore, an integrated solution that can deeply coordinate the control of the pulverizing system with the low-voltage protection of the coal feeder is urgently needed. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method and system for coordinated low-voltage ride-through control of the coal pulverizing system and the coal feeder.

[0005] The technical solution of this invention is as follows: This invention proposes a coordinated control method for low-voltage ride-through of a coal pulverizing system and a coal feeder, comprising the following steps: Real-time data collection of pulverizing system operation and coal feeder voltage is performed, and a physical-data dual-core driven model is used to estimate the coordinated state of the pulverizing system and the coal feeder, thus obtaining the estimated value of the coordinated state of the pulverizing system and the coal feeder. Based on the cooperative state estimation, with the optimization objectives of minimizing energy consumption, equipment wear, environmental protection indicators, and low voltage protection reliability of the pulverizing system, the start-stop control command sequence of the pulverizing system containing adaptive correction parameters for low voltage conditions is generated through model predictive control (MPC) rolling optimization. The low voltage condition is determined by a 3-out-of-2 logic. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in conjunction. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. Based on the coordinated state estimation value and the start-stop control command sequence of the pulverizing system, a four-level active safety protection mechanism including a low-voltage dedicated protection level is constructed to coordinate the handling of fault risks caused by low-voltage operating conditions.

[0006] Preferably, the physical-data dual-core driven model includes: The physical mechanism model is constructed based on the principles of mass conservation, energy conservation, and heat transfer, and is used to describe the mass and energy dynamics of the pulverizing process; The data-driven model is built on a long short-term memory neural network (LSTM) and integrates a Monte Carlo Dropout module in the output part of the LSTM network. This module is used to output the state prediction value and the uncertainty variance of the prediction value through multiple forward propagations during model inference. The Extended Kalman Filter (EKF) algorithm is used to fuse the outputs of the physical mechanism model and the data-driven model. The uncertainty variance is dynamically introduced into the observation noise covariance of the EKF algorithm to adjust the fusion weights of the physical mechanism model and the data-driven model.

[0007] Preferably, the objective function for the model predictive control MPC rolling optimization is: ; In the formula: Let be the comprehensive objective function that needs to be minimized at time k; For prediction in the time domain; Let be the state variables of the pulverizing system predicted at time k+i in the future; The optimized solution for time k represents the control input variables to be applied at future time k+i-1. This is a function representing the energy consumption of the pulverizing system. For equipment wear term functions; For environmental protection indicator terms; This is a function for the low-voltage ride-through reliability term; The parameters are adaptively corrected for the low-voltage condition at time k.

[0008] Preferably, the execution delay of the coal feeder full stop protection is assigned a value based on the operating conditions of the pulverizing system, specifically including: If the pulverizing system is operating in a steady state, the execution delay is the first set value; If the powder-making system is in the start-up / shutdown phase, the execution delay is automatically adjusted to a second set value that is greater than the first set value.

[0009] Preferably, the four-level active safety protection mechanism includes a low-voltage specific protection level, which is handled as follows: when abnormal voltage fluctuations are detected but do not reach the trigger threshold of low-voltage conditions, the control parameters of the coal feeder, cold air damper and hot air damper are collaboratively corrected through feedforward compensation based on the collaborative state estimation value.

[0010] Preferably, the two-out-of-three logic judgment is as follows: continuously monitor the voltage of the feeder inlet circuit, the power circuit voltage, and the control circuit voltage. When any two of the three voltage values ​​are simultaneously lower than a preset threshold and continue to exceed the anti-vibration delay, it is determined that a low voltage condition is triggered.

[0011] Preferably, the method further includes: based on historical data under all operating conditions, online self-learning updates are performed on the optimization parameters of the physical-data dual-core driven model, the model predictive control (MPC) and the judgment threshold of low voltage operating conditions using a machine learning model.

[0012] On the other hand, the present invention also provides a low-voltage ride-through coordinated control system for a coal pulverizing system and a coal feeder, comprising: The state estimation module collects real-time operating data of the pulverizing system and voltage data of the coal feeder. It performs collaborative state estimation through a physical-data dual-core driven model to obtain the collaborative state estimate of the pulverizing system and the coal feeder. The optimization decision module, based on the collaborative state estimate, aims to achieve the lowest energy consumption, minimum equipment wear, optimal environmental protection indicators, and highest reliability of low voltage protection in the pulverizing system. Through model predictive control (MPC) rolling optimization, it generates a sequence of start-stop control commands for the pulverizing system that includes adaptive correction parameters for low voltage conditions. The low voltage protection and linkage execution module uses a 3-out-of-2 logic to determine the low voltage condition. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in linkage. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. The safety protection module, based on the collaborative state estimation value and the start-stop control command sequence of the pulverizing system, constructs a four-level active safety protection mechanism including a low-voltage specific protection level to collaboratively handle fault risks caused by low-voltage operating conditions.

[0013] In another aspect, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the low-voltage ride-through coordinated control method of the coal pulverizing system and the coal feeder as described in any embodiment of the present invention.

[0014] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the low-voltage ride-through coordinated control method for the coal pulverizing system and the coal feeder as described in any embodiment of the present invention.

[0015] This invention offers the following advantages: It deeply integrates the low-voltage ride-through protection of the coal feeder as an embedded collaborative element into the start-up and shutdown control framework of the pulverizing system, breaking down system barriers and achieving a fundamental shift from "equipment-level independent protection" to "system-level collaborative control and protection," ensuring the continuity and stability of the entire pulverizing start-up and shutdown process under voltage fluctuations. Furthermore, through mechanisms such as two-out-of-three logic judgment, condition-based dynamic delay protection, and four-level active collaborative protection, it significantly reduces the interruption rate of the pulverizing system start-up and shutdown process and the risk of false tripping of the coal feeder caused by voltage fluctuations. Based on the collaborative state estimation of the physical-data dual-core model and MPC rolling optimization with low-voltage reliability objectives, it can proactively generate control commands that adapt to voltage fluctuations, improving the stability of the control process. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0018] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0019] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0020] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0021] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0022] Example 1: To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present application and with reference to the accompanying drawings.

[0023] To address the problems of existing technologies, this invention provides a method for coordinated low-voltage ride-through control of a coal pulverizing system and a coal feeder, comprising the following steps: Real-time data collection of pulverizing system operation and coal feeder voltage is performed, and a physical-data dual-core driven model is used to estimate the coordinated state of the pulverizing system and the coal feeder, thus obtaining the estimated value of the coordinated state of the pulverizing system and the coal feeder. In this embodiment, the pulverizing system's operating data is acquired in real time from sensor signals installed at key process points of the pulverizing system via the analog and digital input modules of the distributed control system (DCS). Specifically, this includes: mill current, primary air volume, mill outlet temperature, mill inlet pressure, and coal feed rate.

[0024] The coal feeder voltage data is acquired from the voltage transformer signals in the coal feeder distribution cabinet via the analog input module of the DCS. Specifically, this includes: the coal feeder incoming circuit voltage (i.e., the PC section bus voltage), the power circuit voltage (i.e., the input voltage of the coal feeder frequency converter), and the control circuit voltage (i.e., the operating power supply voltage of the coal feeder controller).

[0025] Real-time collected pulverizing system operation data and coal feeder voltage data are input into a physical-data dual-core driven model. The physical mechanism model performs mechanism inference based on the current input and the state at the previous moment; the LSTM model performs data-driven state prediction. An Extended Kalman Filter (EKF) algorithm is used as the fusion processor. The EKF uses the predictions from the physical mechanism model as prior estimates, and uses the predictions from the LSTM model and its real-time prediction uncertainty as observation values ​​and observation noise covariance for data fusion and state updates.

[0026] In this embodiment, the physical-data dual-core driven model specifically includes: Physical mechanism model: Based on the principles of mass conservation, energy conservation and heat transfer, a set of nonlinear differential equations is established, including equipment such as coal mill, separator, and powder pipe, to describe the mass and energy dynamics of the pulverizing process.

[0027] Data-driven model: Constructed using a Long Short-Term Memory (LSTM) neural network, its input is the aforementioned multi-dimensional time-series operational data, used to learn unmodeled dynamic characteristics and complex relationships in the actual system. At the end of the LSTM, i.e., the output portion, a Monte Carlo Dropout (MC Dropout) module is integrated. This module remains active during inference after model training, enabling the network to generate a prediction distribution for the same set of inputs through multiple forward propagations, rather than a single prediction point. Its output includes not only the predicted state value but also the uncertainty variance of that prediction. Specifically, it can be represented as: ; ; ; In the formula: Let t be the state prediction vector output by LSTM under the tth Monte Carlo Dropout forward propagation sampling at time t. This vector contains key internal state estimates of the pulverizing system and coal feeder, such as the coal powder inventory and the degree of coal powder dryness in the pulverizer. Let be the input data vector fed into the LSTM network at time k; T is the number of Monte Carlo samplings. Let be the final state prediction output by the LSTM model at time k. It is the arithmetic mean of the predictions from T Monte Carlo samplings, serving as the best point estimate of the system state by the data-driven model. Let Variance be the uncertainty variance matrix of the LSTM model prediction at time k. It is calculated by taking T samples of the predicted values. Its mean The average of the squared deviations is used to quantify the confidence or uncertainty of data-driven model predictions.

[0028] In the EKF fusion process, the uncertainty variance of the LSTM model prediction is dynamically introduced. The recursive formula for the EKF prediction and update step is as follows: Prediction steps are based on physical models: ; ; In the formula: This is the prior state prediction vector made by the physical mechanism model at time k based on the state at time k-1; This is the state transition function of the physical mechanism model; Let be the posterior state estimate obtained after EKF fusion at time k, which is the optimal estimate at the previous time. This refers to the control input vector applied to the system at time k-1, such as the setpoint of the coal feeder speed, the opening of the cold air damper, and the opening of the hot air damper. For time k, the prediction is based on the prior state. The corresponding prior estimation error covariance matrix; Let k be the state transition function. Regarding the status The Jacobian matrix is ​​used to linearize the nonlinear model during the filtering process; Let be the posterior estimation error covariance matrix at time k-1; Let be the process noise covariance matrix at time k, used to model the inaccuracies of the physical model itself and unmodeled disturbances.

[0029] Update step fusion data-driven prediction: The output of LSTM and Considered as observation value and its observation noise covariance :

[0030] In the formula: For time k, the adaptive observation noise covariance matrix; and As the adjustment coefficient, and Dynamic association.

[0031] The Kalman gain was then calculated. And update the state estimate: ; ; ; In the formula: Here is the Kalman gain matrix at time k, used to determine whether to trust the predictions of the physical model or the observations of the data-driven model more during state updates; The observation matrix at time k describes how the state vector is mapped to the observation vector. In this embodiment, since the state estimation and LSTM prediction use the same set of state variables, therefore... I is the identity matrix; Let k be the posterior cooperative state estimate after EKF fusion update. The updated posterior estimation error covariance matrix at time k will be used for the prediction step at the next time step (k+1).

[0032] The aforementioned uncertainty-guided adaptive EKF fusion, when the data-driven model predicts uncertainty... When increased, equivalent observation noise Increasing the Kalman gain leads to... Reducing the number of EKF fusion results will increase the reliability of predictions from the physical mechanism model. Conversely, people tend to trust the predictions of data-driven models more. This enables adaptive weight fusion based on online uncertainty, significantly improving the robustness of cooperative state estimation under complex and variable conditions.

[0033] The final output is a coordinated state estimate, which is an enhanced state vector that integrates mechanistic and data information. It includes at least: the coal powder inventory in the coal mill, the degree of coal powder dryness, the thermal stress of key equipment (such as the grinding rollers), the real-time voltage status assessment of the coal feeder, and the inverter health index. These states are key variables that cannot be directly measured by the original sensors but are crucial to control.

[0034] Based on the cooperative state estimation, with the optimization objectives of minimizing energy consumption, equipment wear, environmental protection indicators, and low voltage protection reliability of the pulverizing system, the start-stop control command sequence of the pulverizing system containing adaptive correction parameters for low voltage conditions is generated through model predictive control (MPC) rolling optimization. In this embodiment, the estimated cooperative state is used as the initial state of the MPC controller. A comprehensive objective function is constructed with the optimization objectives of minimizing energy consumption of the pulverizing system, minimizing equipment wear, optimizing environmental performance, and maximizing low-voltage protection reliability: ; In the formula: Let be the comprehensive objective function that needs to be minimized at time k; For prediction in the time domain; Let k be the predicted state variables of the pulverizing system at time k+i in the future. The optimized solution for time k represents the control input variables to be applied at future time k+i-1. Let be the energy consumption term function of the pulverizing system, which is a scalar or vector positively correlated with the system's energy consumption; For equipment wear term function, it is a scalar or vector that is positively correlated with the wear rate of key equipment such as grinding rollers and bearings; For environmental protection indicator terms, there is a scalar or vector that is positively correlated with the pollutant generation potential. For low voltage ride reliability term function, it is a scalar or vector that is negatively correlated with the ability of the pulverizing system to maintain stable operation during voltage dips. The smaller the value, the higher the reliability. The adaptive correction parameters for low-voltage conditions at time k are dynamically adjusted based on the voltage risk reflected in the cooperative state estimation. These parameters are used to strengthen or weaken the impact of low-voltage ride-through reliability targets during optimization. When the estimated voltage drop risk is high... Increasing this makes the optimizer more inclined to generate control instructions that improve system stability.

[0035] In each control cycle (e.g., 1 second), the MPC controller predicts the system behavior for a future period (e.g., 30 minutes) based on the latest cooperative state estimate.

[0036] Under the premise of satisfying various operational constraints, by solving optimization problems, such as using numerical optimization algorithms such as sequential quadratic programming or interior point method, a series of optimal control command sequences in the future control time domain are calculated. The optimal control command sequences include: coal feeder speed setpoint, cold air damper opening, hot air damper opening, primary air volume setpoint, etc.

[0037] During the optimization process, adaptive correction parameters for low-voltage conditions are incorporated as key constraints or penalty terms. For example, when the "coal feeder voltage status assessment" in the coordinated state estimate indicates a risk of voltage decline, the optimization algorithm automatically relaxes the production efficiency target and prioritizes generating control commands that can improve system inertia and slow down the action speed in preparation for possible voltage drops.

[0038] Finally, the first element of the calculated control instruction sequence (i.e. the control instruction to be executed at the current moment) is sent to the output module of the DCS, and the process is repeated in the next cycle.

[0039] The low voltage condition is determined by a 3-out-of-2 logic. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in conjunction. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. The system employs a 3-out-of-2 logic to determine low-voltage conditions, continuously monitoring and collecting the voltage of the coal feeder's inlet circuit, power circuit, and control circuit. When any two of these three voltage values ​​are simultaneously lower than a preset threshold (85% of the rated voltage in this embodiment) and remain below this threshold for more than one anti-jitter delay (20ms in this embodiment), the protection logic block within the DCS determines it as a low-voltage condition and triggers it.

[0040] After a low-voltage condition is triggered, the DCS executes the pre-configured fault recovery sequence: Fault Code Merging: Multiple scattered fault alarms that may be caused by voltage fluctuations (such as "inverter fault" and "speed loss") are merged into a unified "low voltage ride-through" event code to avoid interference from multiple alarms. Specifically, in the DCS logic configuration, related fault signals (such as inverter fault and speed feedback loss) are merged through an 'OR' logic module and a unified low voltage ride-through event flag is output.

[0041] Inverter self-reset: Sends a reset command to the coal feeder inverter and confirms that its parameters have been set according to the preset scheme (such as enabling the automatic reset function for faults such as undervoltage and overcurrent, resetting 3 times with an interval of 10 seconds each time).

[0042] Coal feeder self-start: After receiving the "ready" signal from the frequency converter and the voltage stabilizes, the coal feeder automatically restores to the speed setting value before the fault or the new setting value given by the S2 command according to the preset smooth speed curve.

[0043] Pulverizing system-side linkage adjustment: Simultaneously executed when the low-pressure ride-through protection process is triggered: Based on the severity of the low-voltage event and the adaptive correction parameters in the instruction sequence, normal control instructions are temporarily overridden, and the primary air volume, cold / hot damper opening, etc. are dynamically adjusted to maintain the coal mill outlet temperature and primary air-to-coal ratio within a safe range, preventing coal blockage or overheating.

[0044] Dynamic assignment of full shutdown protection delay: Query the current operating status of the pulverizing system. If the pulverizing system is in normal steady-state operation, the trip delay of the coal feeder's "full shutdown" protection will remain at the normal value (e.g., 1 second); if the pulverizing system is in a vulnerable stage of start-up and shutdown (e.g., starting warm-up mill, stopping purging), the delay will be automatically extended (e.g., to 3 seconds) to buy time for the coordinated adjustment of the pulverizing system and avoid accidental triggering of the full shutdown due to a brief voltage drop, which would lead to the operation of the unit's main protection.

[0045] Based on the coordinated state estimation value and the start-stop control command sequence of the pulverizing system, a four-level active safety protection mechanism including a low-voltage dedicated protection level is constructed to coordinate the handling of fault risks caused by low-voltage operating conditions.

[0046] The four-level collaborative protection includes: Level 1 - Low Voltage Specific Protection Level: When abnormal voltage fluctuations are detected but do not reach the trigger threshold for low voltage conditions, feedforward compensation is performed based on state estimation. For example, the coal feeder rate is slightly reduced in advance, and the hot air damper is closed proportionally to reduce system inertia and actively resist potential voltage disturbances.

[0047] Level 2 - Early Warning Level: When the voltage drop reaches the warning threshold or the estimated value of a critical state (such as the dryness of pulverized coal) deviates from the normal range, an early warning will be issued on the control interface, and all current control parameters and states will be automatically backed up.

[0048] Level 3 - Degradation Level: When a low voltage condition has been triggered or the condition of critical equipment has deteriorated, the control of the pulverizing system will be automatically switched to "safety priority mode", suspending non-core optimization objectives (such as reducing energy consumption) and making every effort to ensure the stability of core safety parameters such as coal mill outlet temperature and primary air volume.

[0049] Level 4 - Emergency Shutdown: When coordinated measures are ineffective and it is determined that the fault (such as prolonged voltage loss or severe coal blockage in the coal mill) will endanger the safety of the equipment or unit, the pulverizing system and related equipment will be shut down in an emergency according to the preset safety sequence, and a trip signal will be sent to the superior system.

[0050] Each of the aforementioned protection levels operates in concert, not in isolation. For example, during degradation control, the control commands are fed back to the MPC optimizer in real time as new constraints, ensuring that the subsequent generated control command sequence remains consistent with the current protection strategy. This achieves closed-loop coordination between "state estimation, optimized control, and active protection."

[0051] In a preferred embodiment of this invention, the method further includes the step of: based on historical data under all operating conditions, performing online self-learning updates on the optimization parameters of the physical-data dual-core driven model, the model predictive control (MPC) model, and the judgment threshold for low-voltage operating conditions using a machine learning model.

[0052] Example 2: This embodiment provides a coordinated control system for a coal pulverizing system and a coal feeder under low voltage ride-through, including: The state estimation module collects real-time operating data of the pulverizing system and voltage data of the coal feeder. It performs collaborative state estimation through a physical-data dual-core driven model to obtain the collaborative state estimate of the pulverizing system and the coal feeder. The optimization decision module, based on the collaborative state estimate, aims to achieve the lowest energy consumption, minimum equipment wear, optimal environmental protection indicators, and highest reliability of low voltage protection in the pulverizing system. Through model predictive control (MPC) rolling optimization, it generates a sequence of start-stop control commands for the pulverizing system that includes adaptive correction parameters for low voltage conditions. The low voltage protection and linkage execution module uses a 3-out-of-2 logic to determine the low voltage condition. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in linkage. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. The safety protection module, based on the collaborative state estimation value and the start-stop control command sequence of the pulverizing system, constructs a four-level active safety protection mechanism including a low-voltage specific protection level to collaboratively handle fault risks caused by low-voltage operating conditions.

[0053] Example 3: This embodiment proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the low-voltage ride-through coordinated control method of the coal pulverizing system and the coal feeder as described in any embodiment of the present invention.

[0054] Example 4: This embodiment proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the low-voltage ride-through coordinated control method for the coal pulverizing system and the coal feeder as described in any embodiment of the present invention.

[0055] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0056] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0057] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0058] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for coordinated low-voltage ride-through control of a coal pulverizing system and a coal feeder, characterized in that, Includes the following steps: Real-time data collection of pulverizing system operation and coal feeder voltage is performed, and a physical-data dual-core driven model is used to estimate the coordinated state of the pulverizing system and the coal feeder, thus obtaining the estimated value of the coordinated state of the pulverizing system and the coal feeder. Based on the cooperative state estimation, with the optimization objectives of minimizing energy consumption, equipment wear, environmental protection indicators, and low voltage protection reliability of the pulverizing system, the start-stop control command sequence of the pulverizing system containing adaptive correction parameters for low voltage conditions is generated through model predictive control (MPC) rolling optimization. The low voltage condition is determined by a 3-out-of-2 logic. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in conjunction. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. Based on the coordinated state estimation value and the start-stop control command sequence of the pulverizing system, a four-level active safety protection mechanism including a low-voltage dedicated protection level is constructed to coordinate the handling of fault risks caused by low-voltage operating conditions.

2. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The physical-data dual-core driven model includes: The physical mechanism model is constructed based on the principles of mass conservation, energy conservation, and heat transfer, and is used to describe the mass and energy dynamics of the pulverizing process; The data-driven model is built on a long short-term memory neural network (LSTM) and integrates a Monte Carlo Dropout module in the output part of the LSTM network. This module is used to output the state prediction value and the uncertainty variance of the prediction value through multiple forward propagations during model inference. The Extended Kalman Filter (EKF) algorithm is used to fuse the outputs of the physical mechanism model and the data-driven model. The uncertainty variance is dynamically introduced into the observation noise covariance of the EKF algorithm to adjust the fusion weights of the physical mechanism model and the data-driven model.

3. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The objective function for the model predictive control MPC rolling optimization is: ； In the formula: Let be the comprehensive objective function that needs to be minimized at time k; For prediction in the time domain; Let be the state variables of the pulverizing system predicted at time k+i in the future; For time k, the optimized solution is used to determine the control input variables to be applied at future time k+i-1. This is a function representing the energy consumption of the pulverizing system. For equipment wear term functions; For environmental protection indicator terms; This is a function for the low-voltage ride-through reliability term; The parameters are adaptively corrected for the low-voltage condition at time k.

4. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The execution delay of the coal feeder full stop protection is assigned a value based on the operating conditions of the pulverizing system, specifically including: If the pulverizing system is operating in a steady state, the execution delay is the first set value; If the powder-making system is in the start-up / shutdown phase, the execution delay is automatically adjusted to a second set value that is greater than the first set value.

5. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The four-level active safety protection mechanism includes a low-voltage specific protection level, which handles the following: when abnormal voltage fluctuations are detected but do not reach the trigger threshold of low-voltage conditions, the control parameters of the coal feeder, cold air damper, and hot air damper are collaboratively corrected through feedforward compensation based on the collaborative state estimate.

6. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The three-out-of-two logic judgment is as follows: continuously monitor the voltage of the feeder inlet circuit, the power circuit voltage, and the control circuit voltage. When any two of the three voltage values ​​are simultaneously lower than the preset threshold and continue to exceed the anti-vibration delay, it is determined that a low voltage condition is triggered.

7. The low-voltage ride-through coordinated control method for the coal pulverizing system and coal feeder according to claim 1, characterized in that: The method further includes: based on historical data under all operating conditions, online self-learning updates are performed on the optimization parameters of the physical-data dual-core driven model, the model predictive control (MPC) and the judgment threshold of low voltage operating conditions through a machine learning model.

8. A low-voltage ride-through coordinated control system for a coal pulverizing system and a coal feeder, characterized in that, include: The state estimation module collects real-time operating data of the pulverizing system and voltage data of the coal feeder. It performs collaborative state estimation through a physical-data dual-core driven model to obtain the collaborative state estimate of the pulverizing system and the coal feeder. The optimization decision module, based on the collaborative state estimate, aims to achieve the lowest energy consumption, minimum equipment wear, optimal environmental protection indicators, and highest reliability of low voltage protection in the pulverizing system. Through model predictive control (MPC) rolling optimization, it generates a sequence of start-stop control commands for the pulverizing system that includes adaptive correction parameters for low voltage conditions. The low voltage protection and linkage execution module uses a 3-out-of-2 logic to determine the low voltage condition. After the low voltage condition is triggered, the low voltage ride-through protection process of the coal feeder is executed, and the start and stop parameters of the pulverizing system are adjusted in linkage. The execution delay of the coal feeder full stop protection is assigned according to the pulverizing system condition. The safety protection module, based on the collaborative state estimation value and the start-stop control command sequence of the pulverizing system, constructs a four-level active safety protection mechanism including a low-voltage specific protection level to collaboratively handle fault risks caused by low-voltage operating conditions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the low-voltage ride-through coordinated control method for the coal pulverizing system and the coal feeder as described in claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the low-voltage ride-through coordinated control method for the coal pulverizing system and the coal feeder as described in claims 1-7.

Images