Millisecond-level cooperative control method and system of solid waste-based heat storage device and thermal power generating unit

Abstract

The application discloses a millisecond-level cooperative control method and system of solid waste-based heat storage device and thermal power generating unit, the method comprising: generating a digital twin of solid waste-based heat storage device and thermal power generating unit by constructing a heat storage and power generation combined simulation model; obtaining the running state data of the solid waste-based heat storage device and the thermal power generating unit in real time through sensors and control equipment deployed in the production site; generating initial real-time control instructions by executing a proportional-integral-derivative control algorithm; performing multi-objective dynamic optimization and conflict resolution on the initial real-time control instructions based on the optimization target output by the digital twin through the edge computing equipment, and generating optimized control instructions; comparing the feedback data and the simulation data of the digital twin by the edge computing equipment and generating a comparison result, and correcting the time-varying parameters related to the performance of the solid waste-based heat storage device in the heat storage and power generation combined simulation model.

Description

Technical Field

[0001] This application relates to the technical field of industrial energy conservation and environmental protection, and in particular to a millisecond-level collaborative control method and system for solid waste-based thermal storage devices and thermal power units. Background Technology

[0002] While existing technologies, such as dynamic feedback control-based swing bridge attitude monitoring systems, achieve multi-source sensor integration and edge computing data fusion, they primarily focus on bridge construction technology and lack millisecond-level collaborative control capabilities for thermal power and solid waste-based thermal storage systems. Furthermore, although the latest publicly available technologies have addressed issues like heterogeneous protocols among multi-brand CNC equipment, their application is limited to CNC equipment within the Industrial Internet, failing to provide solutions for the real-time control needs of complex systems like thermal storage and thermal power units.

[0003] While the aforementioned technical solutions possess a degree of innovation and practicality within their respective application fields, they exhibit significant shortcomings in millisecond-level coordinated control of solid waste-based thermal storage and thermal power units. Specifically, they lack real-time hardware interaction capabilities between the thermal storage medium and the thermal power unit, and are deficient in technologies related to pre-resolution of control conflicts and multi-objective dynamic arbitration.

[0004] Therefore, how to achieve millisecond-level coordinated control between solid waste-based thermal storage devices and thermal power units to improve system response speed and control accuracy has become an urgent technical problem to be solved. Summary of the Invention

[0005] This application proposes a millisecond-level collaborative control method and system for solid waste-based thermal storage devices and thermal power units to address the shortcomings of the aforementioned prior art.

[0006] According to a first aspect of the embodiments of this application, a millisecond-level coordinated control method for a solid waste-based thermal storage device and a thermal power unit is provided, comprising: By constructing a joint simulation model of thermal storage and power generation, a digital twin of the solid waste-based thermal storage device and the thermal power unit is generated. The joint simulation model of thermal storage and power generation is used to simulate the energy and material flow process of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on the received real-time data and output optimization targets. By deploying sensors and control equipment at the production site, the operating status data of the solid waste-based thermal storage device and the thermal power unit are acquired in real time, and the operating status data is transmitted to the edge computing device through a time-sensitive network channel. The edge computing device generates initial real-time control commands based on the received operating status data by executing a proportional-integral-derivative control algorithm. Using the edge computing device, based on the optimization objectives output by the digital twin, the initial real-time control commands are dynamically optimized and conflict resolved through multi-objective processes, and optimized control commands are generated. The optimized control commands are sent to the actuators of the solid waste-based thermal storage device and the thermal power unit, respectively, and the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit are collected as feedback data. The edge computing device compares the feedback data with the simulation data of the digital twin and generates a comparison result. Based on the comparison result, the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation are corrected.

[0007] In some implementations, a digital twin of the solid waste-based thermal storage device and the thermal power unit is generated by constructing a joint simulation model of thermal storage and power generation, including: A joint simulation model for thermal energy storage and power generation was constructed based on the energy storage characteristic parameters of solid waste-based phase change materials. Based on the aforementioned combined thermal storage and power generation simulation model, the process of thermal storage and power generation working together under different operating conditions is simulated and the steady-state operating point is determined. A digital twin of the solid waste-based thermal storage device and the thermal power unit is generated based on the steady-state operating point; The energy storage characteristic parameters include energy storage density, charge / discharge rate, and temperature resistance level.

[0008] In some embodiments, the construction of a joint simulation model for thermal storage and power generation based on the energy storage characteristic parameters of solid waste-based phase change materials includes: The solid waste-based phase change material is prepared using blast furnace steel slag, converter steel slag, carbide slag, phosphogypsum, or tailings filter residue as raw materials. Modified magnesium bricks were selected as high-temperature pressure vessels, and a high-temperature pressure solid waste melting and heat storage device was made based on the solid waste-based phase change material and modified magnesium bricks. A joint simulation model of thermal storage and power generation is constructed based on the physical characteristics of the high-temperature pressure-bearing solid waste melting thermal storage device. The aforementioned joint simulation model of thermal storage and power generation is integrated into the industrial internet platform.

[0009] In some embodiments, the edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit, used for communication and collaborative decision-making between the solid waste-based thermal storage device and the thermal power unit. The method further includes: Based on sensors deployed at the production site, control equipment deployed at the production site, actuators of the solid waste-based thermal storage device, actuators of the thermal power unit, and edge computing equipment, closed-loop control with millisecond-level response is performed on the solid waste-based thermal storage device and the thermal power unit.

[0010] In some embodiments, the edge computing device integrates a field-programmable gate array (FPGA) and a graphics processor (GPU) to accelerate the execution of control algorithms through hardware parallel computing. The method further includes: Based on the received operating status data, vector addition is performed by the field-programmable gate array, and complex function floating-point operations are performed by the graphics processing unit.

[0011] In some embodiments, the method further includes: With the goal of maximizing energy supply and with total energy storage and energy storage density as constraints, a multi-objective evolutionary model for non-dominated ranking is constructed for multi-objective dynamic optimization and conflict resolution in litigation. The multi-objective evolutionary model is implemented based on the Pareto optimal solution set selection method, and the Lagrange multiplier method is used to construct dummy variables to replace the original control variables.

[0012] In some implementations, the actual operating parameters of the solid waste-based thermal storage device and the actuator of the thermal power unit are collected every fifty milliseconds. The time-varying parameters include the decay rate of the thermal conductivity of the solid waste-based thermal storage device with time and temperature.

[0013] In some implementations, the multi-objective dynamic optimization and conflict resolution of the initial real-time control command includes: When the main steam pressure drops below a preset threshold in the initial real-time control command or the solid waste-based thermal storage device triggers a heat overflow alarm, it is determined that a conflict has occurred. Suspend the execution of the current control command and generate a sequence of candidate compensation strategies, including immediate heat release, normal energy storage, delayed energy storage, maintaining the original state, starting tracking, and no operation. The Pareto optimization method is used to sort the candidate compensation strategy sequence to obtain the optimal solution set; The conflict is resolved based on the optimal solution set until the conflict is eliminated.

[0014] According to a second aspect of this application, a millisecond-level coordinated control system for a solid waste-based thermal storage device and a thermal power unit is provided, comprising: The digital twin generation module is used to generate digital twins of solid waste-based thermal storage devices and thermal power units by constructing a joint simulation model of thermal storage and power generation. The joint simulation model of thermal storage and power generation is used to simulate the energy and material flow processes of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on received real-time data and output optimization targets. The status data acquisition module is used to acquire the real-time operating status data of the solid waste-based thermal storage device and the thermal power unit through sensors and control equipment deployed on the production site, and transmit the operating status data to the edge computing device through a time-sensitive network channel; The initial instruction generation module is used by the edge computing device to generate initial real-time control instructions based on the received operating status data by executing a proportional-integral-derivative control algorithm. An optimization instruction generation module is used to perform multi-objective dynamic optimization and conflict resolution on the initial real-time control instruction based on the optimization objectives output by the digital twin through the edge computing device, and generate optimized control instructions. The execution and feedback module is used to send the optimized control commands to the actuators of the solid waste-based thermal storage device and the thermal power unit, respectively, and to collect the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit as feedback data. The parameter correction module is used to compare the feedback data with the simulation data of the digital twin by the edge computing device and generate a comparison result, and correct the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation based on the comparison result.

[0015] In some embodiments, the edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit. Based on sensors deployed at the production site, control equipment deployed at the production site, the actuators of the solid waste-based thermal storage device, the actuators of the thermal power unit, and the edge computing device, a closed-loop control module is constructed to perform millisecond-level responses to the solid waste-based thermal storage device and the thermal power unit.

[0016] The beneficial effects of the millisecond-level coordinated control method and system for solid waste-based thermal storage devices and thermal power units in this application embodiment include at least the following: This application embodiment achieves a comprehensive and high-fidelity digital mapping of the combined operating status of thermal power units and solid waste-based thermal storage devices by constructing a joint simulation model of thermal power storage and power generation and generating a digital twin. This enables prior simulation and optimization analysis of different operating conditions in virtual space, providing an accurate prediction and planning basis for actual control decisions, and significantly improving the predictability and scientific nature of control. This application embodiment ensures the real-time acquisition and reliable transmission of key operating status data by deploying field sensors and using time-sensitive network channels to transmit data to edge computing devices, providing a stable and low-latency data foundation for subsequent millisecond-level real-time decisions, effectively overcoming the constraints of latency and jitter in traditional data transmission on rapid control. This application embodiment generates initial commands by executing proportional-integral-derivative control algorithms on edge computing devices, sinking core control functions to the field, reducing dependence on upper-level systems and communication latency, and realizing localized rapid basic adjustment of system dynamic changes, improving the directness and response speed of control. This application embodiment performs multi-objective dynamic optimization and conflict resolution of initial commands based on the optimization objectives of the digital twin, ensuring that control commands no longer satisfy only a single local objective. By comprehensively optimizing multiple objectives such as energy utilization and system safety, and proactively resolving potential conflicts, this application achieves a leap from basic stability control to system-level collaborative optimization control, enhancing the overall economy and operational safety of the system. This embodiment of the application constructs a real-time control closed loop of perception, decision-making, execution, and feedback by issuing optimization commands to the actuators and collecting feedback from actual operating parameters. This closed loop enables continuous adjustment of control based on execution results, improving adaptability to external disturbances and changes in internal parameters, as well as control accuracy. This embodiment of the application achieves dynamic self-calibration of the digital twin by comparing feedback data with digital twin simulation data and correcting the model's time-varying parameters. This ensures that the simulation model can track long-term changes in the physical entity's characteristics, maintaining the long-term fidelity of the digital twin, thereby guaranteeing the continuous effectiveness of model-based optimization and control strategies. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the millisecond-level collaborative control method between a solid waste-based thermal storage device and a thermal power unit according to an embodiment of this application is shown. Figure 2 This paper illustrates a flowchart of a specific embodiment of the millisecond-level collaborative control method between a solid waste-based thermal storage device and a thermal power unit according to an embodiment of this application. Figure 3 A flowchart illustrating the conflict pre-resolution mechanism of an embodiment of this application is shown; Figure 4 This diagram illustrates a comparison of the response times of the embodiments of this application and conventional methods under different load variations. Figure 5A schematic diagram of the structure of the millisecond-level collaborative control system between the solid waste-based thermal storage device and the thermal power unit according to an embodiment of this application is shown. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following millisecond-level coordinated control method and system for solid waste-based thermal storage devices and thermal power units will be described clearly and completely in conjunction with the accompanying drawings of the embodiments of this application. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed embodiments of the present application, but merely to illustrate selected embodiments of the present application. Other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are all within the scope of protection of the embodiments of the present application.

[0020] It can be noted that similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it will not be further defined and explained in subsequent figures according to the embodiments of this application.

[0021] This application discloses a millisecond-level collaborative control method and system for solid waste-based thermal energy storage devices and thermal power units. This millisecond-level collaborative control method is implemented based on a millisecond-level collaborative control system for solid waste-based thermal energy storage devices and thermal power units. The purpose of this method is to effectively improve system response speed, reduce control latency, ensure equipment safety, and optimize energy distribution by deploying a lightweight real-time operating system (RTOS) and field-programmable gate array (FPGA) hardware acceleration at the equipment edge layer, combined with a joint simulation model constructed from digital twins, and a multi-objective dynamic arbitration algorithm. This technical solution overcomes the shortcomings of existing technologies and achieves efficient, safe, and stable operation of solid waste-based thermal energy storage devices and thermal power units.

[0022] See attached document Figure 1 As shown, the millisecond-level collaborative control method between the solid waste-based thermal energy storage device and the thermal power unit includes steps 110-150 as follows.

[0023] Step 110: By constructing a joint simulation model of thermal storage and power generation, a digital twin of the solid waste-based thermal storage device and the thermal power unit is generated.

[0024] The combined simulation model of thermal storage and power generation is used to simulate the energy and material flow process of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on the received real-time data and output optimization targets.

[0025] In some implementations, a digital twin of a solid waste-based thermal energy storage device and a thermal power unit is generated by constructing a joint simulation model of thermal energy storage and power generation. This includes: constructing a joint simulation model of thermal energy storage and power generation based on the energy storage characteristic parameters of solid waste-based phase change materials; simulating the process of thermal energy storage and power generation working together under different operating conditions based on the joint simulation model and determining the steady-state operating point; and generating a digital twin of a solid waste-based thermal energy storage device and a thermal power unit based on the steady-state operating point.

[0026] The energy storage characteristic parameters include energy density, charge / discharge rate, and temperature resistance level.

[0027] In some implementations, the joint simulation model for thermal storage and power generation based on the energy storage characteristic parameters of solid waste-based phase change materials includes: preparing the solid waste-based phase change material using blast furnace slag, converter slag, carbide slag, phosphogypsum, or tailings filter residue as raw materials; selecting modified magnesia bricks as high-temperature pressure vessels, and fabricating a high-temperature pressure-bearing solid waste melting thermal storage device based on the solid waste-based phase change material and modified magnesia bricks; constructing the joint simulation model for thermal storage and power generation based on the physical characteristics of the high-temperature pressure-bearing solid waste melting thermal storage device; and integrating the joint simulation model for thermal storage and power generation into an industrial internet platform.

[0028] For example, by constructing a joint simulation model for thermal storage and power generation, the following steps are taken: using blast furnace slag, converter slag, carbide slag, phosphogypsum, and tailings filter residue as raw materials, and employing a vacuum degreasing and sintering process, high-temperature solid waste-based phase change materials with a thermal conductivity ≥3.5W / m·K are prepared; modified magnesia bricks are selected as high-temperature pressure vessels, and sealed and sintered at 1000°C~1100°C to establish a physical model of the solid waste melting thermal storage device (i.e., a joint simulation model for thermal storage and power generation); the solid waste melting thermal storage device is connected to the Industrial Internet, for example, integrated into a plant-grid interaction platform, and further, by comprehensively considering the energy and material flow processes of boilers, steam turbines, and solid waste-based thermal storage devices, a joint simulation model for thermal storage and power generation is established to simulate the steady-state operating point under different working conditions; and suitable thermal storage media and container materials are selected by comparing energy storage density, heat charging and releasing rate, and temperature resistance level as evaluation indicators.

[0029] In some implementations, the method further includes: comparing and selecting suitable thermal storage media and container materials using energy storage density, heat charging and discharging rate, and temperature resistance rating as key evaluation indicators. Based on the comparison results, a joint simulation model of thermal storage and power generation is constructed to generate a digital twin of the solid waste-based thermal storage device and the thermal power unit.

[0030] The digital twin constructed in this application is a high-fidelity dynamic mapping in information space of a high-temperature pressure-bearing solid waste melting thermal storage device (i.e., a solid waste-based thermal storage device) and its associated thermal power unit. Its core is a joint simulation model of thermal storage and power generation. This model is built based on the physical properties of solid waste-based phase change materials and system structural parameters, and can simulate the thermodynamic working process under different operating conditions. This digital twin interacts bidirectionally with the physical entity through a real-time data interface: on the one hand, it receives actual operating data uploaded by edge computing devices to dynamically calibrate model parameters (such as thermal conductivity decay rate) to ensure synchronization with the physical entity; on the other hand, it sends simulation-optimized control objectives and strategies to the edge computing devices, thereby achieving closed-loop precise control through deep integration of the physical system and the information model.

[0031] Step 120: Real-time operating status data of the solid waste-based thermal storage device and the thermal power unit are acquired through sensors and control equipment deployed at the production site, and the operating status data is transmitted to the edge computing device through a time-sensitive network channel.

[0032] The edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit, used for communication and collaborative decision-making between the solid waste-based thermal storage device and the thermal power unit.

[0033] To achieve real-time perception and collaborative control, this application embodiment continuously collects operational status data from various sensors (such as temperature and pressure sensors) and underlying control devices deployed at the solid waste-based thermal storage device and the thermal power unit production site. This data is then reliably transmitted to the edge computing device, which serves as the computing core, through a time-sensitive network channel with deterministic and low-latency characteristics, providing a data foundation for subsequent real-time decision-making and control command generation.

[0034] Step 130: The edge computing device generates initial real-time control commands based on the received operating status data by executing a proportional-integral-derivative control algorithm.

[0035] For example, the edge computing device integrates a field-programmable gate array and a graphics processor to accelerate the execution of control algorithms through hardware parallel computing.

[0036] In some implementations, the method further includes: performing vector addition calculations through the field-programmable gate array based on the received operating status data, and performing complex function floating-point operations through the graphics processing unit.

[0037] For example, this application embodiment utilizes edge computing technology to embed a scalable third-party computing module (i.e., an edge computing device) between the programmable logic controller (PLC) of the thermal storage device and the distributed control system (DCS) of the thermal power unit, enabling communication and data interaction between the two. To address the time coordination issue between the unit and the thermal storage device, Time-Sensitive Networking (TSN) technology is employed to resolve delay, jitter, and packet loss issues during signal transmission. Inside the edge computing device, a Field-Programmable Gate Array (FPGA) is used as the intelligent terminal and backplane processor to perform fine-grained decomposition of the Proportional-Integral-Derivative (PID) control program, transforming the serial PID algorithm into independent units that can be executed in parallel. For complex function calculations that cannot be efficiently solved by general-purpose digital signal processors (DSPs) or central processing units (CPUs), a built-in high-performance graphics processing unit (GPU) is used to improve floating-point arithmetic capabilities and accelerate numerical solution efficiency.

[0038] Step 140: Using the edge computing device, based on the optimization objective output by the digital twin, perform multi-objective dynamic optimization and conflict resolution on the initial real-time control command, and generate an optimized control command.

[0039] In some implementations, the method further includes: constructing a multi-objective evolutionary model for non-dominated ranking to perform multi-objective dynamic optimization and conflict resolution in litigation, with the goal of maximizing energy supply and the constraints of total energy storage and energy storage density.

[0040] For example, this multi-objective evolutionary model is performed based on the Pareto optimal solution set selection method, and uses the Lagrange multiplier method to construct dummy variables to replace the original control variables.

[0041] In some implementations, the multi-objective dynamic optimization and conflict resolution of the initial real-time control command includes: determining a conflict when the main steam pressure drop exceeds a preset threshold or the solid waste-based thermal storage device triggers a heat overflow alarm; pausing the execution of the current control command and generating a sequence of candidate compensation strategies including immediate heat release, normal energy storage, delayed energy storage, maintaining the original state, initiating tracking, and no operation; sorting the candidate compensation strategy sequence using the Pareto optimization method to obtain the optimal solution set; and resolving the conflict according to the optimal solution set until the conflict is eliminated.

[0042] For example, this application embodiment constructs a non-dominated ranking multi-objective evolutionary model for a closed-loop energy storage system based on solid waste materials, based on the optimization objectives output by the aforementioned digital twin. For instance, a Pareto frontier screening method based on the goal of maximizing overall benefits is used, with total energy storage and energy density as constraints, and a differential evolution algorithm is employed to seek the optimal solution. Simultaneously, a Lagrange multiplier method is used to construct an energy storage complementarity strategy, and dummy variables with resistance-capacitance characteristics are designed to reduce control conflicts between systems.

[0043] Step 150: The optimized control command is sent to the actuators of the solid waste-based thermal storage device and the thermal power unit respectively, and the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit are collected as feedback data respectively.

[0044] In some embodiments, the method further includes: performing closed-loop control of the solid waste-based thermal storage device and the thermal power unit with millisecond-level response based on sensors deployed at the production site, control equipment deployed at the production site, actuators of the solid waste-based thermal storage device, actuators of the thermal power unit, and the edge computing device.

[0045] For example, the actual operating parameters of the solid waste-based thermal storage device and the actuator of the thermal power unit are collected every fifty milliseconds.

[0046] For example, in this embodiment of the application, during system operation, the edge computing device receives actual operating data every 50ms and performs error analysis with the simulation data of the digital twin. Based on the analysis results, the decay rate of the thermal conductivity of the thermal storage device with time and temperature, as well as other parameters, in the joint simulation model of thermal storage and power generation are corrected. Finally, the global robustness of the closed-loop control system is ensured by using the Lyapunov stability verification method.

[0047] Step 160: The edge computing device compares the feedback data with the simulation data of the digital twin and generates a comparison result. Based on the comparison result, the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation are corrected.

[0048] For example, the time-varying parameter includes the rate of decay of the thermal conductivity of the solid waste-based thermal storage device over time and temperature.

[0049] In order to achieve dynamic fidelity and model self-calibration of the digital twin, the edge computing device compares and analyzes the actual operation feedback data collected with the simulation data generated by the digital twin in real time. Based on the comparison results obtained from the analysis, the device periodically corrects the time-varying parameters (such as the decay rate of the thermal conductivity of the phase change material) related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation. This ensures that the simulation model can continuously and accurately map the real state of the physical system and maintain the effectiveness of closed-loop control.

[0050] This application embodiment achieves a comprehensive and high-fidelity digital mapping of the combined operating status of thermal power units and solid waste-based thermal storage devices by constructing a joint simulation model of thermal power storage and power generation and generating a digital twin. This enables prior simulation and optimization analysis of different operating conditions in virtual space, providing an accurate prediction and planning basis for actual control decisions, and significantly improving the predictability and scientific nature of control. This application embodiment ensures the real-time acquisition and reliable transmission of key operating status data by deploying field sensors and using time-sensitive network channels to transmit data to edge computing devices, providing a stable and low-latency data foundation for subsequent millisecond-level real-time decisions, effectively overcoming the constraints of latency and jitter in traditional data transmission on rapid control. This application embodiment generates initial commands by executing proportional-integral-derivative control algorithms on edge computing devices, sinking core control functions to the field, reducing dependence on upper-level systems and communication latency, and realizing localized rapid basic adjustment of system dynamic changes, improving the directness and response speed of control. This application embodiment performs multi-objective dynamic optimization and conflict resolution of initial commands based on the optimization objectives of the digital twin, ensuring that control commands no longer satisfy only a single local objective. By comprehensively optimizing multiple objectives such as energy utilization and system safety, and proactively resolving potential conflicts, this application achieves a leap from basic stability control to system-level collaborative optimization control, enhancing the overall economy and operational safety of the system. This embodiment of the application constructs a real-time control closed loop of perception, decision-making, execution, and feedback by issuing optimization commands to the actuators and collecting feedback from actual operating parameters. This closed loop enables continuous adjustment of control based on execution results, improving adaptability to external disturbances and changes in internal parameters, as well as control accuracy. This embodiment of the application achieves dynamic self-calibration of the digital twin by comparing feedback data with digital twin simulation data and correcting the model's time-varying parameters. This ensures that the simulation model can track long-term changes in the physical entity's characteristics, maintaining the long-term fidelity of the digital twin, thereby guaranteeing the continuous effectiveness of model-based optimization and control strategies.

[0051] One specific embodiment of this application provides an implementation scheme for a millisecond-level coordinated control method between a solid waste-based thermal storage device and a thermal power unit. (See attached document.) Figure 2As shown, the system is mainly based on three core components: millisecond-level hardware interaction function, conflict pre-resolution mechanism, and multi-objective dynamic arbitration.

[0052] In implementing millisecond-level hardware interaction, real-time operating systems such as VxWorks, FreeRTOS, Mbed-OS, and QNX are deployed on edge computing nodes to reduce system redundancy and complexity, and improve real-time performance and reliability. To address the high-speed interaction requirements between real-time control commands (e.g., changing inlet steam temperature and adjusting turbine load) and the thermal stress state and remaining lifespan of the thermal storage device, command and status information are transmitted via fieldbus and high-speed network. Centralized parsing, forwarding, logical integration, and priority arbitration are performed on the edge computing device. Simultaneously, field-programmable gate arrays (FPGAs) are used to accelerate control commands, for example, by using digital signal processing (DSP) to perform proportional-integral-differential (PI) calculations, thus shortening command parsing and execution time. To reduce the performance requirements of high-arithmetic-intensity calculations and ensure data security and integrity, a partial result upload-based computation method is adopted. Vector addition is performed on the FPGA within the edge computing device, while multiplication is performed on the host CPU. Through these methods, the control cycle can be reduced from the industrial average of 20ms to 5ms while maintaining control accuracy.

[0053] In the process of implementing the conflict pre-de-escalation mechanism, refer to the appendix. Figure 3As shown, the conflict pre-resolution mechanism first requires the establishment of a joint simulation model of the solid waste-based thermal storage device and the thermal power unit. This model is a simplified high-order system model, including six key state variables: main steam pressure, high-temperature superheated steam temperature, low-temperature superheated steam temperature, high-pressure condenser pressure, low-pressure condenser pressure, and feedwater flow rate. The method to obtain the initial state under stable operating conditions is to allow the system to run normally until all relevant parameters tend to stabilize, and then collect data as the initial state. Based on this, a coupling model with a period of 1 millisecond is established between the unit and the thermal storage device using a discrete time series modeling method. Conflict detection is marked by a drop in the unit's main steam pressure exceeding 5% or the thermal storage device triggering a heat overflow alarm. For example, when the unit reduces load at a high flow rate, if the thermal storage device is still in an energy storage state and cannot release heat in time, it will cause a sudden drop in main steam pressure, and at the same time, the temperature inside the thermal storage device will rise sharply and trigger an alarm. The specific method for conflict pre-resolution is as follows: when the above-mentioned conflict is detected, the current control command is not executed immediately, but a series of candidate compensation strategies are generated and sorted, and then the sorted strategies are executed one by one until the conflict is eliminated. When generating candidate strategies, for various typical conflict scenarios that occur under the dual-mode operation of sequential control and heat storage compensation (such as the contradiction between load withdrawal and energy storage heat release, sudden load increase, etc.), strategies such as immediate heat release, normal energy storage, delayed energy storage, maintaining the original state, starting tracking, and no operation are combined and sorted, and finally the optimal solution set is obtained through Pareto front screening.

[0054] In the implementation of multi-objective dynamic arbitration, this arbitration is undertaken by edge computing devices deployed at the production site. These devices employ computing units such as field-programmable gate arrays (FPGAs) and graphics processing units (GPUs) to transfer some computing tasks originally handled by cloud servers to the edge, thereby reducing the computing burden on the central platform and improving the overall system efficiency. Preferably, the method of this application can be deployed in an existing distributed control system for thermal power units as an independent process, or its core functions can be embedded into relevant functional modules of the distributed control system for integration.

[0055] See attached document Figure 4 The figure shows a comparison of response times under different load changes in the embodiments of this application. As can be seen from the figure, the response time of the traditional control system increases significantly with the increase of the percentage of load change. However, the solution of this application benefits from edge computing devices, field-programmable gate array hardware acceleration, and millisecond-level collaborative control mechanisms, resulting in a very gradual increase in response time, which remains at a low level overall. This demonstrates that the embodiments of this application have excellent real-time response capabilities under dynamic operating conditions.

[0056] Table 1: Performance Indicator Comparison Table

[0057] As shown in Table 1 above, the embodiment of this application shortens the control cycle from 20ms to 5ms, reduces the response latency from 50-100ms to 10-20ms, and increases energy utilization from 82-85% to 88-92%. Furthermore, in terms of conflict resolution, an active pre-resolution mechanism replaces the traditional passive response; and in terms of stability assurance, rigorous verification based on the Lyapunov method is introduced. This comparison fully demonstrates the comprehensive improvement of the embodiment of this application in terms of control real-time performance, operational economy, and system stability.

[0058] See attached document Figure 5 As shown, this application also discloses a millisecond-level collaborative control system for a solid waste-based thermal storage device and a thermal power unit, including: a digital twin generation module 510, a status data acquisition module 520, an initial instruction generation module 530, an optimized instruction generation module 540, an execution and feedback module 550, and a parameter correction module 560.

[0059] The digital twin generation module 510 is used to generate a digital twin of the solid waste-based thermal storage device and the thermal power unit by constructing a joint simulation model of thermal storage and power generation. The joint simulation model of thermal storage and power generation is used to simulate the energy and material flow process of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on the received real-time data and output optimization targets.

[0060] The status data acquisition module 520 is used to acquire the real-time operating status data of the solid waste-based thermal storage device and the thermal power unit through sensors and control equipment deployed on the production site, and transmit the operating status data to the edge computing device through a time-sensitive network channel.

[0061] The initial instruction generation module 530 is used by the edge computing device to generate initial real-time control instructions based on the received operating status data by executing a proportional-integral-derivative control algorithm.

[0062] The optimization instruction generation module 540 is used to perform multi-objective dynamic optimization and conflict resolution on the initial real-time control instruction based on the optimization target output by the digital twin through the edge computing device, and generate optimized control instructions.

[0063] The execution and feedback module 550 is used to send the optimized control command to the actuators of the solid waste-based thermal storage device and the thermal power unit respectively, and to collect the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit as feedback data.

[0064] The parameter correction module 560 is used to compare the feedback data with the simulation data of the digital twin by the edge computing device and generate a comparison result, and correct the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation based on the comparison result.

[0065] In some implementations, the edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit. Based on sensors deployed at the production site, control devices deployed at the production site, actuators of the solid waste-based thermal storage device, actuators of the thermal power unit, and the edge computing device, a closed-loop control module is formed to perform millisecond-level responses to the solid waste-based thermal storage device and the thermal power unit.

[0066] The embodiments of this application realize the efficient, safe, and stable collaborative operation of the solid waste-based thermal storage system and the thermal power unit through the above embodiments.

[0067] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of this application, and this application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this application, and these modifications and improvements are also considered to represent the scope of protection of this application.

Claims

1. A millisecond-level collaborative control method for a solid waste-based thermal storage device and a thermal power unit, characterized in that, include: By constructing a joint simulation model of thermal storage and power generation, a digital twin of solid waste-based thermal storage and thermal power unit is generated. The joint simulation model of thermal storage and power generation is used to simulate the energy and material flow process of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on the received real-time data and output optimization targets. By deploying sensors and control equipment at the production site, the operating status data of the solid waste-based thermal storage device and the thermal power unit are acquired in real time, and the operating status data is transmitted to the edge computing device through a time-sensitive network channel. The edge computing device generates initial real-time control commands based on the received operating status data by executing a proportional-integral-derivative control algorithm. Using the edge computing device, based on the optimization objectives output by the digital twin, the initial real-time control commands are dynamically optimized and conflict resolved through multi-objective processes, and optimized control commands are generated. The optimized control commands are sent to the actuators of the solid waste-based thermal storage device and the thermal power unit, respectively, and the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit are collected as feedback data. The edge computing device compares the feedback data with the simulation data of the digital twin and generates a comparison result. Based on the comparison result, the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation are corrected.

2. The method according to claim 1, characterized in that, By constructing a joint simulation model of thermal storage and power generation, a digital twin of solid waste-based thermal storage and thermal power units is generated, including: A joint simulation model for thermal energy storage and power generation was constructed based on the energy storage characteristic parameters of solid waste-based phase change materials. Based on the aforementioned combined thermal storage and power generation simulation model, the process of thermal storage and power generation working together under different operating conditions is simulated and the steady-state operating point is determined. A digital twin of the solid waste-based thermal storage and the thermal power unit is generated based on the steady-state operating point; The energy storage characteristic parameters include energy storage density, charge / discharge rate, and temperature resistance level.

3. The method according to claim 2, characterized in that, The joint simulation model for thermal energy storage and power generation, constructed based on the energy storage characteristic parameters of solid waste-based phase change materials, includes: The solid waste-based phase change material is prepared using blast furnace steel slag, converter steel slag, carbide slag, phosphogypsum, or tailings filter residue as raw materials. Modified magnesium bricks were selected as high-temperature pressure vessels, and a high-temperature pressure solid waste melting and heat storage device was made based on the solid waste-based phase change material and modified magnesium bricks. A joint simulation model of thermal storage and power generation is constructed based on the physical characteristics of the high-temperature pressure-bearing solid waste melting thermal storage device. The aforementioned joint simulation model of thermal storage and power generation is integrated into the industrial internet platform.

4. The method according to claim 1, wherein the edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit, used for communication and collaborative decision-making between the solid waste-based thermal storage device and the thermal power unit, characterized in that, The method further includes: Based on sensors deployed at the production site, control equipment deployed at the production site, actuators of the solid waste-based thermal storage device, actuators of the thermal power unit, and edge computing equipment, closed-loop control with millisecond-level response is performed on the solid waste-based thermal storage device and the thermal power unit.

5. The method according to claim 4, wherein the edge computing device integrates a field-programmable gate array and a graphics processor for accelerating the execution of control algorithms through hardware parallel computing, characterized in that, The method further includes: Based on the received operating status data, vector addition is performed by the field-programmable gate array, and complex function floating-point operations are performed by the graphics processing unit.

6. The method according to claim 1, characterized in that, The method further includes: With the goal of maximizing energy supply and with total energy storage and energy storage density as constraints, a multi-objective evolutionary model for non-dominated ranking is constructed for multi-objective dynamic optimization and conflict resolution in litigation. The multi-objective evolutionary model is implemented based on the Pareto optimal solution set selection method, and the Lagrange multiplier method is used to construct dummy variables to replace the original control variables.

7. The method according to claim 1, characterized in that, The actual operating parameters of the solid waste-based thermal storage device and the actuator of the thermal power unit are collected every fifty milliseconds. The time-varying parameters include the decay rate of the thermal conductivity of the solid waste-based thermal storage device with time and temperature.

8. The method according to claim 1, characterized in that, The multi-objective dynamic optimization and conflict resolution of the initial real-time control command includes: When the main steam pressure drops below a preset threshold in the initial real-time control command or the solid waste-based thermal storage device triggers a heat overflow alarm, it is determined that a conflict has occurred. Suspend the execution of the current control command and generate a sequence of candidate compensation strategies, including immediate heat release, normal energy storage, delayed energy storage, maintaining the original state, starting tracking, and no operation. The Pareto optimization method is used to sort the candidate compensation strategy sequence to obtain the optimal solution set; The conflict is resolved based on the optimal solution set until the conflict is eliminated.

9. A millisecond-level collaborative control system for a solid waste-based thermal storage device and a thermal power unit, characterized in that, include: The digital twin generation module is used to generate a digital twin of solid waste-based thermal storage and a thermal power unit by constructing a joint simulation model of thermal storage and power generation. The joint simulation model of thermal storage and power generation is used to simulate the energy and material flow process of the thermal power unit and the solid waste-based thermal storage device under different operating conditions. The digital twin is used to generate simulation data based on the received real-time data and output optimization targets. The status data acquisition module is used to acquire the real-time operating status data of the solid waste-based thermal storage device and the thermal power unit through sensors and control equipment deployed on the production site, and transmit the operating status data to the edge computing device through a time-sensitive network channel; The initial instruction generation module is used by the edge computing device to generate initial real-time control instructions based on the received operating status data by executing a proportional-integral-derivative control algorithm. An optimization instruction generation module is used to perform multi-objective dynamic optimization and conflict resolution on the initial real-time control instruction based on the optimization objectives output by the digital twin through the edge computing device, and generate optimized control instructions. The execution and feedback module is used to send the optimized control commands to the actuators of the solid waste-based thermal storage device and the thermal power unit, respectively, and to collect the actual operating parameters of the actuators of the solid waste-based thermal storage device and the thermal power unit as feedback data. The parameter correction module is used to compare the feedback data with the simulation data of the digital twin by the edge computing device and generate a comparison result, and correct the time-varying parameters related to the performance of the solid waste-based thermal storage device in the joint simulation model of thermal storage and power generation based on the comparison result.

10. The millisecond-level collaborative control system for a solid waste-based thermal storage device and a thermal power unit according to claim 9, characterized in that, The edge computing device is a third-party computing device embedded between the solid waste-based thermal storage device and the thermal power unit. Based on sensors deployed at the production site, control equipment deployed at the production site, the actuators of the solid waste-based thermal storage device, the actuators of the thermal power unit, and the edge computing device, a closed-loop control module is constructed to perform millisecond-level responses to the solid waste-based thermal storage device and the thermal power unit.

Images