Method and device for controlling peak regulation of thermal power generating unit, electronic equipment and storage medium

Abstract

The invention discloses a thermal power generating unit peak regulation control method and device, electronic equipment and a storage medium, and can integrate historical data, real-time parameters and external environment variables of a thermal power generating unit peak regulation operation interval, and construct a unit peak regulation state multi-dimensional representation space through time alignment and feature extraction. A dynamic response model reflecting actual operation characteristics of equipment is established, pre-compensation and dynamic correction can be performed on a peak regulation instruction, nonlinear time-varying characteristics of a boiler, a steam turbine and an auxiliary engine link are fully considered, and limitation of a layered decoupling control framework is avoided, so that the problem that existing layered decoupling control neglects the nonlinear time-varying characteristics of the equipment, and the nonlinear time-varying characteristics of the boiler, the steam turbine and the auxiliary engine link are not fully considered can be solved. According to the method, the technical problems of model prediction precision degradation, reheater temperature instability and other cascading failures caused by thermal power generating unit peak regulation are solved, and the technical effects that the thermal power generating unit peak regulation model prediction precision is improved, cascading failures in the peak regulation process are prevented, the unit peak regulation operation stability is guaranteed, and the regulation capacity of a thermal power generating unit in a new energy high-permeability power grid is enhanced are achieved.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a control method and apparatus, electronic equipment and storage medium for peak shaving of thermal power units. Background Technology

[0002] Thermal power, as an important regulating resource in the power system, is widely used in frequency stabilization and load balancing of power grids with high penetration of new energy sources. Among related technologies, a technical system for the flexible retrofitting of thermal power units has been constructed through the coordinated operation of boiler thermal storage utilization, turbine bypass heating, and wide load adaptation of the combustion system.

[0003] Existing peak-shaving control methods directly adopt a hierarchical decoupled control architecture, ignoring the nonlinear time-varying characteristics of the boiler, turbine, and auxiliary equipment links. This causes the model prediction accuracy to deteriorate sharply with the deviation of operating conditions, resulting in a chain of failures such as excessive main steam pressure and instability of reheater temperature. Summary of the Invention

[0004] This disclosure provides a control method, device, electronic equipment, and storage medium for peak shaving in thermal power units. Its main purpose is to solve the problem of cascading failures caused by the rapid deterioration of model prediction accuracy with deviations in operating conditions, leading to excessive main steam pressure and reheater temperature instability.

[0005] According to a first aspect of this disclosure, a control method for peak shaving in thermal power units is provided, comprising: Acquire historical operating data, real-time operating parameters, and external environmental variables of the target thermal power unit within the peak-shaving operating range; Time alignment and feature extraction are performed on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state; A dynamic response model of the peak-shaving capacity of thermal power units is constructed based on the multi-dimensional characterization space of the unit's peak-shaving state. Pre-compensation and dynamic correction are performed on the peak shaving command to generate the final control command, and the thermal power unit is controlled to perform peak shaving operation based on the final control command.

[0006] Optionally, the step of performing time alignment and feature extraction on the historical operating data, real-time operating parameters, and external environmental variables to construct a multi-dimensional representation space of the unit's peak-shaving state includes: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

[0007] Optionally, the method further includes: Piecewise linearization modeling is adopted, and load change rate clustering based on historical operating data is used to divide the rated load range into a preset number of sub-ranges.

[0008] Optionally, the method further includes: The model predictive control framework is used to predict the unit output trajectory within a preset time period; where the standard function is to minimize the sum of squared deviations between the predicted trajectory and the dispatching instructions. Calculate the feedforward compensation amount and add it to the original scheduling instruction to form the preliminary control instruction; Set threshold values ​​for boiler water-cooled wall metal temperature gradient, turbine last-stage blade vibration amplitude, generator stator winding temperature rise rate, and main transformer oil temperature change rate. When any threshold index falls below the safety value, the optimal output trajectory is recalculated based on the sequential quadratic programming algorithm; wherein, the sequential quadratic programming algorithm minimizes the deviation between the output trajectory and the pre-compensation command, and the constraint condition is that the margin index of each device is not lower than the safety threshold. Generate a revised sequence of control instructions, with the instruction update cycle occurring every two seconds.

[0009] Optionally, the method further includes: The equipment health status monitoring subsystem collects boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signal. After extracting time-domain statistical features and frequency-domain energy distribution features, it inputs them into a support vector machine classifier to output the operating status. The operating parameter over-limit early warning subsystem sets alarm thresholds of different levels; wherein, the alarm thresholds include design allowable limits, recommended safe operation values, and optimal economic operation values; The system automatically triggers a rapid load reduction procedure when the device health status is detected as faulty or the parameters exceed the design limits.

[0010] According to a second aspect of this disclosure, a control device for peak shaving in thermal power units is provided, comprising: The acquisition unit is used to acquire historical operating data, real-time operating parameters, and external environmental variables of the target thermal power unit within the peak-shaving operating range. The construction unit is used to perform time alignment and feature extraction on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state. The construction unit is also used to construct a dynamic response model of the peak-shaving capacity of the thermal power unit based on the multi-dimensional characterization space of the unit's peak-shaving state; The generation unit is also used to perform pre-compensation and dynamic correction on the peak shaving command to generate the final control command, and control the thermal power unit to perform peak shaving operation based on the final control command.

[0011] Optionally, the building unit is further configured to: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

[0012] Optionally, the device further includes: The unit division is also used to divide the rated load range into a preset number of sub-ranges by using piecewise linearization modeling based on the load change rate clustering of historical operating data.

[0013] Optionally, the device further includes: The prediction unit is used to predict the unit's output trajectory within a preset time period based on the model predictive control framework; where the standard function is to minimize the sum of squares of the deviations between the predicted trajectory and the dispatching instructions. The calculation unit is used to calculate the feedforward compensation amount and superimpose the feedforward compensation amount onto the original scheduling instruction to form the preliminary control instruction; The setting unit is used to set the threshold values ​​for the boiler water-cooled wall metal temperature gradient, the turbine last stage blade vibration amplitude, the generator stator winding temperature rise rate, and the main transformer oil temperature change rate. The calculation unit is also used to recalculate the optimal output trajectory based on a sequential quadratic programming algorithm when any threshold index is lower than the safety value; wherein, the sequential quadratic programming algorithm is to minimize the deviation between the output trajectory and the pre-compensation command, and the constraint condition is that the margin index of each device is not lower than the safety threshold. The update unit is used to generate the corrected control instruction sequence, with an instruction update cycle of once every two seconds.

[0014] Optionally, the device further includes: The data acquisition unit is used by the equipment health status monitoring subsystem to acquire boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signal. After extracting time-domain statistical features and frequency-domain energy distribution features, it inputs them into the support vector machine classifier to output the operating status. The setting unit is also used to set different levels of alarm thresholds for the operating parameter over-limit early warning subsystem; wherein, the alarm thresholds include design allowable limit values, safe operation recommended values, and economic operation optimal values; The control unit is used to automatically trigger a rapid load reduction program when the equipment health status is detected to be faulty or the parameters exceed the design allowable limits.

[0015] According to a third aspect of this disclosure, an electronic device is provided, comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in the first aspect above.

[0016] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method described in the first aspect above.

[0017] According to a fifth aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0018] The control method, device, electronic equipment, and storage medium for peak shaving of thermal power units disclosed herein can integrate historical data, real-time parameters, and external environmental variables of the peak shaving operation range of thermal power units. Through time alignment and feature extraction, a multi-dimensional characterization space of the unit's peak shaving state is constructed, thereby establishing a dynamic response model that reflects the actual operating characteristics of the equipment. It can also perform pre-compensation and dynamic correction on peak shaving commands, fully considering the nonlinear time-varying characteristics of the boiler, turbine, and auxiliary equipment links, and avoiding the limitations of the hierarchical decoupled control architecture. Therefore, it can solve the technical problem that existing hierarchical decoupled control ignores the nonlinear time-varying characteristics of equipment, leading to deterioration of model prediction accuracy, excessive main steam pressure, and cascading failures such as reheater temperature instability. It achieves the technical effects of improving the prediction accuracy of the peak shaving model of thermal power units, preventing cascading failures during peak shaving, ensuring the stability of the unit's peak shaving operation, and enhancing the regulation capability of thermal power units in grids with high penetration of new energy sources.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0020] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein: Figure 1 This is a flowchart illustrating a control method for peak shaving of a thermal power unit provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram of the structure of a control device for peak shaving of a thermal power unit provided in an embodiment of the present disclosure; Figure 3 A schematic diagram of the structure of another control device for peak shaving of thermal power units provided in an embodiment of this disclosure; Figure 4 A schematic block diagram of an example electronic device provided for embodiments of this disclosure. Detailed Implementation

[0021] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0022] The following description, with reference to the accompanying drawings, outlines a control method, apparatus, electronic device, and storage medium for peak shaving of thermal power units according to embodiments of the present disclosure.

[0023] Figure 1 This is a flowchart illustrating a control method for peak shaving of thermal power units provided in an embodiment of this disclosure.

[0024] like Figure 1 As shown, the method includes the following steps: Step 101: Obtain historical operating data, real-time operating parameters, and external environmental variables of the target thermal power unit within the peak-shaving operating range; Historical operating data is a long-term record of the unit's operating trajectory during past participation in peak shaving. It encompasses core information reflecting the unit's load regulation history, dynamic changes in the thermal system, and the operational status of mechanical components. This data can be extracted from the unit's historical database, providing a reference for analyzing the unit's operating patterns, characteristics, trends, and potential problems under different peak shaving conditions. Real-time operating parameters focus on the current operating status of the unit's core equipment and systems, covering key indicators of the boiler, turbine, generator, and auxiliary systems. These parameters need to be collected in real time by high-precision sensors deployed in key parts of the unit to ensure accurate control of the unit's immediate operating status and provide real-time basis for subsequent dynamic adjustment of control strategies.

[0025] External environmental variables include external factors affecting the peak-shaving operation of the generating units. These factors include grid-level dispatching requirements and overall operational status, as well as natural environmental conditions. These variables need to be acquired through data exchange with the grid dispatching center and data collection from monitoring equipment around the generating units to ensure that subsequent models can fully adapt to the actual external conditions of the unit's operation. The collection of these three types of data must ensure comprehensive coverage and continuous time dimension to avoid data gaps or biases, laying a reliable foundation for subsequent data processing.

[0026] Step 102: Time alignment and feature extraction are performed on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state; The time alignment process addresses potential differences in sampling frequencies among historical operating data, real-time operating parameters, and external environmental variables. It employs a unified time synchronization mechanism to standardize data from different sources and frequencies onto the same time base, ensuring that all types of data correspond one-to-one at the time node. This avoids data correlation analysis distortion caused by time misalignment and provides a time dimension guarantee for the accuracy of subsequent feature extraction.

[0027] The feature extraction stage, based on the operating mechanisms and peak-shaving requirements of each system in the unit, filters and extracts information that characterizes the key peak-shaving characteristics of the unit from three types of data. For the different operating characteristics of systems such as boiler, turbine, and electrical systems, features related to the peak-shaving performance of each system are extracted. These features must accurately reflect the dynamic changes and core operating states of each system during the peak-shaving process, avoiding redundant information interference. After time alignment and feature extraction, all extracted feature vectors are processed uniformly and integrated to form a multi-dimensional representation space of the unit's peak-shaving state. This space can comprehensively and systematically depict the overall operating state of the unit during the peak-shaving process from multiple dimensions, encompassing both the independent operating characteristics of each system and the inter-system correlation characteristics. This provides a structured and comprehensive input foundation for the subsequent construction of a dynamic response model for the peak-shaving capacity of thermal power units, ensuring that the model can fully rely on the actual operating characteristics of the unit for subsequent analysis and optimization.

[0028] Step 103: Construct a dynamic response model of the peak-shaving capacity of the thermal power unit based on the multi-dimensional characterization space of the unit's peak-shaving state; In the modeling process, it is necessary to first combine the actual characteristics of the unit's peak-shaving operation, and take into account the possible nonlinear differences in the thermal system, mechanical response and electrical characteristics of the unit in different load ranges. Based on the characteristics of each dimension covered in the multidimensional representation space, the model is reasonably divided according to the load change range to ensure that the unit characteristics are relatively stable in each range, thus laying the foundation for the accuracy of the model.

[0029] For each divided interval, an adapted dynamic model structure is constructed. The model needs to include a state vector that can characterize the internal operating state of the unit, a control vector for regulating the operation of the unit, and a disturbance vector that may affect the unit response. The state vector needs to be associated with the core features in the multi-dimensional representation space, the control vector corresponds to the peak shaving control measures that may be executed later, and the disturbance vector covers various potential interference factors.

[0030] To ensure the model accurately reflects actual unit operation, online parameter identification methods are needed to continuously refine model parameters using historical and real-time data in a multi-dimensional representation space. This eliminates discrepancies between theoretical assumptions and actual operation, ensuring the model accurately predicts the unit's response trends under different peak-shaving commands. The final dynamic response model for thermal power unit peak-shaving capacity must output the unit's expected response to peak-shaving commands based on the current state reflected in the multi-dimensional representation space. This provides reliable model support for pre-compensation and dynamic correction of peak-shaving commands in subsequent steps, achieving a crucial link from state representation to control decision-making.

[0031] Step 104: Perform pre-compensation and dynamic correction on the peak shaving command to generate the final control command, and control the thermal power unit to perform peak shaving operation based on the final control command.

[0032] The pre-compensation stage, based on the previously constructed dynamic response model of the peak-shaving capacity of thermal power units, predicts the unit's output trajectory over a future period. It compares the deviation of this trajectory with the original peak-shaving command, calculates the corresponding feedforward compensation, and adds it to the original command to form a preliminary control command. This proactive approach addresses potential response lag issues and reduces deviations after command execution, laying the foundation for peak-shaving accuracy. The dynamic correction stage focuses on unit operational safety, monitoring the operational margin indicators of key equipment in real time. It determines whether these indicators meet safe operation requirements. If any margin indicator approaches or falls below the safety boundary, a constraint optimization algorithm is immediately initiated to recalculate the optimal output trajectory. While ensuring that the margin indicators of each piece of equipment meet safety requirements, the new trajectory is made as close as possible to the pre-compensated preliminary command, avoiding operational risks caused by equipment exceeding limits.

[0033] After pre-compensation and dynamic correction are completed, the final control command is formed. This command must balance peak-shaving accuracy and equipment safety, ensuring that the command parameters are compatible with the unit's current operating status, external environment, and equipment tolerance. Finally, the final control command is sent to each control subsystem of the unit, driving the core equipment such as the boiler, turbine, and generator to coordinate their actions and execute specific peak-shaving operations, so that the actual output of the unit is adjusted according to the command requirements to achieve the peak-shaving target.

[0034] In some embodiments, the step of performing time alignment and feature extraction on the historical operating data, real-time operating parameters, and external environmental variables to construct a multi-dimensional characterization space for the unit's peak-shaving state includes: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

[0035] For the boiler side, the combustion stability index is calculated by weighting the amplitude of furnace negative pressure fluctuation and the deviation of oxygen setpoint. The amplitude of furnace negative pressure fluctuation directly reflects the stability of combustion conditions inside the furnace, while the deviation of oxygen setpoint is related to the rationality of fuel-air ratio. The weighted calculation of the two can comprehensively quantify the stability of the combustion system during peak shaving, avoiding the impact of combustion fluctuations on peak shaving accuracy. The evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The steam drum water level is a key indicator of the boiler evaporation system's operating status, and the feedwater flow rate directly determines the evaporation supply. This ratio can accurately reflect the dynamic response efficiency of the boiler evaporation system during peak shaving, providing a basis for judging whether the evaporation is suitable for load adjustment requirements.

[0036] For the turbine side, the valve action delay coefficient is calculated based on the time interval between the issuance of the control valve command and the actual change in opening. The turbine control valve opening directly controls the steam intake, thus affecting the unit's output. This time interval directly reflects the lag in the control valve's mechanical response and is a core characteristic for evaluating the turbine's output adjustment speed during peak shaving. For the electrical side, the reactive power support capacity attenuation rate is calculated as the ratio of the generator terminal voltage change to the reactive power output change. Generator terminal voltage stability is a crucial guarantee for the safe operation of the power grid, and changes in reactive power output directly affect the terminal voltage level. This ratio effectively reflects the unit's ability to maintain terminal voltage stability and provide reactive power support during peak load fluctuations. Integrating these extracted features into the overall feature set constitutes a multi-dimensional representation space of the unit's peak shaving state, providing accurate feature inputs for subsequent dynamic response model construction.

[0037] In some embodiments, the method further includes: Piecewise linearization modeling is adopted, and load change rate clustering based on historical operating data is used to divide the rated load range into a preset number of sub-ranges.

[0038] In the process of constructing a dynamic response model for the peak-shaving capacity of thermal power units, in order to adapt to the nonlinear operating characteristics of the units in different load ranges and improve the model's accuracy in depicting the peak-shaving response law of the units, it is necessary to adopt a piecewise linearization modeling method and divide the rated load range based on the load change rate clustering results of historical operating data.

[0039] Load change rate clustering uses the load adjustment rate of past peak-shaving processes of units in historical operating data as the core analysis object. By mining the similarity and difference of load change rates under different load levels, it groups load ranges with similar load change patterns and convergent unit response characteristics into one category, avoiding accuracy deviations caused by directly modeling the entire rated load range. The partitioning operation takes the rated load range as the whole scope and, based on the grouping results obtained from cluster analysis, divides it into a preset number of sub-intervals. Within each sub-interval, the dynamic characteristics of the unit's thermal system, the response speed of mechanical components, and the power regulation patterns on the electrical side are relatively stable, effectively reducing the complexity of adapting a single model to the entire load range.

[0040] This segmentation method based on historical load change rate clustering can fully fit the actual operating patterns of the units, ensuring that the division of each sub-interval has data support and physical meaning. This lays the foundation for building accurate dynamic response models in each sub-interval, enabling the models to more accurately capture the response characteristics of the units to peak-shaving commands under different load intervals, thereby improving the reliability of subsequent peak-shaving command optimization and control.

[0041] In some embodiments, the method further includes: The model predictive control framework is used to predict the unit output trajectory within a preset time period; where the standard function is to minimize the sum of squared deviations between the predicted trajectory and the dispatching instructions. Calculate the feedforward compensation amount and add it to the original scheduling instruction to form the preliminary control instruction; Set threshold values ​​for boiler water-cooled wall metal temperature gradient, turbine last-stage blade vibration amplitude, generator stator winding temperature rise rate, and main transformer oil temperature change rate. When any threshold index is lower than the safety value, the optimal output trajectory is recalculated based on the sequential quadratic programming algorithm; wherein, the sequential quadratic programming algorithm is to minimize the deviation between the output trajectory and the pre-compensation command, and the constraint condition is that the margin index of each device is not lower than the safety threshold. Generate a revised sequence of control instructions, with the instruction update cycle occurring every two seconds.

[0042] In the process of pre-compensation and dynamic correction of peak-shaving commands, the unit output trajectory is first predicted based on the model predictive control framework. The prediction range focuses on the trend of unit output change within a preset time period. This process uses a sequential quadratic programming algorithm to minimize the sum of squared deviations between the predicted trajectory and the dispatch command. This algorithm ensures that the predicted output trajectory closely matches the dispatch command requirements, reducing initial prediction deviations and providing a precise basis for subsequent compensation operations. Then, the feedforward compensation amount is calculated. Combining the unit's past operating patterns with the current prediction deviation, a compensation value that can offset potential response lags in advance is determined. This feedforward compensation amount is then added to the original dispatch command to form a preliminary control command. This step can proactively avoid command execution deviations that may occur due to dynamic response delays, laying the foundation for peak-shaving accuracy.

[0043] To ensure the safe operation of core equipment, specific safety thresholds need to be set for critical equipment, including the boiler water-cooled wall metal temperature gradient threshold, the turbine last-stage blade vibration amplitude threshold, the generator stator winding temperature rise rate threshold, and the main transformer oil temperature change rate threshold. These thresholds are determined based on equipment design standards and long-term safe operation experience, and are directly related to the equipment's service life and safety boundaries. When any threshold is detected to be below the safe value, the optimal output trajectory is immediately recalculated based on a sequential quadratic programming algorithm. The algorithm minimizes the deviation between the output trajectory and the pre-compensation command, ensuring the adjusted trajectory is as close as possible to the initial control direction. Simultaneously, the margin indicators of each piece of equipment are kept above the safety threshold as a constraint to prevent equipment failures due to parameter exceeding limits. Finally, a corrected control command sequence is generated based on the recalculation results. To ensure that the commands can adapt to changes in the unit's operating status in real time, the command update cycle is set to once every two seconds, ensuring the real-time performance and safety of the control commands and achieving coordinated optimization of peak-shaving commands and equipment safety.

[0044] In some embodiments, the method further includes: The equipment health status monitoring subsystem collects boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signal. After extracting time-domain statistical features and frequency-domain energy distribution features, it inputs them into a support vector machine classifier to output the operating status. The operating parameter over-limit early warning subsystem sets alarm thresholds of different levels; wherein, the alarm thresholds include design allowable limits, recommended safe operation values, and optimal economic operation values; The system automatically triggers a rapid load reduction procedure when the device health status is detected as faulty or the parameters exceed the design limits.

[0045] During peak-shaving operation of thermal power units, a multi-level safety protection mechanism is needed to ensure equipment safety and system stability. This mechanism includes three core components: equipment health status monitoring, early warning of exceeding operating parameter limits, and emergency load control. The equipment health status monitoring subsystem focuses on the health status of key equipment in the unit, collecting real-time data on boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signals. The boiler heating surface wall temperature distribution directly reflects whether there is a risk of local overheating. The turbine bearing vibration spectrum is related to rotor running stability and mechanical wear. The generator partial discharge signal is an important basis for judging insulation performance and internal faults.

[0046] The acquired signals need to undergo feature extraction processing to extract time-domain statistical features and frequency-domain energy distribution features that can characterize the equipment status. Time-domain statistical features can reflect the overall change pattern of the signal in the time dimension, while frequency-domain energy distribution features can capture abnormal fluctuations in specific frequency bands of the signal. After these features are input into the support vector machine classifier, the classifier can output three operating states of the equipment based on the preset model: normal, warning, and fault, so as to achieve accurate judgment of the health status of the equipment.

[0047] The operating parameter limit exceedance early warning subsystem sets different levels of alarm thresholds for key parameters during unit operation, namely the design allowable limit, the recommended safe operation value, and the optimal economic operation value. The design allowable limit is the safety baseline for equipment operation; exceeding this value will directly cause equipment damage or safety accidents. The recommended safe operation value is the threshold that needs to be monitored in daily operation; exceeding this value means that the parameter deviates from the safe operation range and timely intervention is required. The optimal economic operation value is the ideal parameter range that balances operating efficiency and energy consumption, providing a reference for optimizing operating economy.

[0048] When the equipment health status monitoring subsystem outputs a fault status, or the operating parameter over-limit warning subsystem detects that the parameter exceeds the design allowable limit value, the system will automatically trigger the load reduction program. By quickly reducing the unit load, the load pressure on the equipment is reduced, the fault is prevented from expanding further, the safety of the unit's core equipment is ensured, and peak shaving interruption or more serious operational accidents are prevented due to equipment failure.

[0049] Corresponding to the aforementioned control method for peak shaving of thermal power units, this invention also proposes a control device for peak shaving of thermal power units. Since the device embodiments of this invention correspond to the aforementioned method embodiments, details not disclosed in the device embodiments can be referred to in the aforementioned method embodiments, and will not be repeated here.

[0050] Figure 2 This is a schematic diagram of the structure of a control device for peak shaving of a thermal power unit provided in an embodiment of this disclosure, as shown below. Figure 2 As shown, it includes: Acquisition unit 21 is used to acquire historical operating data, real-time operating parameters and external environmental variables of the target thermal power unit within the peak-shaving operating range; Construction unit 22 is used to perform time alignment and feature extraction on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state; The construction unit 22 is also used to construct a dynamic response model of the peak-shaving capacity of the thermal power unit based on the multi-dimensional characterization space of the unit's peak-shaving state; The generation unit 23 is also used to perform pre-compensation and dynamic correction on the peak shaving command to generate the final control command, and control the thermal power unit to perform peak shaving operation based on the final control command.

[0051] Furthermore, in one possible implementation of this disclosure embodiment, the construction unit 22 is further configured to: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

[0052] Furthermore, in one possible implementation of the embodiments of this disclosure, such as Figure 3 As shown, the device further includes: The division unit 24 is also used to divide the rated load range into a preset number of sub-ranges by using piecewise linearization modeling based on the load change rate clustering of historical operating data.

[0053] Furthermore, in one possible implementation of the embodiments of this disclosure, such as Figure 3 As shown, the device further includes: Prediction unit 25 is used to predict the unit output trajectory within a preset time period based on the model predictive control framework; wherein, the standard function is to minimize the sum of squares of the deviations between the predicted trajectory and the dispatching instructions; The calculation unit 26 is used to calculate the feedforward compensation amount and superimpose the feedforward compensation amount onto the original scheduling instruction to form the preliminary control instruction; Setting unit 27 is used to set the threshold values ​​for the boiler water-cooled wall metal temperature gradient, the threshold values ​​for the vibration amplitude of the turbine last stage blade, the threshold values ​​for the temperature rise rate of the generator stator winding, and the threshold values ​​for the oil temperature change rate of the main transformer. The calculation unit 26 is also used to recalculate the optimal output trajectory based on the sequential quadratic programming algorithm when any threshold index is lower than the safety value; wherein, the sequential quadratic programming algorithm is to minimize the deviation between the output trajectory and the pre-compensation command, and the constraint condition is that the margin index of each device is not lower than the safety threshold. Update unit 28 is used to generate the corrected control instruction sequence, with an instruction update cycle of once every two seconds.

[0054] Furthermore, in one possible implementation of the embodiments of this disclosure, such as Figure 3 As shown, the device further includes: The acquisition unit 29 is used by the equipment health status monitoring subsystem to acquire the boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signal. After extracting the time domain statistical features and frequency domain energy distribution features, it is input into the support vector machine classifier to output the operating status. The setting unit 27 is also used to set different levels of alarm thresholds for the operating parameter over-limit early warning subsystem; wherein, the alarm thresholds include design allowable limit values, safe operation recommended values, and economic operation optimal values; Control unit 210 is used to automatically trigger a rapid load reduction program when the health status of the equipment is detected to be faulty or the parameters exceed the design allowable limits.

[0055] It should be noted that the foregoing explanation of the method embodiments also applies to the apparatus of the embodiments of this disclosure, and the principle is the same. Therefore, the embodiments of this disclosure are not limited thereto.

[0056] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0057] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0058] like Figure 4As shown, device 400 includes a computing unit 401, which can perform various appropriate actions and processes based on a computer program stored in ROM (Read-Only Memory) 402 or a computer program loaded from storage unit 408 into RAM (Random Access Memory) 403. RAM 403 may also store various programs and data required for the operation of device 400. The computing unit 401, ROM 402, and RAM 403 are interconnected via bus 404. I / O (Input / Output) interface 405 is also connected to bus 404.

[0059] Multiple components in device 400 are connected to I / O interface 405, including: input unit 406, such as keyboard, mouse, etc.; output unit 407, such as various types of monitors, speakers, etc.; storage unit 408, such as disk, optical disk, etc.; and communication unit 409, such as network card, modem, wireless transceiver, etc. Communication unit 409 allows device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0060] The computing unit 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, CPUs (Central Processing Units), GPUs (Graphics Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, DSPs (Digital Signal Processors), and any suitable processor, controller, microcontroller, etc. The computing unit 401 performs the various methods and processes described above, such as the control method for peak shaving of thermal power units. For example, in some embodiments, the control method for peak shaving of thermal power units can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on device 400 via ROM 402 and / or communication unit 409. When the computer program is loaded into RAM 403 and executed by the computing unit 401, one or more steps of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the aforementioned control method for peak shaving of thermal power units by any other suitable means (e.g., by means of firmware).

[0061] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System-on-Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0062] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0063] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0064] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0065] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, and blockchain networks.

[0066] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0067] It's important to note that artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily include computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graph technologies.

[0068] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0069] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A control method for peak shaving in thermal power units, characterized in that, include: Acquire historical operating data, real-time operating parameters, and external environmental variables of the target thermal power unit within the peak-shaving operating range; Time alignment and feature extraction are performed on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state; A dynamic response model of the peak-shaving capacity of thermal power units is constructed based on the multi-dimensional characterization space of the unit's peak-shaving state. Pre-compensation and dynamic correction are performed on the peak shaving command to generate the final control command, and the thermal power unit is controlled to perform peak shaving operation based on the final control command.

2. The method according to claim 1, characterized in that, The step of performing time alignment and feature extraction on the historical operating data, real-time operating parameters, and external environmental variables to construct a multi-dimensional characterization space for the unit's peak-shaving state includes: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

3. The method according to claim 1, characterized in that, The method further includes: Piecewise linearization modeling is adopted, and load change rate clustering based on historical operating data is used to divide the rated load range into a preset number of sub-ranges.

4. The method according to claim 1, characterized in that, The method further includes: The model predictive control framework is used to predict the unit output trajectory within a preset time period; where the standard function is to minimize the sum of squared deviations between the predicted trajectory and the dispatching instructions. Calculate the feedforward compensation amount and add it to the original scheduling instruction to form the preliminary control instruction; Set threshold values ​​for boiler water-cooled wall metal temperature gradient, turbine last-stage blade vibration amplitude, generator stator winding temperature rise rate, and main transformer oil temperature change rate. When any threshold index falls below the safety value, the optimal output trajectory is recalculated based on the sequential quadratic programming algorithm; wherein, the sequential quadratic programming algorithm minimizes the deviation between the output trajectory and the pre-compensation command, and the constraint condition is that the margin index of each device is not lower than the safety threshold. Generate a revised sequence of control instructions, with the instruction update cycle occurring every two seconds.

5. The method according to claim 1, characterized in that, The method further includes: The equipment health status monitoring subsystem collects boiler heating surface wall temperature distribution, turbine bearing vibration spectrum, and generator partial discharge signal. After extracting time-domain statistical features and frequency-domain energy distribution features, it inputs them into a support vector machine classifier to output the operating status. The operating parameter over-limit early warning subsystem sets alarm thresholds of different levels; wherein, the alarm thresholds include design allowable limits, recommended safe operation values, and optimal economic operation values; The system automatically triggers a rapid load reduction procedure when the device health status is detected as faulty or the parameters exceed the design limits.

6. A control device for peak shaving in thermal power units, characterized in that, include: The acquisition unit is used to acquire historical operating data, real-time operating parameters, and external environmental variables of the target thermal power unit within the peak-shaving operating range. The construction unit is used to perform time alignment and feature extraction on the historical operating data, real-time operating parameters and external environmental variables to construct a multi-dimensional characterization space of the unit's peak-shaving state. The construction unit is also used to construct a dynamic response model of the peak-shaving capacity of the thermal power unit based on the multi-dimensional characterization space of the unit's peak-shaving state; The generation unit is also used to perform pre-compensation and dynamic correction on the peak shaving command to generate the final control command, and control the thermal power unit to perform peak shaving operation based on the final control command.

7. The apparatus according to claim 6, characterized in that, The building unit is also used for: The combustion stability index is calculated by weighting the fluctuation amplitude of the furnace negative pressure with the deviation of the oxygen setpoint, and the evaporation fluctuation rate is quantified based on the ratio of the change in steam drum water level to the change in feedwater flow rate per unit time. The valve action delay coefficient in the turbine-side characteristics is calculated by the time interval between the issuance of the valve command and the actual change in valve opening. The reactive power support capacity attenuation rate is calculated by the ratio of the change in generator terminal voltage to the change in reactive power output.

8. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-5.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1-5.

Images