Online evaluation method and device for primary frequency modulation performance of thermal power plant unit

Abstract

The invention discloses an online evaluation method and device for the primary frequency modulation performance of a thermal power plant unit, and the method comprises the steps: building a dynamic response model of a thermal power plant unit by considering the coupling effect of the valve adjustment and main steam pressure change of a steam turbine of the thermal power plant unit; equivalent electric quantity generated during energy conversion between the boiler and the steam turbine is calculated, and sensitivity analysis is carried out on key operation parameters, collected in real time, of the thermal power generating unit according to the equivalent electric quantity; and constructing an online evaluation model for evaluating the primary frequency modulation capability of the thermal power generating unit in real time. According to the method, the primary frequency modulation performance of the thermal power plant under different loads can be accurately evaluated, and real-time and accurate decision support is provided for power grid dispatching.

Description

Technical Field

[0001] The present invention relates to the field of power grid frequency fluctuation control, and in particular to an online evaluation method and device for primary frequency regulation performance of thermal power plant units based on dynamic energy conversion process analysis and equivalent power quantification. Background Art

[0002] With the rapid development of renewable energy, the frequency fluctuation control in the power grid has become increasingly complex. As the main component of traditional energy, the primary frequency regulation capability of thermal power has become increasingly important in the process of deep peak regulation. When evaluating the participation of thermal power energy storage systems in primary frequency regulation, existing technologies often lack accurate consideration of dynamic energy conversion details and sufficient quantification of energy storage potential, especially safety evaluation under low-load operating conditions. Summary of the invention

[0003] In order to solve the problems existing in the background technology, the present invention proposes an online evaluation method and device for the primary frequency regulation performance of thermal power plant units. Through in-depth analysis of the dynamic energy conversion process and quantification of equivalent power, the primary frequency regulation performance of the thermal power plant under different loads is accurately evaluated, providing real-time and accurate decision support for power grid dispatching.

[0004] On one hand, the present invention provides an online evaluation method for primary frequency regulation performance of a thermal power plant unit, comprising:

[0005] Considering the coupling effect of valve regulation and main steam pressure change of the steam turbine of the thermal power plant unit, a dynamic response model of the thermal power unit is constructed;

[0006] Calculate the equivalent amount of electricity generated by the energy conversion between the boiler and the steam turbine, and perform sensitivity analysis on the key operating parameters of the thermal power unit collected in real time based on the equivalent amount of electricity;

[0007] An online evaluation model is constructed to evaluate the primary frequency regulation capability of thermal power units in real time.

[0008] In one embodiment, calculating the equivalent amount of electricity generated when energy is converted between a boiler and a steam turbine includes:

[0009] Under different operating conditions, the thermal power unit converts the energy stored in the boiler with the energy between the steam turbine and determines the energy conversion efficiency and energy change corresponding to each operating condition;

[0010] The equivalent electric quantity in each operating state is determined according to the product of the energy conversion efficiency corresponding to each operating state and the energy variation.

[0011] In one embodiment, a sensitivity analysis is performed on key operating parameters, including:

[0012] Using the constructed deep learning clustering model, different operating conditions are identified through key operating parameters;

[0013] Perform adaptive parameter weight calculation in each operating condition to obtain the weight of each key operating parameter;

[0014] According to the weight of each key operating parameter, the sensitivity of each key operating parameter to the equivalent electric quantity is determined.

[0015] In one embodiment, before using the constructed deep learning clustering model to identify different operating conditions through key operating parameters, the method further includes:

[0016] The deep learning clustering model is trained with the calculated equivalent power of thermal power units under different operating conditions as input, and the equivalent power under the same operating condition is clustered into one category.

[0017] In one embodiment, the deep learning clustering model adopts a multi-layer neural network structure and constructs a loss function using softmax fusion center loss.

[0018] In one embodiment, a coding model of a VAE variational autoencoder network is adopted, key operating parameters are used as the number of input neurons, and the weights of each key parameter under corresponding working conditions are output.

[0019] In one embodiment, the steam flow rate is collected in real time using an edge device to determine the rate of change of the steam flow rate and the steam turbine power;

[0020] The online evaluation model is constructed by taking into account the equivalent electrical quantity and the steam turbine power.

[0021] In one embodiment, the online evaluation method further comprises:

[0022] Set the safety margin and evaluate the safety margin of thermal power units during frequency regulation, including:

[0023] Calculate the difference between the maximum power variation that the thermal power unit can theoretically withstand under different operating conditions and the power variation of the thermal power unit during the actual frequency modulation process.

[0024] The safety margin is obtained according to the ratio of the difference to the maximum power variation.

[0025] In one embodiment, the online evaluation method further comprises:

[0026] Set an early warning threshold to warn of the risks of thermal power units during frequency regulation, including:

[0027] Calculate the critical equivalent power value of the thermal power unit under different operating conditions, and set the warning threshold based on the critical equivalent power value.

[0028] In one embodiment, the online evaluation method further comprises:

[0029] The frequency regulation strategy of the primary frequency regulation of the thermal power unit is adaptively optimized, including:

[0030] Automatically optimize the frequency regulation strategy of primary frequency regulation of thermal power units through learning algorithms;

[0031] Use closed-loop feedback control to dynamically adjust the control parameters of the primary frequency regulation of thermal power units.

[0032] In one embodiment, the coupling effect of steam turbine valve regulation and main steam pressure change is considered, and the turbine speed change is determined based on the main steam pressure, steam enthalpy, steam mass flow rate, and steam pressure drop change caused by valve displacement to construct the dynamic response model.

[0033] On the other hand, the present invention also provides an online evaluation device for the primary frequency regulation performance of a thermal power plant unit, which adopts the above-mentioned online evaluation method for the primary frequency regulation performance of a thermal power plant unit. The online evaluation device for the primary frequency regulation performance of a thermal power plant unit comprises:

[0034] Dynamic response model building module, which is used to consider the coupling effect of valve regulation and main steam pressure change of the steam turbine of the thermal power plant unit and build the dynamic response model of the thermal power unit;

[0035] The parameter optimization module is used to calculate the equivalent amount of electricity generated by the energy conversion between the boiler and the steam turbine. Based on the equivalent amount of electricity, a sensitivity analysis is performed on the key operating parameters of the thermal power unit collected in real time.

[0036] The evaluation model building module is used to build an online evaluation model to evaluate the primary frequency regulation capability of thermal power units in real time.

[0037] In addition, another aspect of the present invention also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above-mentioned online evaluation method for the primary frequency regulation performance of a thermal power plant unit are implemented, and the same technical effect can be achieved.

[0038] The present invention has outstanding substantive features and significant progress compared to the prior art, specifically:

[0039] The present invention proposes an online evaluation method for the primary frequency regulation performance of a thermal power plant unit. The method considers the coupling effect of valve regulation of the steam turbine of the thermal power plant unit and the change in main steam pressure, and constructs a dynamic response model of the thermal power unit to accurately capture the physical phenomena and behaviors of the thermal power unit in the frequency regulation process, and provides an accurate frequency regulation performance evaluation. At the same time, the equivalent electric quantity is introduced, and based on the equivalent electric quantity, a sensitivity analysis is performed on the key operating parameters of the thermal power unit collected in real time, and the complex energy conversion process is quantified into indicators that are easy to understand and evaluate. In addition, an online evaluation model is constructed using edge computing technology to evaluate the primary frequency regulation capability of the thermal power unit in real time, ensuring the real-time and accuracy of the evaluation system, so that the evaluation process can timely reflect the current operating status and grid demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the overall process of an online evaluation method for primary frequency regulation performance of a thermal power plant unit according to an embodiment of the present invention;

[0041] Figure 2 Shows Figure 1 A detailed schematic diagram of the process of performing sensitivity analysis on key operating parameters in step S2;

[0042] Figure 3 It is a schematic block diagram of the overall structure of an online evaluation device for primary frequency regulation performance of a thermal power plant unit according to an embodiment of the present invention;

[0043] Wherein, the accompanying drawings are marked as follows:

[0044] 300-Online evaluation device for primary frequency regulation performance of thermal power plant units;

[0045] 301-dynamic response model building module;

[0046] 302- parameter optimization module;

[0047] 303-Evaluation model building module. DETAILED DESCRIPTION

[0048] In order to make the above features and effects of the present invention more clearly understood, embodiments are given below and described in detail with reference to the accompanying drawings.

[0049] The purpose of this invention is to overcome the problems of the prior art in accurately quantifying the frequency regulation capability of thermal power plants under different working conditions, especially the dynamic response under low load operation, and propose an online evaluation method for the primary frequency regulation performance of thermal power plant units based on dynamic energy conversion process analysis and equivalent power quantification. Figure 1 As shown, Figure 1 The overall flow chart of the online evaluation method for the primary frequency regulation performance of the thermal power plant unit is shown.

[0050] An online evaluation method for primary frequency regulation performance of a thermal power plant unit comprises at least the following steps:

[0051] Step S1: Considering the coupling effect of the valve regulation of the steam turbine and the main steam pressure change, a dynamic response model of the thermal power unit is constructed.

[0052] The dynamic response model of the thermal power unit is constructed based on the analysis of the frequency regulation capability of the thermal power unit. The dynamic response model of the thermal power unit takes into account the coupling effect of the valve regulation of the steam turbine in the thermal power unit and the change of the main steam pressure, ensuring that the model can accurately capture the physical phenomena in the frequency regulation process, laying a solid foundation for the subsequent equivalent power quantification and evaluation. The dynamic response in the model reflects the actual frequency regulation operation, providing a reference for real-time monitoring and performance evaluation in subsequent steps.

[0053] Specifically, considering the coupling effect of steam turbine valve regulation and main steam pressure change, the change in turbine speed is determined based on the change in steam pressure drop caused by main steam pressure, steam enthalpy, steam mass flow rate, and valve displacement, and the dynamic response model is constructed as follows:

[0054]

[0055] Where: P main is the main steam pressure, h st is the steam enthalpy, is the steam mass flow rate, ΔP valve is the change in steam pressure drop caused by valve displacement, ΔN turbine is the change in turbine speed.

[0056] In this embodiment, by constructing a dynamic response model of the thermal power unit, the physical phenomena and behaviors of the thermal power unit during the frequency regulation process are accurately captured, providing an accurate frequency regulation performance evaluation.

[0057] Step S2: Calculate the equivalent amount of electricity generated during energy conversion between the boiler and the steam turbine, and perform sensitivity analysis on key operating parameters of the thermal power unit collected in real time based on the equivalent amount of electricity.

[0058] In one embodiment, the thermal power unit converts the energy stored in the boiler and the energy between the steam turbine under different operating conditions, and determines the energy conversion efficiency and energy change amount corresponding to each operating condition, specifically:

[0059] ΔE storage =η(ΔE input -ΔE output )

[0060] Where: ΔE storage is the change in energy stored in the boiler, η is the conversion efficiency, ΔE inputis the input energy of the steam turbine, ΔE output is the output energy of the steam turbine.

[0061] Furthermore, the equivalent electric quantity under each operating state is determined according to the product of the energy conversion efficiency and the energy change under each operating state. The equivalent electric quantity is directly related to the efficiency of the frequency modulation operation. The introduction of the equivalent electric quantity is to quantify the complex physical process into an indicator that is easy to evaluate. The equivalent electric quantity is expressed as:

[0062] ΔEE=η·ΔE h

[0063] Where: ΔEE is the equivalent amount of electricity, η is the conversion efficiency, ΔE h is the energy change, ΔE h =ΔE storage .

[0064] Furthermore, based on the equivalent power, the sensitivity of the key operating parameters of the thermal power units collected by real-time monitoring is analyzed. The sensitivity analysis of key operating parameters further refines which factors have a significant impact on the frequency regulation capability and guides the optimization of operating parameters to improve the frequency regulation efficiency. The output of this step provides quantitative input for the next step of online evaluation.

[0065] Specifically, deep learning technology is used to perform sensitivity analysis and calculate the contribution of key operating parameters to equivalent power. Different key operating parameters are assigned different weights, and the weights change dynamically according to different data of key operating parameters under different working conditions. For details, see Figure 2 As shown in Figure 2 A detailed schematic diagram of the process of performing sensitivity analysis on key operating parameters in step S2 is shown.

[0066] First, the constructed deep learning clustering model is used to identify different operating conditions through key operating parameters (S21); then, adaptive parameter weight calculation is performed in each operating condition to obtain the weight of each key operating parameter (S22). According to the weight of each key operating parameter, the sensitivity of each key operating parameter to the equivalent power is determined (S23). The key operating parameter x i The sensitivity to the equivalent electric charge ΔEE is denoted by S i .

[0067] In step S21, in one embodiment, the deep learning clustering model uses multi-layer neurons for feature extraction and uses improved softmax as a loss function. The improved softmax is implemented by integrating traditional softmax with centor loss.

[0068] For traditional softmax, the formula is:

[0069]

[0070] Where W is the weight of parameter x, and b is the bias of parameter x. Using only the traditional softmax loss function to train the network, the deep features obtained only have separability, but cannot constrain the distance between parameters under different working conditions. Such features are not sufficient to effectively identify the operating conditions.

[0071] Therefore, in this embodiment, center loss is introduced to specify a category center for each working condition. The features corresponding to the parameters of the same working condition are as close to their own category center as possible, and the category centers of different working conditions are as far away as possible. In the training process, the distance constraint between the sample and the class center in the feature space after network mapping is added, so as to take into account both intra-class aggregation and inter-class separation. The Centor loss formula is expressed as:

[0072]

[0073] in, It is the center of each working condition. Its data dimension is consistent with the number of key operating parameters.

[0074] Therefore, the loss function obtained by fusing the traditional softmax with the centor loss is:

[0075]

[0076] In this embodiment, the deep learning clustering model is trained with the calculated equivalent power of the thermal power unit under different operating conditions as input, the equivalent power under the same operating condition is clustered into one category, and the clustering set of different operating conditions is output. Then, the trained deep learning clustering model is used to identify different operating conditions with key operating parameters as input.

[0077] In step S22, in one embodiment, the neural network concept is adopted, and the key operating parameters are used as the number of input neurons. The factor analysis network of different key operating parameters under the same working conditions is learned through the codec network. In one embodiment, a VAE variational autoencoder network is specifically adopted, in which the encoder uses an M*N multi-layer neural network, and the decoder uses an N*M multi-layer neural network. The final network model is: M*N*X*N*M. The number of nodes in the X layer is equal to the number of key operating parameters, and its value represents the weight of each key operating parameter.

[0078] In this embodiment, equivalent electric quantity is introduced to quantify the complex energy conversion process into an index that is easy to understand and evaluate, providing a new method for evaluating the frequency regulation capability of thermal power units under different loads. In addition, based on equivalent electric quantity, deep learning technology is used to perform sensitivity analysis, calculate the contribution of key operating parameters to equivalent electric quantity, and guide the optimization of key operating parameters to improve frequency regulation efficiency.

[0079] Step S3: construct an online evaluation model to evaluate the primary frequency regulation capability of the thermal power unit in real time.

[0080] In one embodiment, the edge device is first used for preliminary processing to monitor the steam flow in real time, and the change rate of the steam flow ΔQ is obtained by the difference method. st ,Right now:

[0081]

[0082] Among them, Q st1 is the steam flow rate measured at time t1, Q st2 is the steam flow measured at time t2.

[0083] After determining the rate of change of the steam flow rate, the steam turbine power can be further determined based on the rate of change of the steam flow rate. The steam turbine power is the product of the steam flow rate and the steam enthalpy value.

[0084] Furthermore, by comprehensively considering parameters such as equivalent power, steam turbine power and time, real-time feedback of scheduling decisions is provided, and an online evaluation model is constructed to evaluate the primary frequency regulation capability in real time. turbine , ΔEE,t) is expressed as:

[0085] F=α·N turbine +β·ΔEE+γ·t

[0086] Where: α, β, γ are weight coefficients, N turbine is the turbine output power and t is the time.

[0087] Through real-time data collection and preliminary processing of edge devices, the real-time and accuracy of the evaluation is ensured, and the model prediction is closely linked to the actual operation. turbie , ΔEE,t) uses the equivalent power ΔEE, turbine power ΔN turbine The dynamics of this stage ensures the interaction between the model and actual operations, improving the response speed.

[0088] In addition, in some embodiments, the safety margin is further calculated to evaluate the safety margin of the thermal power unit during the frequency regulation process.

[0089] Safety margin S margin Defines the actual power change ΔN achieved by a thermal power unit in a frequency regulation operation actual The maximum theoretical power variation ΔN max The difference between the two is given in dimensionless form. The safety margin reflects the stability and frequency regulation capability margin of the system when facing frequency disturbances, that is, the safety margin S margin It is expressed as:

[0090]

[0091] Among them, ΔN max Indicates the maximum power change that a thermal power unit can theoretically withstand, that is, the upper limit of frequency regulation. ΔN actual It indicates the power change of the thermal power unit during an actual frequency regulation process.

[0092] Safety margin S margin Indicates the relative space of the actual frequency modulation operation distance from the theoretical maximum safety limit. If the safety margin S margin A larger value indicates that the system has a larger margin for safe operation; otherwise, it means that it is approaching or exceeding the safety limit and there is a potential risk.

[0093] By calculating the safety margin S of thermal power units under different operating conditions margin , the power system can better understand the safety status of thermal power units when performing a frequency regulation task, and adjust the strategy in time to ensure the stable operation and safety of the power grid.

[0094] In addition, in some embodiments, a warning threshold is further set to provide early warning of the risks of thermal power units during the frequency regulation process.

[0095] Specifically, the warning threshold W under different working conditions can be set according to the historical operating conditions of the thermal power units and the power grid demand. thresh , when the system approaches or exceeds the warning threshold, an alert is triggered:

[0096] W threh =α·ΔEE crt +β

[0097] Where: W thresh is the warning threshold, α, β are constants, ΔEE crit is the critical equivalent electrical quantity value.

[0098] Safety margin S margin The assessment provides a safe boundary for system operation, ensuring that the frequency modulation is within a controllable range and avoiding instability. The risk warning system sets the warning threshold W threshEarly warning of potential risks, combining historical and real-time data, and timely adjustment of strategies ensure that frequency modulation operations are carried out under safe conditions and avoid potential system shocks.

[0099] In addition, in some embodiments, the frequency regulation strategy of the primary frequency regulation of the thermal power unit is further adaptively optimized, and then the optimized frequency regulation strategy can be evaluated using the evaluation method disclosed in the above steps.

[0100] The frequency regulation strategy optimization process consists of two parts. The first part is to automatically optimize the frequency regulation strategy of the primary frequency regulation of the thermal power unit through the learning algorithm. The optimized frequency regulation strategy is:

[0101]

[0102] Where: π new is the optimized frequency modulation strategy, π old is the frequency modulation strategy before optimization, η learn is the learning rate, It’s a reward signal.

[0103] The frequency regulation strategy is updated through the learning algorithm and automatically optimized based on the reward signal, thus enhancing the adaptability to power grid changes.

[0104] Second: Use closed-loop feedback control to dynamically adjust the control parameters of the primary frequency regulation of thermal power units to improve the accuracy and stability of frequency regulation. The closed-loop control law formula is as follows:

[0105]

[0106] Where: u k 、u k+1 is the control input, e j is the error, K p , K i , K d is the gain of the PID controller, e k is the current error, and de / dt is the rate of change of the error.

[0107] The closed-loop control law dynamically adjusts the control parameters according to the error and its rate of change, improving the accuracy and stability of frequency regulation. The final closed loop ensures the dynamic optimization of the frequency regulation capability of the thermal power unit, responding to the grid demand in the most efficient and safe way.

[0108] In summary, the online evaluation method for the primary frequency regulation performance of a thermal power plant unit disclosed in the present invention considers the coupling effect of valve regulation of the steam turbine of the thermal power plant unit and the change of main steam pressure, and constructs a dynamic response model of the thermal power unit to accurately capture the physical phenomena and behaviors of the thermal power unit in the frequency regulation process, and provides an accurate frequency regulation performance evaluation. At the same time, the equivalent electric quantity is introduced, and based on the equivalent electric quantity, a sensitivity analysis is performed on the key operating parameters of the thermal power unit collected in real time, and the complex energy conversion process is quantified into indicators that are easy to understand and evaluate. In addition, an online evaluation model is constructed using edge computing technology to evaluate the primary frequency regulation capability of the thermal power unit in real time, ensuring the real-time and accuracy of the evaluation system, so that the evaluation process can timely reflect the current operating status and grid demand.

[0109] In addition, corresponding to the above-mentioned on-line evaluation method for the primary frequency regulation performance of a thermal power plant unit, another embodiment of the present invention further discloses an on-line evaluation device for the primary frequency regulation performance of a thermal power plant unit, referring to Figure 3 , Figure 3 The structure diagram of the on-line evaluation device 300 for the primary frequency regulation performance of the thermal power plant unit is shown, which can be realized by Figure 1 , Figure 2 The various processes of the online evaluation method for primary frequency regulation performance of thermal power plant units are shown.

[0110] An online evaluation device 300 for primary frequency regulation performance of a thermal power plant unit, the device at least comprising:

[0111] A dynamic response model building module 301 is used to consider the coupling effect of valve regulation and main steam pressure change of the steam turbine of the thermal power plant unit to build a dynamic response model of the thermal power unit;

[0112] The parameter optimization module 302 is used to calculate the equivalent electric quantity generated when the energy is converted between the boiler and the steam turbine, and to perform sensitivity analysis on the key operating parameters of the thermal power unit collected in real time based on the equivalent electric quantity;

[0113] The evaluation model building module 303 is used to build an online evaluation model for real-time evaluation of the primary frequency regulation capability of the thermal power unit.

[0114] In addition, corresponding to the above-mentioned online evaluation method for the primary frequency regulation performance of thermal power plant units, another embodiment of the present invention also discloses an electronic device, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of the above-mentioned online evaluation method for the primary frequency regulation performance of thermal power plant units are implemented, and the same technical effect can be achieved.

[0115] The processor is the processor in the electronic device in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0116] Another embodiment of the present invention further provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the above-mentioned steps of wide-band oscillation monitoring are implemented and the same technical effect can be achieved.

[0117] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be applied, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0119] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. An online evaluation method for primary frequency regulation performance of thermal power plant units, characterized in that: Include: Considering the coupling effect of valve regulation and main steam pressure change of the steam turbine of the thermal power plant unit, a dynamic response model of the thermal power unit is constructed; Calculate the equivalent amount of electricity generated by the energy conversion between the boiler and the steam turbine, and perform sensitivity analysis on the key operating parameters of the thermal power unit collected in real time based on the equivalent amount of electricity; An online evaluation model is constructed to evaluate the primary frequency regulation capability of thermal power units in real time.

2. The method according to claim 1, characterized in that Calculate the equivalent amount of electricity generated when converting energy between boilers and steam turbines, including: Under different operating conditions, the thermal power unit converts the energy stored in the boiler with the energy between the steam turbine and determines the energy conversion efficiency and energy change corresponding to each operating condition; The equivalent electric quantity in each operating state is determined according to the product of the energy conversion efficiency corresponding to each operating state and the energy variation.

3. The method according to claim 1 or 2, characterized in that: Conduct sensitivity analysis on key operating parameters, including: Using the constructed deep learning clustering model, different operating conditions are identified through key operating parameters; Perform adaptive parameter weight calculation in each operating condition to obtain the weight of each key operating parameter; According to the weight of each key operating parameter, the sensitivity of each key operating parameter to the equivalent electric quantity is determined.

4. The method according to claim 3, characterized in that Before using the constructed deep learning clustering model to identify different operating conditions through key operating parameters, it also includes: The deep learning clustering model is trained with the calculated equivalent power of thermal power units under different operating conditions as input, and the equivalent power under the same operating condition is clustered into one category.

5. The method according to claim 3, characterized in that: The deep learning clustering model adopts a multi-layer neural network structure and uses softmax fusion center loss to construct a loss function.

6. The method according to claim 3, characterized in that The encoding model of VAE variational autoencoder network is adopted, and the key operating parameters are used as the input neuron number to output the weights of each key parameter under the corresponding working conditions.

7. The method according to claim 1, characterized in that Use edge devices to collect steam flow in real time, determine the rate of change of steam flow, and steam turbine power; The online evaluation model is constructed by taking into account the equivalent electrical quantity and the steam turbine power.

8. The method according to claim 1, characterized in that Also includes: Set the safety margin and evaluate the safety margin of thermal power units during frequency regulation, including: Calculate the difference between the maximum power variation that the thermal power unit can theoretically withstand under different operating conditions and the power variation of the thermal power unit during the actual frequency modulation process. The safety margin is obtained according to the ratio of the difference to the maximum power variation.

9. The method according to claim 1, characterized in that: Also includes: Set an early warning threshold to warn of the risks of thermal power units during frequency regulation, including: Calculate the critical equivalent power value of the thermal power unit under different operating conditions, and set the warning threshold based on the critical equivalent power value.

10. The method according to claim 1, characterized in that Also includes: The frequency regulation strategy of the primary frequency regulation of the thermal power unit is adaptively optimized, including: Automatically optimize the frequency regulation strategy of primary frequency regulation of thermal power units through learning algorithms; Use closed-loop feedback control to dynamically adjust the control parameters of the primary frequency regulation of thermal power units.

Images