Method and device for predicting primary frequency modulation margin of thermal power generating unit
By obtaining the actual working conditions parameters and simulation results of the thermal power unit, a constraint model of the valve step amount is generated to predict its first frequency modulation margin, which solves the problem of the thermal power unit's frequency modulation capability decrease after ultra-high voltage transmission, and improves the frequency modulation margin prediction Accuracy and grid stability.
Patent Information
- Application Number
- CN202410914118.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-08
AI Technical Summary
The frequency modulation capability of thermal power units decreases after ultra-high voltage transmission is connected. It is difficult for the existing technology to accurately reflect the frequency modulation performance of the unit in real time, resulting in an increase in frequency modulation instability and low-frequency impact risks.
By obtaining the turbine power, main steam parameters and valve opening, a constraint model of the valve step amount is generated. The pre-trained frequency modulation margin prediction model is used, and the valve step amount is determined based on the actual working condition constraints and simulation results, and then the valve step amount is predicted once Frequency modulation margin improves the accuracy of frequency modulation capabilities.
The prediction of the maximum output response of the unit's primary frequency modulation unit based on the current working conditions is achieved, providing a decision-making basis for power plant operation and power grid scheduling, improving the accuracy of frequency modulation margin prediction, and reducing frequency modulation instability and low-frequency impact risks.
Smart Images

Figure CN118920504A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of primary frequency regulation control, and in particular to a method and device for predicting primary frequency regulation margin of a thermal power unit. Background Art
[0002] In order to realize the transmission of clean energy from the west to the load center and solve the contradiction between the layout of energy resources and load demand, it is necessary to vigorously develop UHV transmission technology, complete the long-distance low-loss transmission of electric energy, and realize the high-quality interconnection of electric energy. However, the access of UHV transmission will change the supply and demand situation of local power input, which will greatly change the distribution of power grid flow, and the power supply and demand situation will also change from relative balance to absolute surplus. The impact on the primary frequency regulation of the receiving power grid is mainly reflected in two points: on the one hand, since the UHV parameters are constant and do not have the frequency regulation capability, once this part of the large proportion of power supply fails, it will bring serious low-frequency impact to the receiving power grid and increase the frequency regulation burden of other units. To some extent, it increases the frequency regulation risk of the receiving end; on the other hand, in order to ensure the balance of supply and demand, large-scale UHV transmission access will occupy the power generation space of the receiving power grid units and reduce the load rate of the units. The primary frequency regulation capability of the units under low load rate is relatively poor, which will also increase the instability of frequency regulation. Therefore, the access of UHV transmission brings higher requirements and greater challenges to the primary frequency regulation of the receiving power grid.
[0003] The main primary frequency regulation method of thermal power units at present is to change the steam inlet of the unit by relying on the spontaneous action of the valve under the stimulation of the frequency deviation signal. Although each unit has basically invested in the primary frequency regulation function technically, in actual operation, the primary frequency regulation effect of some units is not as good as expected under the disturbance of large frequency difference. The reasons are analyzed as follows:
[0004] (1) Insufficient valve adjustment margin: In order to improve the economical operation of the unit, some units set the valve opening to a larger value when optimizing the sliding pressure operation parameters, so that the power increment brought by the valve margin during the primary frequency modulation does not meet the requirements;
[0005] (2) System control logic conflict: Since the main steam pressure will drop in a short time when the valve is increased, when the drop reaches a threshold, the protection logic of the main steam pressure regulation loop will be triggered, thereby reducing the valve opening to reduce the fluctuation of the main steam pressure, resulting in insufficient heat storage release of the unit;
[0006] (3) Insufficient heat storage of the units: Faced with frequent primary frequency regulation, the primary frequency regulation capacity of some units is consumed, and their contribution to supporting the next round of primary frequency regulation is insufficient;
[0007] (4) Problems with the equipment itself: Due to equipment aging and line problems, the primary frequency regulation performance of some units is poor, and even the primary frequency regulation circuit is cut off.
[0008] In general, when the control system is in normal operation, the primary frequency regulation capability of thermal power units is relatively dependent on their own operating conditions. Especially when large-scale renewable energy systems and ultra-high voltage transmission are connected to the grid, making low-load operation of the units a normal practice, the primary frequency regulation capability of the units will become worse.
[0009] In the prior art, a single decoupling model of the steam turbine is established, and its parameters are identified based on historical data; the overall heat storage of the boiler is calculated based on the heat storage of the cold fluid working medium at the boiler outlet and the heat storage of the boiler metal, and the dynamic response process equation of the primary frequency regulation performance of the steam turbine is obtained; the heat storage increment is combined with the steam turbine model to obtain the primary frequency regulation capability prediction. This method establishes a mechanism model. However, in actual operation, there are many factors affecting the primary frequency regulation of thermal power units, and there is a strong nonlinearity between the power change of the unit and the various variables. The mechanism model is difficult to accurately reflect the primary frequency regulation capability of the unit in real time, so the effect in practical applications needs to be improved. Summary of the invention
[0010] In view of the problems in the prior art, the embodiments of the present application provide a method and device for predicting the primary frequency regulation margin of a thermal power unit, which can at least partially solve the problems existing in the prior art.
[0011] In a first aspect, the present application provides a method for predicting a primary frequency regulation margin of a thermal power unit, comprising:
[0012] Obtain steam turbine power, main steam parameters and valve opening;
[0013] Generate a constraint model of valve step quantity according to the acquired actual working condition constraints;
[0014] Determining a valve step amount using the constraint model;
[0015] The steam turbine power, the main steam parameters, the valve opening and the valve step amount are input into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using acquired primary frequency regulation experimental data.
[0016] Furthermore, the generation of a constraint model for the valve step amount according to the acquired actual working condition constraint includes:
[0017] Perform valve step simulation according to actual working condition constraints and obtain simulation results;
[0018] The constraint model is generated according to the actual working condition constraints and the simulation results.
[0019] Furthermore, the simulation results include an upward step simulation result and a downward step simulation result; the valve step simulation according to the actual working condition constraints includes:
[0020] Performing a valve upward step simulation according to actual working condition constraints to obtain the upward step simulation result;
[0021] A valve downward step simulation is performed according to actual working condition constraints to obtain the downward step simulation result.
[0022] Furthermore, the step of pre-training the frequency modulation margin prediction model includes:
[0023] Get the data of a frequency modulation experiment;
[0024] Preprocessing the primary frequency modulation experimental data to obtain a training set and a test set;
[0025] The training set and the test set are used to perform model training to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights of hidden layer output to output layer.
[0026] In a second aspect, the present application provides a device for predicting primary frequency regulation margin of a thermal power unit, comprising:
[0027] Actual data acquisition unit, used to obtain turbine power, main steam parameters and valve opening;
[0028] A constraint model generating unit, used for generating a constraint model of a valve step amount according to the acquired actual working condition constraints;
[0029] A valve step amount determination unit, used for determining the valve step amount using the constraint model;
[0030] The frequency regulation margin prediction unit is used to input the turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data.
[0031] Furthermore, the constraint model generating unit includes:
[0032] A simulation result generation module is used to perform valve step simulation according to actual working condition constraints to obtain simulation results;
[0033] A constraint model generation module is used to generate the constraint model according to the actual working condition constraints and the simulation results.
[0034] Furthermore, the simulation result generation module includes:
[0035] An upward step simulation submodule, used to perform an upward step simulation of the valve according to actual working condition constraints to obtain the upward step simulation result;
[0036] The downward step simulation submodule is used to perform a downward step simulation of the valve according to actual working condition constraints to obtain the downward step simulation result.
[0037] Furthermore, it also includes:
[0038] An experimental data acquisition unit, used for acquiring primary frequency modulation experimental data;
[0039] A data preprocessing unit, used to preprocess the primary frequency modulation experiment data to obtain a training set and a test set;
[0040] The model training unit is used to perform model training using the training set and the test set to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights of the hidden layer output to the output layer.
[0041] In a third aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any one of the above embodiments when executing the computer program.
[0042] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.
[0043] In a fifth aspect, the present application provides a computer program product, wherein the computer program product includes a computer program, and when the computer program is executed by a processor, the method described in any of the above embodiments is implemented.
[0044] The method and device for predicting the primary frequency regulation margin of a thermal power unit provided in the present application obtain the steam turbine power, main steam parameters and valve opening; generate a constraint model of the valve step amount according to the actual operating condition constraints obtained; determine the valve step amount using the constraint model; input the steam turbine power, the main steam parameters, the valve opening and the valve step amount into the pre-trained frequency regulation margin prediction model to obtain the frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data, and realizes the prediction of the maximum output response of the primary frequency regulation of the unit based on the current operating condition of the generator set, providing a decision-making basis for power plant operators and power grid dispatching staff. Among them, by generating a constraint model of the valve step amount according to the actual operating condition constraints obtained; and determining the valve step amount using the constraint model, the accuracy of the valve step amount is improved, thereby improving the accuracy of the prediction of the primary frequency regulation margin of the unit. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0046] Figure 1 It is a flow chart of a method for predicting primary frequency regulation margin of a thermal power unit provided in one embodiment of the present application;
[0047] Figure 2 It is a flow chart of a method for predicting primary frequency regulation margin of a thermal power unit provided in one embodiment of the present application;
[0048] Figure 3 It is a flow chart of a method for predicting primary frequency regulation margin of a thermal power unit provided in one embodiment of the present application;
[0049] Figure 4 It is a flow chart of a method for predicting primary frequency regulation margin of a thermal power unit provided in one embodiment of the present application;
[0050] Figure 5 It is a structural schematic diagram of a primary frequency regulation margin prediction device for a thermal power unit provided in one embodiment of the present application;
[0051] Figure 6 It is a structural schematic diagram of a primary frequency regulation margin prediction device for a thermal power unit provided in one embodiment of the present application;
[0052] Figure 7 It is a structural schematic diagram of a primary frequency regulation margin prediction device for a thermal power unit provided in one embodiment of the present application;
[0053] Figure 8 It is a structural schematic diagram of a primary frequency regulation margin prediction device for a thermal power unit provided in one embodiment of the present application;
[0054] Fig. 9 is a schematic diagram of the physical structure of an electronic device provided in one embodiment of the present application;
[0055] Fig.10 It is a constant pressure-sliding pressure-constant pressure curve in actual operation of a power plant provided by an embodiment of the present application;
[0056] Fig.11 is a structural diagram of an extreme learning machine provided by an embodiment of the present application;
[0057] Fig.12 It is a schematic diagram of model training provided in one embodiment of the present application. DETAILED DESCRIPTION
[0058] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but are not intended to limit the present application. It should be noted that, in the absence of conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other.
[0059] The following takes the server as an example of the execution subject to illustrate the specific implementation process of the primary frequency regulation margin prediction method of the thermal power unit provided in the embodiment of the present application.
[0060] Figure 1 FIG. 1 is a flow chart of a method for predicting a primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 1 As shown, the method for predicting the primary frequency regulation margin of a thermal power unit provided in this application includes:
[0061] S101: Obtaining steam turbine power, main steam parameters and valve opening;
[0062] S102: generating a constraint model of the valve step amount according to the acquired actual working condition constraints;
[0063] S103: Determine the valve step amount using the constraint model;
[0064] S104: Input the steam turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using acquired primary frequency regulation experimental data.
[0065] from Figure 1 It can be seen from the illustrated process that the method for predicting the primary frequency regulation margin of a thermal power unit provided in the present application obtains the steam turbine power, main steam parameters and valve opening; generates a constraint model of the valve step amount according to the actual operating condition constraints obtained; determines the valve step amount using the constraint model; inputs the steam turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data, and realizes the prediction of the maximum output response of the primary frequency regulation of the unit based on the current operating condition of the generator set, providing a decision-making basis for power plant operators and power grid dispatching staff.
[0066] Each step is explained in detail below.
[0067] S101: Obtaining steam turbine power, main steam parameters and valve opening;
[0068] Specifically, the primary frequency regulation capability refers to the maximum power change contributed by the unit's own heat storage capacity when the grid frequency is disturbed. The server uses the primary frequency regulation margin as an indicator parameter to evaluate the primary frequency regulation capability of the thermal power unit and makes a prediction based on the unit's operating status. The primary frequency regulation margin is the maximum tolerable power change value constrained by the unit's main steam parameters. To obtain the primary frequency regulation margin, it is necessary to consider the valve initial value, valve step amount, main steam flow, main steam pressure, and operating conditions.
[0069] The server predicts the primary frequency regulation margin of the thermal power unit using the turbine power, main steam parameters, valve opening and valve step amount. The main steam parameters include main steam pressure and main steam flow. Among them, the turbine power, main steam parameters and valve opening (valve initial value) can be read in real time from the measuring point of the unit, and the valve step amount needs to be obtained through simulation. Therefore, before predicting the primary frequency regulation margin of the unit, the server obtains the turbine power, main steam parameters and valve opening read in real time from the measuring point of the unit.
[0070] S102: generating a constraint model of the valve step amount according to the acquired actual working condition constraints;
[0071] Specifically, the valve step amount is subject to the constraints of actual operating conditions, which include valve upper limit constraints, valve lower limit constraints and main steam pressure constraints. The server generates a constraint model for the valve step amount based on the current operating condition constraints, valve upper limit constraints, valve lower limit constraints and main steam pressure constraints, etc., to determine the valve step amount according to the constraint model.
[0072] Figure 2 FIG. 1 is a flow chart of a method for predicting a primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 2 As shown, S102 includes:
[0073] S201: Perform valve step simulation according to actual working condition constraints to obtain simulation results;
[0074] Specifically, Fig.10 The following is the constant pressure-sliding pressure-constant pressure curve in the actual operation of the power plant. Fig.10 The unit can be operated in sliding pressure mode under 30%-90% load condition, and the main steam pressure changes with the load. The existing technology does not take into account the change of pressure in the primary frequency modulation, and the valve can be adjusted without limit. If the change of main steam pressure is introduced, the problem of margin limitation can be better considered.
[0075] When the valve steps, the main steam pressure will change in the opposite direction, and the pressure fluctuation will have a great impact on the unit. If the steam pressure is too low, the work done by the steam in the turbine will also be reduced, and the unit efficiency will decrease; if the steam pressure is too high, the mechanical stress of the unit will increase, causing safety failures. At the same time, large steam pressure fluctuations and frequent steam pressure fluctuations will also affect the safety of the boiler and turbine. Therefore, the change amplitude and rate of change of the main steam pressure must be strictly monitored and controlled. Therefore, when making a frequency regulation margin, the pressure limit needs to be considered. The server performs valve step simulation according to the main steam pressure constraint in the actual operating condition constraint to determine the valve step amount.
[0076] Figure 3 FIG. 1 is a flow chart of a method for predicting a primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 3 As shown, S201 includes:
[0077] S301: Performing a valve upward step simulation according to actual working condition constraints to obtain the upward step simulation result;
[0078] Specifically, during the primary frequency modulation process, the pressure fluctuation amplitude needs to be limited. According to the control logic of the power plant engineering master control, the bias limit of the pressure can be obtained to be ±2MPa. The extreme value of the valve step amount is inferred from the pressure change, and the power change value obtained is the primary frequency modulation margin based on the actual pressure limit under this working condition.
[0079] The server generates simulation results according to the actual working condition constraints, including upward step simulation results and downward step simulation results. The server performs valve upward step simulation under different working conditions, monitors the change value of the pressure curve, finds the valve step amount when the pressure change is reached, and records the current power change value as a frequency modulation margin.
[0080] The server obtains the upward step simulation results shown in Table 1 by performing valve upward step simulation. The upward step simulation results include the valve initial value, valve step amount, power initial value, the frequency regulation margin of each load point upward frequency regulation (when power increases) under different working conditions, pressure initial value, pressure change, flow initial value and flow change. When the pressure change is -2Mpa, or the final valve opening (valve initial value plus valve step amount) is 100%, the corresponding power change is the upper limit margin of the unit frequency regulation at this working point. It can be seen from Table 1 that the main steam pressure is higher and the valve is adjusted upward. When the valve initial value is high, the valve is fully opened and has not reached the pressure limit; as the main steam pressure of the unit increases, the valve step amount becomes lower and lower, and finally tends to be flat, reaching about 8%.
[0081] Table 1
[0082]
[0083] In one embodiment, the server may utilize a unit with a capacity of 330MW to perform valve upward step simulation, and may also utilize units with other capacities to perform valve upward step simulation, but the present application is not limited thereto.
[0084] S302: Perform valve downward step simulation according to actual working condition constraints to obtain the downward step simulation result.
[0085] Specifically, during the primary frequency modulation process, the pressure fluctuation amplitude needs to be limited. According to the control logic of the power plant engineering master control, the bias limit of the pressure can be obtained to be ±2MPa. The extreme value of the valve step amount is inferred from the pressure change, and the power change value obtained is the primary frequency modulation margin based on the actual pressure limit under this working condition.
[0086] The server generates simulation results according to the actual working condition constraints, including upward step simulation results and downward step simulation results. The server performs downward step simulation of the valve under different working conditions, monitors the change value of the pressure curve, finds the valve step amount when the pressure change is reached, and records the current power change value as the primary frequency regulation margin.
[0087] The server obtains the downward step simulation results shown in Table 2 by performing valve downward step simulation. The downward step simulation results include the initial value of the valve, the valve step amount, the initial power value, the primary frequency regulation margin of each load point when the power is reduced, the initial pressure value, the pressure change, the initial flow value and the flow change under different working conditions. If the initial value of the main steam pressure is high, when the valve is adjusted downward, the pressure is easy to reach the high alarm value and cause an overpressure alarm. It can be seen from Table 2 that in the three groups of working conditions below 300MW, the valve step amount decreases with the increase of the main steam pressure; in the four groups of working conditions above 300MW, due to the high initial pressure, the margin of the valve downward adjustment is very small, so the primary frequency regulation margin is small.
[0088] Table 2
[0089]
[0090] In one embodiment, the server may utilize a unit with a capacity of 330MW to perform valve step-down simulation, or may utilize units with other capacities to perform valve step-down simulation, but the present application is not limited thereto.
[0091] S202: Generate the constraint model according to the actual working condition constraints and the simulation results.
[0092] Specifically, the server generates a constraint model for the valve step amount based on the actual working condition constraints and simulation results. First, consider whether the maximum step amount of the valve is higher than the fully open state of the valve (valve upper limit constraint). When the fully open valve still does not meet the pressure limit requirements, the valve step amount is as follows:
[0093] Δu=100%-u 当前 (1)
[0094] Where Δu represents the valve step amount, u 当前 Indicates the current valve opening.
[0095] Next, consider the valve lower limit constraint, that is, when the valve steps downward, it cannot fall below the preset lower limit value.
[0096] In one embodiment, the lower limit value may be 30%, and the valve step amount is limited as follows:
[0097] Δu+u 当前 ≥30% (2)
[0098] Where Δu represents the valve step amount, u 当前 Indicates the current valve opening.
[0099] Considering the main steam pressure constraint, the relationship between the valve upward step amount and pressure is fitted according to the upward step simulation results. When the pressure is 15-22MPa, it changes approximately linearly. When the pressure is 22-24.5MPa, the valve is almost unchanged, which is expressed as:
[0100]
[0101] Among them, P represents the main steam pressure value under a certain working condition, and Δu represents the valve step amount.
[0102] Considering the main steam pressure constraint, the relationship between the valve downward adjustment step amount and pressure is obtained by fitting according to the downward step simulation results, which is expressed as:
[0103]
[0104] Among them, P represents the main steam pressure value under a certain working condition, and Δu represents the valve step amount.
[0105] Finally, in addition to considering the margin of the unit in the non-frequency modulation section, it is also necessary to consider that when the unit has undergone a frequency modulation, the valve value has changed. At this time, the step value of the valve in the frequency modulation section needs to be corrected based on the frequency modulation:
[0106] Δu 调频 =Δu 非调频 -(|u 当前 -u 调频起始点 |) (5)
[0107] Among them, u 当前 Indicates the current valve opening, u 调频起始点 Indicates the valve opening at the start of frequency modulation, Δu 调频 Indicates the valve step amount after a frequency modulation, Δu 非调频 Indicates the valve step amount before a frequency modulation.
[0108] The above formulas (1) to (5) are the constraint models of valve step quantities generated by the server according to actual working condition constraints and simulation results.
[0109] S103: Determine the valve step amount using the constraint model;
[0110] Specifically, the valve step amount is subject to the constraints of the actual operating conditions, which include the current operating condition constraints, the valve upper limit constraints, the valve lower limit constraints and the main steam pressure constraints. After the server generates a constraint model for the valve step amount based on the actual operating condition constraints, it determines the valve step amount based on the constraint model and the current valve opening (valve initial value).
[0111] S104: Input the steam turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using acquired primary frequency regulation experimental data.
[0112] Specifically, the server uses the acquired primary frequency regulation experimental data to pre-train a frequency regulation margin prediction model. The input of the frequency regulation margin prediction model is turbine power, main steam parameters, valve opening (valve initial value) and valve step amount. The output of the frequency regulation margin prediction model is the frequency regulation margin prediction result (primary frequency regulation margin).
[0113] In one embodiment, the steam turbine power, i.e., the load range of a conventional thermal power unit, is 50% of the rated load to 100% of the rated load.
[0114] Figure 4 FIG. 1 is a flow chart of a method for predicting a primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 4 As shown, the steps of pre-training the frequency modulation margin prediction model include:
[0115] S401: Obtaining a frequency modulation experiment data;
[0116] Specifically, the server obtains primary frequency modulation experimental data, which includes turbine power, main steam parameters, valve opening (valve initial value), valve step amount and primary frequency modulation margin, which are used for model training to obtain a frequency modulation margin prediction model.
[0117] S402: preprocessing the primary frequency modulation experiment data to obtain a training set and a test set;
[0118] Specifically, the server preprocesses the frequency modulation experiment data, including data cleaning, data normalization, data completion and data partitioning, to remove low-quality data and obtain high-quality training sets and test sets, so as to use the high-quality training sets and test sets for model training to generate a frequency modulation margin prediction model.
[0119] S403: Perform model training using the training set and the test set to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights from the hidden layer output to the output layer.
[0120] Specifically, the server uses the training set and the test set to train the model and obtain the frequency margin prediction model. The frequency margin prediction model is obtained based on neural network training. The server first randomly initializes the weights and biases from the input layer to the hidden layer, then performs a nonlinear transformation on the input data through the hidden layer to obtain the hidden layer output matrix, and finally uses the hidden layer output matrix and the target output of the training data to solve the weights from the hidden layer output to the output layer. The input of the frequency margin prediction model is the turbine power, main steam parameters, valve opening (valve initial value) and valve step amount, and the output of the frequency margin prediction model is the frequency margin prediction result (primary frequency margin).
[0121] In one embodiment, the server performs model training based on a single hidden layer feedforward neural network (SLFN), which consists of an input layer, a hidden layer and an output layer, including a perceptron, a multi-layer perceptron (MLP), a radial basis function network (RBFN), a self-organizing map (SOM) and an extreme learning machine (ELM), etc., but the present application is not limited to this.
[0122] In one embodiment, the server obtains a frequency modulation margin prediction model based on extreme learning machine (ELM) training, and its structure is as follows: Fig.11 As shown, the input weights and hidden layer thresholds are randomly generated and independent of the training sample data, and the output layer weight matrix is obtained by one-step analytical calculation.
[0123] For the observed sample (x i ,ti ), where the input vector x i =[x i1 ,x i2 ,x i3 ,…x in ] T ∈R n , (i=1,2,3,…,N), N is the number of observed samples, n is the dimension of the sample input vector, and the output vector t i =[t i1 ,t i2 ,…t im ] T ∈R m , m is the output sample dimension, that is, the number of nodes in the output layer of the ELM model. Assuming the number of nodes in the hidden layer of the model is l, and the activation function is g(·), the ELM model is:
[0124]
[0125] In the formula, β i =[β i1 ,β i2 ,…,β im ] T is the weight from the i-th node in the hidden layer to the output layer of the model, w i =[w i1 ,w i2 ,…,w in ] T is the weight of the i-th node in the model input layer and hidden layer, b i represents the threshold of the i-th node in the hidden layer. The output value of the ELM model can fit the sample with zero error, that is:
[0126]
[0127] That is, there is β i , w i and b i satisfy:
[0128]
[0129] It can be shortened to:
[0130] Hβ=T (9)
[0131] in:
[0132]
[0133] In the formula, H is called the hidden layer output matrix. In the model training stage, the input weights and bias values of the feedforward neural network are randomly set. By calculating the output matrix H, the ELM learning and training problem is transformed into a least squares norm problem for solving the output weight matrix β, that is:
[0134]
[0135] Where H + is the generalized inverse matrix of the matrix H.
[0136] like Fig.12 As shown, the server uses the training set and the test set to train the model based on the extreme learning machine to obtain the frequency regulation margin prediction model. The input of the frequency regulation margin prediction model is the turbine power, main steam parameters, valve opening (valve initial value) and valve step amount, and the output of the frequency regulation margin prediction model is the frequency regulation margin prediction result (primary frequency regulation margin).
[0137] The present application provides a method for predicting the primary frequency regulation margin of a thermal power unit, by obtaining the steam turbine power, main steam parameters and valve opening; generating a constraint model of the valve step amount according to the actual operating condition constraints obtained; determining the valve step amount using the constraint model; inputting the steam turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data, and realizes the prediction of the maximum output response of the primary frequency regulation of the unit based on the current operating condition of the generator unit, providing a decision-making basis for power plant operators and power grid dispatching staff. Among them, by generating a constraint model of the valve step amount according to the actual operating condition constraints obtained; and determining the valve step amount using the constraint model, the accuracy of the valve step amount is improved, thereby improving the accuracy of the prediction of the primary frequency regulation margin of the unit.
[0138] Based on the same inventive concept, the embodiment of the present application also provides a device for predicting the primary frequency regulation margin of a thermal power unit, which can be used to implement the method described in the above embodiment, as described in the following embodiment. Since the principle of solving the problem by the device for predicting the primary frequency regulation margin of a thermal power unit is similar to the method for predicting the primary frequency regulation margin of a thermal power unit, the implementation of the device for predicting the primary frequency regulation margin of a thermal power unit can refer to the implementation of the method for determining the software performance benchmark, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements predetermined functions. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.
[0139] Figure 5 : is a schematic diagram of the structure of a device for predicting the primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application, such as Figure 5 As shown, the device comprises:
[0140] Actual data acquisition unit 501, used to acquire steam turbine power, main steam parameters and valve opening;
[0141] A constraint model generating unit 502, used to generate a constraint model of the valve step amount according to the acquired actual working condition constraints;
[0142] A valve step amount determination unit 503, used to determine the valve step amount using the constraint model;
[0143] The frequency regulation margin prediction unit 504 is used to input the turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data.
[0144] Figure 6 is a schematic diagram of the structure of a device for predicting the primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 5 Based on the embodiment, further, Figure 6 As shown, the primary frequency regulation margin prediction device for a thermal power unit provided in the present application also includes:
[0145] A simulation result generation module 601 is used to perform valve step simulation according to actual working condition constraints to obtain simulation results;
[0146] The constraint model generation module 602 is used to generate the constraint model according to the actual working condition constraints and the simulation results.
[0147] Figure 7 is a schematic diagram of the structure of a device for predicting the primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 6 Based on the embodiment, further, Figure 7 As shown, the primary frequency regulation margin prediction device for a thermal power unit provided in the present application also includes:
[0148] An upward step simulation submodule 701 is used to perform an upward step simulation of the valve according to actual working condition constraints to obtain the upward step simulation result;
[0149] The downward step simulation submodule 702 is used to perform a downward step simulation of the valve according to the actual working condition constraints to obtain the downward step simulation result.
[0150] Figure 8 is a schematic diagram of the structure of a device for predicting the primary frequency regulation margin of a thermal power unit provided in an embodiment of the present application. Figure 5 Based on the embodiment, further, Figure 8As shown, the primary frequency regulation margin prediction device for a thermal power unit provided in the present application also includes:
[0151] The experimental data acquisition unit 801 is used to acquire the primary frequency modulation experimental data;
[0152] A data preprocessing unit 802 is used to preprocess the primary frequency modulation experiment data to obtain a training set and a test set;
[0153] The model training unit 803 is used to perform model training using the training set and the test set to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights of the hidden layer output to the output layer.
[0154] The present application provides a method and device for predicting the primary frequency regulation margin of a thermal power unit, by obtaining the steam turbine power, main steam parameters and valve opening; generating a constraint model of the valve step amount according to the actual operating condition constraints obtained; determining the valve step amount using the constraint model; inputting the steam turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data, and realizes the prediction of the maximum output response of the primary frequency regulation of the unit based on the current operating condition of the generator unit, providing a decision-making basis for power plant operators and power grid dispatching staff. Among them, by generating a constraint model of the valve step amount according to the actual operating condition constraints obtained; and determining the valve step amount using the constraint model, the accuracy of the valve step amount is improved, thereby improving the accuracy of the prediction of the primary frequency regulation margin of the unit.
[0155] Fig. 9 is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application, such as Fig. 9 As shown, the electronic device may include: a processor 901, a communication interface 902, a memory 903 and a communication bus 904, wherein the processor 901, the communication interface 902 and the memory 903 communicate with each other through the communication bus 904. The processor 901 may call the logic instructions in the memory 903 to execute the following method: obtain the steam turbine power, the main steam parameters and the valve opening; generate a constraint model of the valve step amount according to the obtained actual working condition constraints; determine the valve step amount by using the constraint model; input the steam turbine power, the main steam parameters, the valve opening and the valve step amount into the pre-trained frequency regulation margin prediction model to obtain the frequency regulation margin prediction result; the frequency regulation margin prediction model is trained by using the obtained primary frequency regulation experimental data.
[0156] In addition, the logic instructions in the above-mentioned memory 903 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk.
[0157] The present embodiment discloses a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining turbine power, main steam parameters and valve opening; generating a constraint model of valve step amount according to the obtained actual operating condition constraints; determining the valve step amount by using the constraint model; inputting the turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the obtained primary frequency regulation experimental data.
[0158] The present embodiment provides a computer-readable storage medium, which stores a computer program. The computer program enables the computer to execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining turbine power, main steam parameters and valve opening; generating a constraint model of valve step amount according to the obtained actual operating condition constraints; determining the valve step amount by using the constraint model; inputting the turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the obtained primary frequency regulation experimental data.
[0159] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0160] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0161] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0162] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0163] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0164] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.
Claims
1. A method for predicting primary frequency regulation margin of a thermal power unit, characterized in that: include: Obtain steam turbine power, main steam parameters and valve opening; Generate a constraint model of valve step quantity according to the acquired actual working condition constraints; Determining a valve step amount using the constraint model; The steam turbine power, the main steam parameters, the valve opening and the valve step amount are input into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using acquired primary frequency regulation experimental data.
2. The method for predicting primary frequency regulation margin of a thermal power unit according to claim 1, characterized in that: The step of generating a constraint model for the valve step amount according to the acquired actual working condition constraint comprises: Perform valve step simulation according to actual working condition constraints and obtain simulation results; The constraint model is generated according to the actual working condition constraints and the simulation results.
3. The method for predicting primary frequency regulation margin of a thermal power unit according to claim 2, characterized in that: The simulation results include upward step simulation results and downward step simulation results; The valve step simulation according to the actual working condition constraints includes: Performing a valve upward step simulation according to actual working condition constraints to obtain the upward step simulation result; A valve downward step simulation is performed according to actual working condition constraints to obtain the downward step simulation result.
4. The method for predicting primary frequency regulation margin of a thermal power unit according to claim 1, characterized in that: The steps for pre-training the frequency modulation margin prediction model include: Get the data of a frequency modulation experiment; Preprocessing the primary frequency modulation experimental data to obtain a training set and a test set; The training set and the test set are used to perform model training to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights of hidden layer output to output layer.
5. A device for predicting primary frequency regulation margin of a thermal power unit, characterized in that: include: Actual data acquisition unit, used to obtain turbine power, main steam parameters and valve opening; A constraint model generating unit, used for generating a constraint model of a valve step amount according to the acquired actual working condition constraints; A valve step amount determination unit, used for determining the valve step amount using the constraint model; The frequency regulation margin prediction unit is used to input the turbine power, the main steam parameters, the valve opening and the valve step amount into a pre-trained frequency regulation margin prediction model to obtain a frequency regulation margin prediction result; the frequency regulation margin prediction model is trained using the acquired primary frequency regulation experimental data.
6. The device for predicting primary frequency regulation margin of a thermal power unit according to claim 5, characterized in that: The constraint model generating unit comprises: A simulation result generation module is used to perform valve step simulation according to actual working condition constraints to obtain simulation results; A constraint model generation module is used to generate the constraint model according to the actual working condition constraints and the simulation results.
7. The device for predicting primary frequency regulation margin of a thermal power unit according to claim 6, characterized in that: The simulation result generation module comprises: An upward step simulation submodule, used to perform an upward step simulation of the valve according to actual working condition constraints to obtain the upward step simulation result; The downward step simulation submodule is used to perform a downward step simulation of the valve according to actual working condition constraints to obtain the downward step simulation result.
8. The device for predicting primary frequency regulation margin of a thermal power unit according to claim 5, characterized in that: Also includes: An experimental data acquisition unit, used for acquiring primary frequency modulation experimental data; A data preprocessing unit, used to preprocess the primary frequency modulation experiment data to obtain a training set and a test set; The model training unit is used to perform model training using the training set and the test set to generate a frequency modulation margin prediction model; the frequency modulation margin prediction model is implemented based on random initialization of hidden layer parameters and solving the weights of the hidden layer output to the output layer.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.
11. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.
Citation Information
Patent Citations
Thermal power generating unit primary frequency modulation power response performance prediction method and system
CN113031565A
Method, device and equipment for predicting primary frequency modulation regulation capacity of thermal power generating unit and medium
CN115099486A
Method and device for analyzing electric power step response simulation identification error
CN115374555A
Method and device for predicting primary frequency modulation capability of thermal power generating unit based on improved SVM (Support Vector Machine)
CN117094425A
Small-disturbance primary frequency modulation control system for thermal power generating unit
WO2023241406A1
