Scheduling side optimization method of power grid safety control system and computer equipment

Abstract

The invention belongs to the technical field of power systems, and particularly relates to a dispatching side optimization method of a power grid safety control system and computer equipment. The method comprises the following steps: S1, acquiring real-time operation data of a power grid; s2, establishing a power grid dynamic operation model; the power grid dynamic operation model comprises a state space equation taking power grid operation state parameters as state variables; updating the state space equation in real time according to the real-time operation data by adopting a Kalman filtering algorithm; s3, taking the dynamic operation model of the power grid as a constraint condition to construct an optimal scheduling model, and solving the optimal scheduling model to obtain the output of each new energy power generation unit and the input or output power of each energy storage device in the optimal power grid; and S4, executing the optimal output of each new energy power generation unit and the optimal input or output power of each energy storage device. According to the method, the technical problem that the optimization result is unreasonable due to the fact that the model parameters of the power grid operation model constructed in the prior art are fixed and cannot adapt to the rapid dynamic change of the power grid is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power systems, and in particular relates to a dispatching side optimization method and computer equipment for a power grid safety control system. Background Art

[0002] The distribution network refers to the power grid that receives electric energy from the transmission network or regional power plants and distributes it locally or step by step according to voltage to various users through distribution facilities. It is composed of overhead lines, cables, towers, distribution transformers, disconnectors, reactive power compensators and some ancillary facilities, and plays an important role in distributing electric energy in the power grid.

[0003] With the construction of the UHV grid, the "strong DC and weak AC" feature is obvious, and the risk of stable operation of the power grid is increasing. In order to ensure the safe, stable and economical operation of the power grid, a large number of power grid safety and stability control systems are configured at the transmission and receiving end power grids, that is, the safety and stability control system of the second safety line of the power grid. At present, a typical power grid safety and stability control system includes a security control substation and several security control operation devices. The substation collects plant and station information, identifies the operation mode, and realizes control strategy retrieval. The execution device collects information, fault identification, and realizes local control.

[0004] At present, the new energy consumption, power flow and voltage problems of regional power grids with high penetration of distributed energy access usually show the characteristics of full network and decentralization. Only investigating a single source and adjusting a single-end device, on the one hand, the adjustment capacity is limited and cannot effectively solve the system problems in this operating scenario; on the other hand, with the continuous increase in the scale of distribution networks, there is still a lack of research on the corresponding control framework and the interaction mode between different levels.

[0005] When optimizing power grid dispatching, it is generally necessary to establish a power grid operation model to reflect the actual operating status of the power grid. The power grid operation models established in the prior art are mostly offline and static. For example, the Chinese invention patent application with application publication number CN112350380A and application publication date 2021.02.09 discloses a method and system for constructing a dispatching side operation model of a power grid safety and stability control system. The construction of the security and control operation model in this method includes the following steps: constructing an offline information model; constructing a quasi-real-time dispatching model. The security and control operation model constructed by it is offline and static, and cannot adapt to the rapid dynamic changes of the power grid, which affects the optimization results when performing optimal dispatching, resulting in unreasonable optimization results. Summary of the invention

[0006] The purpose of the present invention is to provide a dispatching side optimization method and computer equipment for a power grid safety control system to solve the technical problem that the model parameters of the power grid operation model constructed in the prior art are fixed and cannot adapt to the rapid dynamic changes of the power grid, resulting in unreasonable optimization results.

[0007] In order to solve the above technical problems, the present invention provides a technical solution for a dispatching side optimization method of a power grid security control system: a dispatching side optimization method of a power grid security control system, the method comprising:

[0008] S1. Obtain real-time operation data of the power grid;

[0009] S2. Establish a dynamic operation model of the power grid;

[0010] The power grid dynamic operation model includes a state space equation with power grid operation state parameters as state variables; and uses a Kalman filter algorithm to update the state variables of the state space equation in real time according to real-time operation data;

[0011] S3. Constructing an optimization dispatching model with the dynamic operation model of the power grid as a constraint condition, and solving the optimization dispatching model to obtain the optimal output of each new energy power generation unit in the power grid and the input or output power of each energy storage device;

[0012] S4. Execute S3 to obtain the optimal output of each new energy power generation unit and the input or output power of each energy storage device.

[0013] The beneficial effect of the above technical solution is that the technical solution of the dispatching side optimization method of a power grid safety control system of the present invention belongs to an improved invention. When establishing a dynamic operation model of the power grid, the present invention adopts a Kalman filter algorithm to update the model parameters in real time to achieve a better dynamic tracking effect. This real-time update capability greatly improves the actual power grid operation accuracy represented by the dynamic operation model of the power grid, provides a more accurate dynamic operation model, and makes the final optimization result more reasonable. The present invention solves the technical problem that the model parameters of the power grid operation model constructed in the prior art are fixed and cannot adapt to the rapid dynamic changes of the power grid, resulting in unreasonable optimization results.

[0014] Furthermore, S4 also includes: during the execution process, using a reinforcement learning algorithm to adjust the grid operation status parameters in real time so that the grid meets the conditions for stable operation; the conditions for stable operation include the grid power supply and demand balance, the power transmission loss does not exceed the preset transmission loss threshold or the grid load fluctuation does not exceed the preset load fluctuation threshold.

[0015] Furthermore, the value function in the reinforcement learning algorithm is adjusted according to the following formula:

[0016]

[0017] Among them, s t represents the real-time operating state parameters of the power grid at time t, a t represents the control decision, α represents the learning rate, r trepresents the immediate reward, γ represents the discount factor, a' represents all possible actions at the next moment, Q(s t , a t ) indicates that it is in the running state t Take action a t The value function of Q(s t+1 , a t+1 ) indicates that it is in the running state t+1 Take action a t+1 The value function of Q(s t+1 , a') indicates that it is in the running state s t+1 The value function of taking action a'.

[0018] Furthermore, the optimization goal of the optimization scheduling model in S3 is to maximize the consumption of new energy in the region, and the objective function is:

[0019]

[0020] Where I represents the set of new energy power generation units, c i is the contribution coefficient of the i-th renewable energy power generation unit, x i is the output of the i-th renewable energy power generation unit; J represents the energy storage device set, d j is the benefit coefficient of the jth energy storage device, y j is the input or output power of the jth energy storage device.

[0021] Furthermore, the grid operation status parameter includes at least one of node voltage, line flow, grid frequency, harmonic current, harmonic voltage, and three-phase imbalance degree.

[0022] Furthermore, S2 also includes establishing a power generation prediction model for a new energy power generation unit, the input parameters of the power generation prediction model for the new energy power generation unit include at least one of solar radiation intensity, ambient temperature, ambient humidity, sunshine duration, photovoltaic module performance, degree of contamination of photovoltaic panels and inverter parameter performance on the output side of the new energy power generation unit; the constraint conditions of the optimization scheduling model in S3 also include the power generation prediction model.

[0023] Furthermore, the power generation prediction model is a least squares support regression vector machine model, and its training formula is as follows:

[0024]

[0025] sty i -w T Φ(x i )-b≤ε+ζ i

[0026]

[0027] i=1,2,...,N

[0028] Among them, w is the weight vector, b is the bias, Φ(x i ) is the sample x i The nonlinear mapping of y i is the true value of the i-th sample, ε is the error threshold, C is the penalty factor, ζ i and i * is the slack variable and N is the number of samples.

[0029] Furthermore, S2 also includes establishing a load forecasting model for load forecasting, the input parameters of the load forecasting model include meteorological data, and the load forecasting model adopts a deep learning model; the constraint conditions of the optimization scheduling model in S3 also include the load forecasting model.

[0030] Furthermore, the constraints of the optimization scheduling model also include at least one of the following: the transmission power of each line in the power grid is not greater than the maximum power allowed to be transmitted by the corresponding line, the operating voltage of each device in the power grid is not greater than the rated voltage of the corresponding device, the voltage of each node in the power grid is not greater than the maximum value threshold of the corresponding node voltage and not less than the minimum value threshold of the corresponding node voltage, the power grid frequency is not greater than the maximum value threshold of the frequency and not less than the minimum value threshold of the frequency, and the load of each transformer does not exceed the rated capacity of the corresponding transformer.

[0031] The present invention also provides a technical solution for a computer device: a computer device, comprising a processor, wherein the processor is used to execute a computer program to implement the steps of the dispatching side optimization method of the power grid security control system as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The present invention is a flowchart of an embodiment of a method for optimizing the dispatching side of a power grid security control system according to the present invention. DETAILED DESCRIPTION

[0033] The present invention adopts the Kalman filter algorithm to update the model parameters in real time when establishing the dynamic operation model of the power grid. This real-time update capability greatly improves the accuracy of the actual power grid operation represented by the dynamic operation model of the power grid, making the final optimization result more reasonable. The present invention solves the technical problem that the model parameters of the power grid operation model constructed in the prior art are fixed and cannot adapt to the rapid dynamic changes of the power grid, resulting in unreasonable optimization results.

[0034] Implementation example of dispatching side optimization method for power grid security control system:

[0035] A dispatching side optimization method for a power grid security control system, such as Figure 1As shown, the following steps are included:

[0036] S1. Obtain real-time operation data of the power grid.

[0037] This embodiment collects real-time operation data through the power grid dispatching automation system, and the real-time operation data includes but is not limited to node voltage, line flow, load change, distributed power output and power grid equipment status;

[0038] Specifically, the grid operation data is collected in real time through sensors, measuring equipment and grid dispatching automation systems deployed at each node. The real-time operation data includes node voltage V(n), line flow P / Q (kW / kVar), load change ΔL(kW), distributed power output DPS(kW), and grid equipment status information E (including equipment operation status, temperature, load conditions, etc.). Specifically, the grid equipment mainly includes power generation equipment, power transmission and transformation equipment, and power distribution equipment on the transmission and distribution side, such as transformer load and switchgear status.

[0039] Real-time data is transmitted to the data center in real time through sensors, smart meters and SCADA systems. For example, the acquisition formula of a node voltage V can be simplified to: Vn(t) = f(Sensor(t)), where Sensor(t) represents the sensor reading of the nth node at time t, and f() represents the data processing function, such as filtering, calibration and other operations.

[0040] S2. Establish a dynamic operation model of the power grid.

[0041] The dynamic operation model of the power grid includes a state space equation with the power grid operation state parameters as state variables; and a Kalman filter algorithm is used to update the state variables of the state space equation in real time according to real-time operation data; the power grid operation state parameters include at least one of node voltage, line flow, power grid frequency, harmonic current, harmonic voltage, and three-phase imbalance degree.

[0042] The basic equation of Kalman filtering is:

[0043] x(k)=F(k-1)x(k-1)+B(k)u(k-1)+w(k-1)

[0044] z(k)=H(k)x(k)+v(k)

[0045] Among them, x represents the state vector, that is, the power grid operation state vector; F is the state transfer matrix; B is the control input matrix; u is the control input; w is the process noise; z is the observation vector; H is the observation matrix; v is the observation noise.

[0046] Kalman filtering is a recursive filtering algorithm based on statistical estimation. It recursively estimates the state of the system based on observed data and system models.

[0047] This embodiment also establishes a power generation prediction model for each new energy unit. The power generation prediction model adopts a least squares support vector regression (LSSVR) model, and its model training formula is as follows:

[0048]

[0049] sty i -w T Φ(x i )-b≤ε+ζ i

[0050]

[0051] i=1,2,...,N

[0052] Among them, w is the weight vector, b is the bias, Φ(x i ) is the sample x i The nonlinear mapping of y i is the true value of the i-th sample, ε is the error threshold, C is the penalty factor, ζ i and i * are slack variables, and N is the number of samples.

[0053] In this embodiment, x i Specifically, it refers to a series of input parameter samples used in the prediction of renewable energy power generation. These parameters are crucial for accurately predicting the power generation output of renewable energy (such as wind power and solar power). The power generation output of renewable energy is highly dependent on natural conditions and other external factors. Therefore, the prediction model needs to comprehensively consider multiple variables, such as solar radiation intensity, ambient temperature, ambient humidity, sunshine duration, photovoltaic module performance, degree of contamination of photovoltaic panels, inverter parameter performance on the output side of renewable energy power generation units, etc.

[0054] This embodiment also establishes a load forecasting model to perform load forecasting based on real-time data. The forecasting model adopts a time series forecasting method based on deep learning to improve the accuracy of load forecasting. The forecasting model adopts a time series forecasting method based on deep learning to improve the accuracy of load forecasting.

[0055] The load forecasting model is mainly established based on the following types of data: 1. Time series historical data, load data over a period of time, reflecting the pattern of load changes over time, including daily load curves, weekly load curves, and seasonal changes; 2. Meteorological data: Meteorological conditions such as temperature, humidity, and wind speed have a significant impact on certain types of loads (such as air conditioning loads). For example, high temperatures may lead to an increase in cooling loads; 3. Holidays and special events: Holidays, large-scale events, factory rest days, etc. can cause significant changes in load patterns.

[0056] S3. Construct an optimization dispatching model with the dynamic operation model of the power grid as a constraint condition, and solve the optimization dispatching model to obtain the optimal output of each new energy unit power generation unit and the input or output power of each energy storage device.

[0057] Once the model parameters are updated in real time through the Kalman filter algorithm, the model can be applied to the optimization of the grid security control strategy. Specifically, the mixed integer linear programming (MILP) method is used to maximize the consumption of new energy while maintaining grid stability. The standard form of the MILP problem is as follows:

[0058] min c T x

[0059] stAx=b

[0060] x l ≤x≤x u

[0061] x j ∈Z,j∈J

[0062] Where c is the objective function coefficient vector, x is the decision variable vector, A and b define the linear relationship constraints, and x l and x u are the lower and upper bounds of the decision variables respectively, and J is a set of integer variables.

[0063] The optimization goal of MILP is to maximize the consumption of renewable energy while maintaining grid stability, which constitutes the core of the MILP problem. Based on this, the objective function can be summarized as maximizing the total output of renewable energy (assuming it is a part of the decision variables) while satisfying all grid stability and security constraints. If the decision variable x contains the output of each renewable energy source and the power of each energy storage device, whether the load side parameters are included depends on the specific problem setting. In practical applications, the load forecast results can indirectly affect the decision process, such as as an exogenous condition input.

[0064] In this embodiment, the goal of the MILP model is to maximize the consumption of new energy while ensuring the stable operation of the power grid. The objective function (maximizing the consumption of new energy) is:

[0065]

[0066] Where I represents the set of new energy power generation units, c i is the contribution coefficient of the i-th new energy unit, x i is the output of the i-th new energy unit; J represents the set of energy storage devices, d j is the benefit coefficient of the jth energy storage device, y j is the charge and discharge power of the jth energy storage device. This objective function aims to reflect the use of renewable energy for power generation as much as possible while ensuring the stability of the power grid.

[0067] The constraints of the optimization scheduling model also include that the transmission power of each line of the power grid is not greater than the maximum power allowed to be transmitted by the corresponding line, the working voltage of each device in the power grid is not greater than the rated voltage of the corresponding device, the voltage of each node in the power grid is not greater than the maximum value threshold of the corresponding node voltage and not less than the minimum value threshold of the corresponding node voltage, the power grid frequency is not greater than the maximum value threshold of the frequency and not less than the minimum value threshold of the corresponding node voltage, the load of each transformer does not exceed the rated capacity of the corresponding transformer, each transformer is not overloaded, and the above-mentioned establishment of power generation prediction model, load prediction model and probability prediction model.

[0068] S4. Execute S3 to obtain the optimal output of each new energy power generation unit and the input or output power of each energy storage device.

[0069] In order to better adapt to the changes in the power grid environment, the reinforcement learning algorithm is introduced to dynamically adjust the model parameters. The Q-learning algorithm is used to update the control strategy of the model:

[0070] 1. Definition of reinforcement learning environment: The power grid system is abstracted as a Markov decision process (MDP), with state s t Including the real-time operation status of the power grid (such as node voltage, line flow, etc.), action a t Represents control decisions (such as adjusting generator output, changing reactive power compensation, etc.).

[0071] 2. Update rules: Use the Q-learning algorithm to update the Q-table (state-action value function), update the Q-value after each decision, and adjust the value function according to the following formula:

[0072]

[0073] Among them, s t represents the real-time operating state parameters of the power grid at time t (such as node voltage, line flow, etc.), a t represents the control decision (such as adjusting the generator output, changing the reactive power compensation, etc.), α represents the learning rate, and r trepresents the immediate reward, γ represents the discount factor, a' represents all possible actions at the next moment, Q(s t , a t ) indicates that it is in the running state t Take action a t The value function of Q(s t+1 , a t+1 ) indicates that it is in the running state t+1 Take action a t+1 The value function of Q(s t+1 , a') indicates that it is in the running state s t+1 The value function of taking action a'.

[0074] After the value function is updated, it is consistent with the updated value function Q(s t+1 , a t+1 ) corresponds to the real-time operation status of the power grid s t+1 As the update target of the power grid operation status, the power grid operation status is updated to s t+1 .

[0075] Reinforcement learning algorithms can gradually explore optimal control strategies in unknown environments and dynamically adjust model parameters to adapt to changes in power grid operation, thereby improving the robustness and adaptability of the entire system.

[0076] Computer equipment example:

[0077] A computer device includes a processor, the processor is used to execute a computer program to implement the steps of the dispatching side optimization method of the power grid security control system as described above. Specifically, the dispatching side optimization method of the power grid security control system has been introduced in sufficient detail in the above-mentioned dispatching side optimization method embodiment of the power grid security control system, and will not be repeated here.

[0078] The present invention has the following characteristics:

[0079] The present invention realizes the accurate collection of all-round real-time data such as node voltage, line flow, load change, distributed power output and power grid equipment status through the power grid dispatching automation system, which helps to more accurately reflect the actual operation status of the power grid and improve the response speed and control accuracy of the system; the multivariable dynamic modeling technology is used, combined with the improved Kalman filter algorithm to update the model parameters in real time, which solves the problem that the traditional model parameters are fixed and cannot adapt to the rapid dynamic changes of the power grid. This real-time update capability greatly improves the prediction accuracy and adaptability of the model, which is crucial for the safe and stable operation of the power grid. In traditional offline and static models, the model parameters are determined at the beginning of model construction, and it is assumed that these parameters remain unchanged during the entire operation. In the dynamic model, the state variables change with time, and can be updated according to real-time data through methods such as the Kalman filter algorithm, so as to achieve a more accurate estimation of the state variables to adapt to the dynamic changes of the power grid, thereby improving the accuracy and applicability of the model. By applying the constructed dynamic operation model to the power grid safety control strategy, the dispatching strategy of the power grid can be optimized in real time, including voltage regulation, power flow distribution, etc., so as to effectively prevent power grid overload, ensure power quality, and improve the reliability and stability of the entire power grid system; the reinforcement learning algorithm is introduced to dynamically adjust the model parameters, which is an intelligent adaptive optimization method. When facing changes in the power grid environment and load, the system can autonomously learn and adjust the control strategy, which is more adaptable to complex and unpredictable operating conditions than static preset control methods.

[0080] Through reinforcement learning algorithms, the system can accumulate experience from continuous operations and gradually optimize and improve its own control strategies, which not only reduces the need for human intervention but also enables the discovery and application of optimal control schemes, thereby significantly improving the operating efficiency and safety of the system.

[0081] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments without creative work, or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A dispatching side optimization method for a power grid security control system, characterized in that: The method comprises the following steps: S1. Obtain real-time operation data of the power grid; S2. Establish a dynamic operation model of the power grid; The power grid dynamic operation model includes a state space equation with power grid operation state parameters as state variables; and uses a Kalman filter algorithm to update the state variables of the state space equation in real time according to real-time operation data; S3, constructing an optimization dispatching model with the dynamic operation model of the power grid as a constraint condition, and solving the optimization dispatching model to obtain the optimal output of each new energy power generation unit in the power grid and the input or output power of each energy storage device; S4. Execute S3 to obtain the optimal output of each new energy power generation unit and the input or output power of each energy storage device.

2. The dispatching side optimization method of the power grid security control system according to claim 1 is characterized in that: The S4 also includes: during the execution process, using a reinforcement learning algorithm to adjust the power grid operation status parameters in real time so that the power grid meets the conditions for stable operation; the conditions for stable operation include the balance of power supply and demand of power grid electricity, the transmission loss of power energy does not exceed a preset transmission loss threshold, or the power grid load fluctuation does not exceed a preset load fluctuation threshold.

3. The dispatching side optimization method of the power grid security control system according to claim 2 is characterized in that: The value function in the reinforcement learning algorithm is adjusted as follows: Among them, s t represents the real-time operating state parameters of the power grid at time t, a t represents the control decision, α represents the learning rate, r t represents the immediate reward, γ represents the discount factor, a' represents all possible actions at the next moment, Q(s t , a t ) indicates that it is in the running state t Take action a t The value function of Q(s t+1 , a t+1 ) indicates that it is in the running state t+1 Take action a t+1 The value function of Q(s t+1 , a') indicates that it is in the running state s t+1 The value function of taking action a'.

4. The dispatching side optimization method of the power grid security control system according to claim 1, characterized in that: The optimization goal of the optimization scheduling model in S3 is to maximize the consumption of new energy in the region, and the objective function is: Where I represents the set of new energy power generation units, c i is the contribution coefficient of the i-th renewable energy power generation unit, x i is the output of the i-th renewable energy power generation unit; J represents the energy storage device set, d j is the benefit coefficient of the jth energy storage device, y j is the input or output power of the jth energy storage device.

5. The dispatching side optimization method of a power grid security control system according to any one of claims 1 to 4, characterized in that: The grid operation status parameter includes at least one of node voltage, line flow, grid frequency, harmonic current, harmonic voltage, and three-phase imbalance degree.

6. The dispatching side optimization method of the power grid security control system according to claim 1, characterized in that: S2 also includes establishing a power generation prediction model for a new energy power generation unit, the input parameters of the power generation prediction model for the new energy power generation unit include at least one of solar radiation intensity, ambient temperature, ambient humidity, sunshine duration, photovoltaic module performance, degree of contamination of photovoltaic panels, and inverter parameter performance on the output side of the new energy power generation unit; the constraint conditions of the optimization scheduling model in S3 also include the power generation prediction model.

7. The dispatching side optimization method of the power grid security control system according to claim 6, characterized in that: The power generation prediction model is a least squares support regression vector machine model, and its training formula is as follows: style i -w T Φ(x i )-b≤e+z i i=1,2,...,N Among them, w is the weight vector, b is the bias, Φ(x i ) is the sample x i The nonlinear mapping of y i is the true value of the i-th sample, ε is the error threshold, C is the penalty factor, ζ i and i * is the slack variable and N is the number of samples.

8. The dispatching side optimization method of the power grid security control system according to claim 1, characterized in that: S2 also includes establishing a load forecasting model for load forecasting, the input parameters of the load forecasting model include meteorological data, and the load forecasting model adopts a deep learning model; the constraint conditions of the optimization scheduling model in S3 also include the load forecasting model.

9. The dispatching side optimization method of the power grid security control system according to claim 1, 6 or 8, characterized in that: The constraints of the optimization scheduling model also include at least one of the following: the transmission power of each line of the power grid is not greater than the maximum power allowed to be transmitted by the corresponding line, the operating voltage of each device in the power grid is not greater than the rated voltage of the corresponding device, the voltage of each node in the power grid is not greater than the maximum value threshold of the corresponding node voltage and not less than the minimum value threshold of the corresponding node voltage, the power grid frequency is not greater than the maximum value threshold of the frequency and not less than the minimum value threshold of the frequency, and the load of each transformer does not exceed the rated capacity of the corresponding transformer.

10. A computer device comprising a processor, characterized in that: The processor is used to execute a computer program to implement the steps of the dispatching side optimization method of a power grid security control system as described in any one of claims 1 to 9.

Images