Cluster analysis model and modeling method for multi-zone-area intelligent terminal stations

Abstract

The invention belongs to the technical field of power system automation, and particularly relates to a cluster analysis model and a modeling method of a multi-zone-area intelligent terminal station. The model can effectively improve the data analysis efficiency of the intelligent terminal station in the transformer area, and support is provided for operation scheduling and control decision making of a load aggregator; according to the method, aggregation of a plurality of transformer area intelligent terminal stations is converted into parallel operation of a plurality of virtual synchronous units, and only overall external characteristics of the plurality of transformer area intelligent terminal stations represented at the PCC need to be clear, and specific parameters in the transformer area intelligent terminal stations do not need to be known, so that the order can be reduced and the solving speed can be improved while the model precision is kept during cluster modeling; the method comprises the following three steps: (1) establishing a unit model of a transformer area intelligent terminal station, (2) establishing a cluster aggregation model of the multi-transformer area intelligent terminal station, and (3) determining network-related power characteristics of the cluster aggregation model of the multi-transformer area intelligent terminal station. The method has the advantages that (1) the data analysis efficiency is improved, (2) the model precision is improved, and (3) the operation decision is effectively supported.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system automation, and particularly relates to a cluster analysis model and modeling method for a multi-substation intelligent terminal station. Background Art

[0002] With the deepening of the power market reform and the rapid development of renewable energy, load aggregators will play an increasingly important role in the future power system, and will develop towards specialization, scale, and intelligence, playing a huge role in building a clean, low-carbon, safe, and efficient energy system. Load aggregators are important participants in the power market. Their main role is to integrate dispersed and adjustable power load resources and participate in power market transactions as a virtual power plant, providing flexibility and ancillary services for the power system.

[0003] Load aggregators build important nodes that connect the distribution network and substation users through multiple substation intelligent terminal stations in the service area, effectively controlling the information flow and energy flow, and realizing the following functions:

[0004] 1. Improve the flexibility of the power system: The traditional power system mainly relies on power generation side regulation to meet load demands, while load aggregators can aggregate demand-side resources through a large number of substation intelligent terminal stations to achieve flexible load regulation and improve the overall flexibility of the power system. When the power supply is tight, load aggregators can reduce the aggregated load to relieve the grid pressure; when the power supply is excessive, load aggregators can increase the aggregated load to absorb the excess power.

[0005] 2. Provide ancillary services: Load aggregators can aggregate demand-side resources through substation intelligent terminal stations to participate in power market ancillary services (frequency regulation, reserve, black start, etc.), providing guarantee for the safe and stable operation of the power system. Load aggregators can quickly adjust the aggregated load through substation intelligent terminal stations to participate in power system frequency regulation and maintain system frequency stability.

[0006] 3. Reduce the electricity cost of users: Load aggregators can participate in power market transactions through substation intelligent terminal stations to strive for more favorable electricity prices for users. Substation intelligent terminal stations can increase user electricity consumption when the electricity price is low and reduce user electricity consumption when the electricity price is high, thereby reducing the overall electricity cost of users.

[0007] 4. Promote the consumption of renewable energy: Load aggregators can aggregate adjustable loads through substation intelligent terminal stations to cooperate with the volatility of local renewable energy generation and promote the consumption of renewable energy. When the output of wind power and photovoltaic power generation is large, substation intelligent terminal stations can increase adjustable electricity loads (fast charging of electric vehicles, energy storage charging, etc.) to absorb the excess renewable energy power.

[0009] The intelligent terminal station of the substation area is an effective means for the load aggregator to integrate dispersed and adjustable power load resources. However, with the explosive growth of the number of active substation areas, a large amount of state data of intelligent terminal stations in the substation area, coordinated control strategies, and interactive coupling effects have brought huge challenges to the dispatching operation analysis of the load aggregator. Therefore, the hierarchical and zonal control of aggregated resources has become an inevitable trend. The traditional modeling method for a single intelligent terminal station in the substation area is difficult to meet the requirements of large-scale application analysis and flexible management and control. Therefore, there is an urgent need for a method that can cluster-model multiple intelligent terminal stations in the substation area to improve the data analysis efficiency and provide operation decision support for the load aggregator to effectively manage a large number of power users. Summary of the Invention

[0010] The technical problem to be solved by the present invention is to provide a cluster analysis model and modeling method for multiple intelligent terminal stations in the substation area, which can effectively improve the data analysis efficiency of the intelligent terminal stations in the substation area and provide support for the operation dispatching and control decision-making of the load aggregator.

[0011] The grid-following intelligent terminal station in the substation area cannot construct the voltage in the AC distribution network as a voltage source like a traditional synchronous machine, and it is difficult to meet the support requirements for the distribution network voltage. In order to enable the intelligent terminal station in the substation area to autonomously form the electric potential at the end of the distribution network, the grid-forming grid-connection technology with the characteristics of a virtual synchronous machine is adopted.

[0012] The grid-forming grid-connection technology adopted by the intelligent terminal station in the substation area is significantly different from the traditional grid-following grid-connection technology. The grid-following grid-connection technology uses a phase-locked loop to measure and lock the phase of the voltage at the Point of Common Coupling (PCC) in real time and control the current output, and passively accesses the power grid in the form of an additional current source. While the grid-forming type actively forms a voltage output at the PCC by simulating the power angle characteristics of a synchronous machine and controlling the virtual rotation angle according to the active power generation. In the present invention, the aggregation of multiple intelligent terminal stations in the substation area is transformed into the parallel operation of multiple virtual synchronous machine sets.

[0013] The order and accuracy of the model are contradictory indicators to the solution speed of the model. How to accelerate the model solution speed while maintaining the model accuracy is the key to the aggregated modeling of multiple intelligent terminal stations in the substation area. For an active distribution network, the intelligent terminal station in the substation area needs to meet the grid standards to access the grid. As long as the overall external characteristics represented by multiple intelligent terminal stations at the PCC are clear, without having to understand their specific internal parameters, the order can be reduced and the solution speed can be improved while maintaining the model accuracy during cluster modeling.

[0014] To solve the above technical problems, the present invention adopts the following technical solutions:

[0015] A cluster analysis model and modeling method for multiple intelligent terminal stations in the substation area, characterized by including the following steps:

[0016] 1. Establish the unit model of the substation area intelligent terminal station

[0017] The unit control of the substation area intelligent terminal station adopts virtual synchronous machine control, which can meet the requirements of different application scenarios (such as droop control, power synchronization control, etc.). The unit model is represented by the swing equation:

[0018]

[0019] In the formula, M i is the virtual inertia constant of the substation area intelligent terminal station i, D i is the damping constant, K d,i is the droop coefficient, P i is the output power of the substation area intelligent terminal station, P i * is the command power of the substation area intelligent terminal station i, ω i is the virtual angular frequency of the substation area intelligent terminal station, ω g is the grid angular frequency, ω * is the rated angular frequency.

[0020] Taking the phase of the grid voltage phase A as the reference, the voltage output phase of the substation area intelligent terminal station at the PCC point is:

[0021]

[0022] The output voltage amplitude of the substation area intelligent terminal station is E i :

[0023] E i = V * + K V,i (V * - V pcc ) (3)

[0024] In the formula, V * is the rated voltage, V pcc is the voltage at the PCC point, K v,i is the voltage regulation coefficient.

[0025] Formulas (1) to (3) are the control outer loop (including active power regulation and voltage regulation) and grid connection synchronization of the substation area intelligent terminal station. The input of the control outer loop of the substation area intelligent terminal station is the grid frequency and voltage amplitude, and the output E i and δ i are used as the reference values for the inner loop control. The inner loop control is voltage and current control, which makes the output voltage of the converter track its reference value.

[0026] Ignoring the line resistance and assuming the line inductance is l i , then the output power of the substation area intelligent terminal station is:

[0027]

[0028] Equations (4) and (5) constitute the second-order power model of the intelligent terminal station in the distribution area, which is convenient for calculating and analyzing its ability to actively support the grid frequency and voltage.

[0029] 2. Establish a multi-distribution area intelligent terminal station cluster aggregation model

[0030] The intelligent terminal station in the distribution area can be regarded as the parallel connection of N terminal station units. Under dynamic conditions, the output voltages of each unit in the intelligent terminal station in the distribution area are close. That is, the whole area can be regarded as the parallel connection of N voltage sources, and its overall equivalent impedance is:

[0031]

[0032] According to the definition of the central inertia, the aggregated inertia of the multi-distribution area intelligent terminal station is the sum of the inertias of each unit, and its equivalent output frequency is the weight of each output frequency to its inertia:

[0033]

[0034] When the intelligent terminal station in the distribution area is operating stably or adjusting gradually, its phase difference δ i is far from 90°, and its dynamic characteristics can be approximately a second-order linear model. The sum of second-order linear models is still a second-order linear model. Then, the aggregated swing equation of the multi-distribution area intelligent terminal station has the same form as the unit model, as shown in Equation (8), where D a is the aggregated damping coefficient.

[0035]

[0036] The expression of the aggregated damping coefficient D a is:

[0037]

[0038] where Di is the damping coefficient of the i-th intelligent terminal station.

[0039] Mi is the inertia constant of the i-th intelligent terminal station.

[0040] Generally, the control time of the phase-locked loop PLL is short and can be ignored. Then, the aggregated droop coefficient of the multi-distribution area intelligent terminal station is:

[0041]

[0042] 3. Determine the grid-connected power characteristics of the multi-distribution area intelligent terminal station aggregation model

[0043] The total power P a and Qa is:

[0044]

[0045] The total power of the multi - sub - area intelligent terminal station passes through the line inductance l g and is input into the power grid, that is:

[0046]

[0047]

[0048] V is obtained by simultaneously solving equations (13) and (14) pcc The voltage is:

[0049]

[0050] Under various dynamic operating conditions of the distribution network, the changes in the active and reactive power injected by the aggregated sub - area intelligent terminal stations into the PCC will affect the power grid frequency and voltage, and the changing power grid frequency and voltage will in turn affect the active and reactive power outputs of the sub - area intelligent terminal stations.

[0051] Furthermore, for the unit model of the sub - area intelligent terminal station established in step 1, its voltage, current, and power are measurable parameters, while inertia, damping, and inductive reactance (inductance) are unknown parameters and need to be determined through parameter identification. The damping of the multi - sub - area intelligent terminal station consists of the damping and losses of each unit, which is obtained by applying a specific perturbation, measuring the unit response, and then using the swing equation; the inertia of each unit can be obtained by applying a specific perturbation, measuring the unit response, and then using the inertia evaluation method; the impedance of each unit can be obtained by applying a specific perturbation, measuring the unit response, and then using the change in active power flow.

[0052] Beneficial effects

[0053] A cluster analysis model and modeling method for a multi - sub - area intelligent terminal station provided by the present invention have the following advantages:

[0054] 1. Improve data analysis efficiency: By aggregating and modeling the sub - area intelligent terminal stations, the data scale can be effectively reduced, and the data analysis efficiency can be improved.

[0055] 2. Improve model accuracy: By aggregating and modeling the multi - sub - area intelligent terminal stations in the region, the generalization ability and prediction accuracy of the model can be improved, and at the same time, the overall power and voltage regulation ability of the region can be effectively improved.

[0056] 3. Effectively support operation decision - making: The aggregation model can be used for the evaluation of the operation status of the sub - area, the optimization of the power grid power flow, fault early warning, load forecasting, etc., providing support for actively participating in the optimization operation decision - making of the distribution network. Description of the drawings

[0057] Figure 1 This is the circuit topology diagram of the unit system of the intelligent terminal station in a multi-substation area for the cluster analysis model and modeling method provided by the embodiments of the present invention. The power grid supplies power to the loads in the substation area through the intelligent terminal station in the substation area. The intelligent terminal station in the substation area opens the DC bus externally and connects flexible and controllable resources in the substation area, including distributed photovoltaics, DC charging piles, and energy storage systems.

[0058] Figure 2 It is a 34-node distribution network system diagram, where the red nodes are the substations with intelligent terminal stations connected, and the black nodes are ordinary substations. To establish a cluster analysis model for multi-substation intelligent terminal stations, the Figure 2 system is equivalent to the Figure 3 system wiring shown.

[0059] Figure 3 It is the equivalent system wiring diagram of a 34-node distribution network with multi-substation intelligent terminal stations, which can be used to construct a cluster analysis model for multi-substation intelligent terminal stations. Specific implementation manners

[0060] The following further describes the specific implementation manners of the present invention with reference to the accompanying drawings.

[0061] As Figure 1 shown, a cluster analysis model and modeling method for multi-substation intelligent terminal stations provided by the embodiments of the present invention include the following steps:

[0062] 1. Establish a unit model for the intelligent terminal station in the substation area

[0063] The unit control of the intelligent terminal station in the substation area adopts virtual synchronous machine control, which can meet the requirements of different application scenarios (such as droop control, power synchronization control, etc.). The unit model is represented by the swing equation:

[0064]

[0065] In the formula, M i is the virtual inertia constant of the intelligent terminal station i in the substation area, D i is the damping constant, K d,i is the droop coefficient, P i is the output power of the intelligent terminal station in the substation area, P i * is the command power of the intelligent terminal station i in the substation area, ω i is the virtual angular frequency of the intelligent terminal station in the substation area, ω g is the grid angular frequency, ω * is the rated angular frequency.

[0066] Taking the phase of the A-phase of the grid voltage as the reference, the voltage output phase of the intelligent terminal station in the substation area at the PCC point is:

[0067]

[0068] The output voltage amplitude of the substation area intelligent terminal station is E i :

[0069] E i = V * + K V,i (V * - V pcc )

[0070] Where V * is the rated voltage, V pcc is the voltage at the PCC point, and K v,i is the voltage regulation coefficient.

[0071] The above three equations are for the control outer loop (including active power regulation and voltage regulation) and grid connection synchronization of the substation area intelligent terminal station. The input of the control outer loop of the substation area intelligent terminal station is the grid frequency and voltage amplitude, and the output E i and δ i are used as the reference values for the inner loop control. The inner loop control is voltage and current control to make the output voltage of the converter track its reference value.

[0072] Ignoring the line resistance and assuming the line inductance is l i , then the output power of the substation area intelligent terminal station is:

[0073]

[0074] The above two equations constitute the second-order power model of the substation area intelligent terminal station, which is convenient for calculating and analyzing its ability to actively support the grid frequency and voltage.

[0075] 2. Establish a multi-substation area intelligent terminal station cluster aggregation model

[0076] The substation area intelligent terminal station can be regarded as N terminal station units in parallel. Under dynamic conditions, the output voltages of each unit in the substation area intelligent terminal station are close, that is, the whole area can be regarded as N voltage sources in parallel, and its overall equivalent impedance is:

[0077]

[0078] According to the definition of the central inertia, the aggregated inertia of the multi-substation area intelligent terminal stations is the sum of the inertias of each unit, and its equivalent output frequency is the weight of each output frequency to its inertia:

[0079]

[0080] When the substation area intelligent terminal station is operating stably or undergoing gradual adjustment, its phase difference δ iAway from 90°, its dynamic characteristics can be approximated as a second-order linear model, and the sum of second-order linear models is still a second-order linear model. Then, the aggregated swing equation of the multi-substation intelligent terminal station has the same form as the unit model, as shown in the following equation, where D a is the aggregated damping coefficient.

[0081]

[0082] The aggregated damping coefficient D a has the following expression:

[0083]

[0084] where D i is the damping coefficient of the i-th intelligent terminal station.

[0085] M i is the inertia constant of the i-th intelligent terminal station.

[0086] Generally, the control time of the phase-locked loop PLL is short and can be ignored. Then, the aggregated droop coefficient of the multi-substation intelligent terminal station is:

[0087]

[0088] 3. Determine the grid-connected power characteristics of the aggregated model of the multi-substation intelligent terminal station

[0089] The total power P a and Q a collected by each unit of the substation intelligent terminal station at the PCC point is:

[0090]

[0091] The total power of the multi-substation intelligent terminal station is input into the power grid through the line inductance l g , that is:

[0092]

[0093] By combining the above two equations, the V pcc voltage is obtained as:

[0094]

[0095] Under various dynamic operating conditions of the distribution network, the changes in the active and reactive power injected by the aggregated substation intelligent terminal stations into the PCC will affect the power grid frequency and voltage, and the changing power grid frequency and voltage will in turn affect the active and reactive power outputs of the substation intelligent terminal stations.

[0096] Furthermore, for the unit model of the substation intelligent terminal established in step 1, its voltage, current, and power are measurable parameters, while inertia, damping, and inductive reactance (inductance) are unknown parameters, and their values need to be determined through parameter identification. The damping of the multi-substation intelligent terminal consists of the damping and losses of each unit, which can be obtained by applying a specific perturbation, measuring the unit response, and then using the swing equation; the inertia of each unit can be obtained by applying a specific perturbation, measuring the unit response, and then using the inertia evaluation method; the impedance of each unit can be obtained by applying a specific perturbation, measuring the unit response, and then using the change in active power flow.

Claims

1. A cluster analysis model and modeling method for multi-station intelligent terminal stations, characterized in that: The following steps are involved: (1) Establish a unit model of the intelligent terminal station in the substation area; The unit control of the intelligent terminal station in the substation area adopts virtual synchronous machine control, which can adapt to the needs of different application scenarios (such as droop control, power synchronization control, etc.). The unit model is expressed by the swing equation: Taking the phase of grid voltage phase A as reference, the voltage output phase of the intelligent terminal station at the PCC point is: The output voltage amplitude of the intelligent terminal station in the substation area is E i : E i =V * +K V,i (V * -V pcc ) Ignore the line resistance and assume the line inductance is l i , then the output active and reactive power of the intelligent terminal station in the substation area is: (2) Establish a cluster aggregation model of intelligent terminal stations in multiple stations; The intelligent terminal station in the substation area can be regarded as N terminal station units in parallel. Under dynamic conditions, the output voltage of each unit in the intelligent terminal station in the substation area is close, that is, the entire area can be regarded as N voltage sources in parallel, and its overall equivalent impedance is: According to the definition of central inertia, the aggregate inertia of multi-station intelligent terminal stations is the sum of the inertia of each unit, and its equivalent output frequency is the weight of each output frequency to its inertia: The aggregate swing equation of multiple intelligent terminal stations is: Where D a is the aggregate damping coefficient, D a The expression is: Where Di is the damping coefficient of the i-th intelligent terminal station. Mi is the inertia constant of the i-th intelligent terminal station. Generally, the control time of the phase-locked loop (PLL) is short and can be ignored. Then the aggregate droop coefficient of the multi-station intelligent terminal station is: (3) Determine the network-related power characteristics of the multi-area intelligent terminal station aggregation model. The total power P collected by each unit of the intelligent terminal station in the substation area at the PCC point a and Q a for: The total power of multiple intelligent terminal stations is transmitted through the line inductance l g Input into the grid, that is: PCC point voltage V pcc for:

2. The cluster analysis model of multi-station intelligent terminal stations according to claim 1 is characterized in that: The unit model of the intelligent terminal station in the substation area in step 1 is established, and its voltage, current, and power are measurable parameters, while inertia, damping, and inductive reactance (inductance) are unknown parameters, and the values ​​need to be determined through parameter identification. The damping of the multi-area intelligent terminal station is composed of the damping and loss of each unit. By applying a specific disturbance and measuring the unit response, it is obtained by the swing equation; The inertia of each unit can be obtained by applying a specific disturbance, measuring the unit response, and then using the inertia evaluation method; the impedance of each unit can be obtained by applying a specific disturbance, measuring the unit response, and then using the change in active power flow.

Images