Method for estimating line resistance of communication-free power transmission line and compensating droop control coefficient

Abstract

The invention discloses a method for estimating line resistance of a communication-free power transmission line and compensating a droop control coefficient in the technical field of electric power grids, which comprises the following steps of: injecting high-frequency pulse disturbance into a reference value of a current loop controller of a direct current-direct current converter, and sampling voltage and current; calculating a real-time output current based on a Boost converter, and calculating a fluctuation quantity; constructing a proportional-integral link; taking the output quantity of the PI link as an estimated value of the line resistance of the power transmission line of the system, and obtaining a corresponding bus voltage fluctuation quantity; sending the bus voltage fluctuation quantity into a first-order Kalman filter to obtain a mean value; the output of the first-order Kalman filter and the output of the PI link are fed back to the input side of a PI loop together, and one-time closed-loop operation is completed; after the PI link is steady, outputting an estimated value of the line resistance, and compensating the estimated value to a droop control coefficient; the method is higher in estimation precision, and is suitable for a single-bus DC micro-grid system or a multi-bus DC micro-grid system.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and particularly relates to a method for estimating wire resistance and compensating the droop control coefficient. Background Art

[0002] The basic principle of droop control is as shown in (1): , where is the voltage loop reference value, is the nominal voltage, is the output current of the DC-DC converter, is the droop control coefficient, also known as virtual impedance.

[0003] When the DC microgrid system contains multiple parallel converters, as Figure 7 shown, each converter is controlled by droop control, and when its nominal voltage is equal, the proportion of the output power between the converters is constrained by the corresponding droop coefficients.

[0004] In the ideal state, taking two parallel converters as an example, the situation of the output power ratio is as shown in (2):

[0005]

[0006] where is the power ratio, , , are respectively the output power, output current, and virtual impedance of the th converter; however, in actual situations, the wire resistance of the transmission line will inject uncontrollable factors into the power distribution performance of the DC microgrid under droop control, and its mathematical expression is as shown in (3):

[0007]

[0008] where is the resistance of the transmission line between the output port of the th converter and the DC bus; it can be seen that the transmission line resistance destroys the strict power constraint of droop control, and easily leads to problems such as the system power not reaching the expected level and the reduction of the converter life; when the virtual impedance is large, the influence of the wire resistance is small; when the virtual impedance is small, the influence of the wire resistance is large.

[0009] In the prior art, there are mainly three ways to solve the above problems: First, by sharing information between converters or within the entire DC system, average power control or power control in a specific ratio is achieved; Second, by calculating the line resistance of the transmission line and compensating the droop coefficient, the influence of the line resistance on power distribution in the DC microgrid is eliminated; Third, based on the load information, the virtual impedance is dynamically adjusted, increasing the virtual impedance under heavy load and decreasing the virtual impedance under light load, thereby relatively reducing the influence of the transmission line resistance. However, such a solution cannot completely solve engineering problems, so it is not widely applied.

[0010] In the first two of the above solutions, the first type of solution is completely based on the communication network and communication hardware. In the second type of solution, there are also some communication-based strategies, which all increase the complexity of the system, raise the construction cost of the system, increase the control difficulty and data processing complexity of the system; it is difficult to be popularized and applied in most low-cost DC microgrid systems; In addition, some line resistance estimation strategies that do not require communication are mostly open-loop calculations, ignoring the voltage fluctuations of the DC bus, and are only applicable to single-bus DC microgrid systems. Such a solution is difficult to ensure the accuracy of line resistance estimation, the reliability of the algorithm, and wide applicability. Summary of the Invention

[0011] Aiming at the deficiencies existing in the prior art, the present invention provides a method for estimating and compensating the droop control coefficient of the transmission line resistance without communication, introducing a Kalman filter to comprehensively consider the fluctuation amount of the DC bus voltage, and through transforming the theory to generalize the estimation algorithm to a multi-bus DC microgrid system, and completely localizing the control, without additional communication facilities, reducing the construction and later maintenance costs.

[0012] The purpose of the present invention is achieved as follows: A method for estimating and compensating the droop control coefficient of the transmission line resistance without communication includes the following steps:

[0013] 1) Under steady-state conditions, inject a pulse disturbance into the reference value of the current loop in the voltage-current double-loop control of the DC-DC converter, and sample its output voltage and inductor current through analog-to-digital conversion within the DC-DC converter. Calculate the real-time output current based on the relationship between the inductor current and the output current of the Boost converter, and calculate their respective fluctuation amounts;

[0014] 2) Based on the mathematical model of the system from the output port of the DC-DC converter to the DC bus in the DC microgrid, construct a PI link, use the output of the PI link as the estimated value of the transmission line resistance of the system, substitute it into the mathematical model, and obtain the bus voltage fluctuation amount;

[0015] 3) Send the calculated bus voltage fluctuation amount into a first-order Kalman filter to obtain the mean value of the fluctuation amount;

[0016] 4) Feed the estimated value and the mean value of the fluctuation quantity back to the input side of the PI link to complete a closed-loop operation; after the PI link reaches a steady state, output the estimated value of the line resistance and compensate it to the droop coefficient;

[0017] 5) Repeat to enable this algorithm, and the repeat enable time interval is greater than 10,000 times the system switching period, and select according to the application scenario as an integer multiple of 10 times of 10,000 times the system switching period.

[0018] Further, before step 1), first judge whether the target system is a single-bus DC microgrid system or a multi-bus DC microgrid system. If it is a single-bus DC microgrid system, directly enter step 1). If it is a multi-bus DC microgrid system, first perform transformation to equivalent the multi-bus DC microgrid system to a single-bus DC microgrid system, and then enter step 1).

[0019] Further, step 1) specifically includes:

[0020] When the coefficient reaches a steady state, inject a pulse disturbance with a fixed frequency and a dynamic amplitude into the current loop reference value in the DC-DC converter, where , is the switching frequency of the Boost converter, , is the real-time inductor current of the Boost converter, and the specific implementation is shown in Equation (4):

[0021]

[0022] where is the current loop reference value after disturbance. According to the controller parameter design, Equation (4) is executed every 20 switching cycles; sample the voltage at the output port and the inductor current at the input port of the Boost converter after disturbance, and calculate the real-time output current based on the relationship between the inductor current and the output current of the Boost converter, as shown in Equation (5):

[0023]

[0024] where is the duty cycle, is the inductor current of the Boost converter; calculate the fluctuation values of the real-time output voltage and the output current intersecting the steady-state quantity , as shown in (6):

[0025]

[0026] where and are the steady-state values of the voltage and current at the output port of the converter.

[0027] Furthermore, step 2) specifically includes:

[0028] Substitute the calculated , and the initialized parameters , into Equation (7) together, and assume to calculate the input quantity of the PI link;

[0029]

[0030] The calculation of the PI link is shown in Equation (8):

[0031]

[0032] where and are the parameters of the proportional link and the integral link respectively, is the output of the PI link, that is, the estimated value of the transmission line resistance; substitute the obtained by the calculation of the PI link into Equation (9) to calculate the corresponding to the current ;

[0033] .

[0034] Furthermore, step 3) specifically includes:

[0035] Send into the first-order Kalman filter for data processing. The calculation process is divided into two parts. First, the state quantity is predicted, and then the state quantity is updated based on Bayesian probability;

[0036] 3-1) Prediction of state quantity:

[0037]

[0038] As shown in Equation (10), use the first-order identity matrix as the state transition matrix, and the predicted value of the next moment directly inherits the output of the current moment;

[0039] Prediction of state estimation covariance matrix:

[0040]

[0041] As shown in (11), use the first-order identity matrix as the state transition matrix, and the predicted value of the covariance matrix of the next moment directly inherits the covariance prediction matrix of the current moment;

[0042] 3 - 2) State variable update:

[0043]

[0044] Where is the state variable observation value, is the Kalman gain matrix, is the state variable update value;

[0045] State estimation covariance matrix update:

[0046]

[0047] Where is the state estimation covariance matrix at the current moment after update;

[0048] 3 - 3) Kalman gain matrix Update:

[0049]

[0050] Where is the covariance matrix of the observation error;

[0051] (10) - (14) are the calculation processes of the first - order Kalman filter, and its output is the mean value of the DC bus voltage fluctuation .

[0052] Furthermore, the specific calculation process function of the mean value is: The state variable is the DC bus voltage fluctuation value ; In the first calculation cycle, it is calculated by (9) , and is sent into the Kalman filter as the observed quantity of the state variable, that is , combined with the initialization parameters in the filter, the first output quantity is calculated through (12), that is the initial value of; According to the update and prediction formulas of (10) - (14), the Kalman filter continuously outputs new state variable estimation values, and finally reaches a steady state. At this time, the output value is the mean value of the DC bus voltage fluctuation .

[0053] Furthermore, step 4) specifically includes:

[0054] After each cycle of calculation is completed, and are updated by sampling, is updated through the first - order Kalman filter, and then sent into equation (7) to calculate the PI - link input , start the calculation of the next cycle; until the input of the PI link When it is reached, stop the loop and output the output of the PI link at this moment, that is, the line resistance estimation value ; Compensate the output to the droop control coefficient as shown in Equation (15);

[0055]

[0056] where is the compensated droop coefficient.

[0057] Furthermore, the DC-DC converter selects a model that supports floating-point operations.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention improves the accuracy of line resistance estimation, strictly constrains the power distribution of parallel converters under droop control, enhances the reliability of the DC microgrid system, extends the service life of the converter, does not require communication facilities, is completely locally controlled, and reduces the initial investment and later maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0060] Figure 1 It is a schematic diagram of a Boost converter and its voltage and current double-loop control circuits, and emphasizes the pulse disturbance injection method.

[0061] Figure 2 It is a flow chart of the algorithm of the present invention.

[0062] Figure 3 It is a single-bus hardware platform for verifying the present invention.

[0063] Figure 4 It is the waveform result and line resistance estimation value obtained by running the present invention on the single-bus hardware platform.

[0064] Figure 5 It is a multi-bus hardware platform for the present invention to conduct eyes.

[0065] Figure 6 It is the waveform result and line resistance estimation value obtained by running the present invention on the multi-bus hardware platform.

[0066] Figure 7 It is a schematic diagram of the existing DC microgrid system. Detailed implementation manners

[0067] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0068] A method for estimating the line resistance of a communication-free transmission line and compensating the sag control coefficient includes the following steps:

[0069] 1) Determine whether the target system is a single-bus DC microgrid system or a multi-bus DC microgrid system. If it is a single-bus DC microgrid system, directly proceed to 2). If it is a multi-bus DC microgrid system, first perform a transformation on the network side and the load side to equivalent the multi-bus DC microgrid system to a single-bus DC microgrid system, and then proceed to 2).

[0070] 2) Connect multiple Boost converters and their designed controllers in accordance with Figure 3 or Figure 5 ways to ensure the reliability of the selected power supply and load devices; then, as shown in Figure 1 , perform primary control design on the Boost converter. The voltage-current double-loop control needs to ensure the stable operation and fast response of the target system.

[0071] 3) Select parameters for the injected disturbance pulse signal. The disturbance frequency is a fixed frequency , and the disturbance amplitude is a dynamic amplitude , where , is the switching frequency of the Boost converter, , is the real-time inductor current of the Boost converter. The specific implementation is as shown in (4); then, as shown in Figure 1 , inject the disturbance into the control loop of the Boost converter; and after the disturbance, measure the output port voltage and inductor current of the Boost converter, and calculate , ; specifically:

[0072]

[0073] where is the reference value of the current loop after perturbation; designed according to the controller parameters, (4) is executed every 20 switching cycles; sample the voltage at the output port and the inductor current at the input port of the Boost converter after perturbation, and calculate the real-time output current based on the relationship between the inductor current and the output current of the Boost converter, as shown in (5):

[0074]

[0075] where is the duty cycle; calculate the fluctuation value of the intersection steady-state quantity of the real-time output voltage and the output current, and , as shown in (6):

[0076]

[0077] where and are the steady-state values of the voltage and current at the output port of the converter;

[0078] 4) Combine the calculated , with the initialized parameters , , and assume , and calculate the input of the PI link according to (7);

[0079]

[0080] The PI link calculation is shown in (8):

[0081]

[0082] where and are the proportional link parameter and the integral link parameter respectively, is the output of the PI link, that is, the estimated value of the transmission line resistance.

[0083] Store the output of the PI link and output it to (9) to calculate the corresponding DC bus voltage fluctuation value ;

[0084] .

[0085] 5) Input the DC bus voltage fluctuation value into the first-order Kalman filter and collect the output of the first-order Kalman filter, that is, the average value of the DC bus voltage fluctuation ; The average value of the DC bus voltage fluctuation , the estimated value of the output line resistance of the PI link and those obtained in the new cycle 、 The input (7) calculates the input of the PI link again , completing one cycle of calculation; The calculation process in the first-order Kalman filter specifically includes:

[0086] First, perform state quantity prediction, and then update the parameters based on Bayesian probability;

[0087] State quantity prediction:

[0088]

[0089] As shown in (10), the present invention does not rely on a model for prediction, and only uses a first-order identity matrix as the state transition matrix. The predicted value at the next moment directly inherits the output at the current moment;

[0090] Prediction of the state estimation covariance matrix:

[0091]

[0092] As shown in (11), the present invention does not rely on a model for covariance matrix prediction, and only uses a first-order identity matrix as the state transition matrix. The predicted value of the covariance matrix at the next moment directly inherits the covariance prediction matrix at the current moment;

[0093] State quantity update:

[0094]

[0095] Where is the state quantity observation value, is the Kalman gain matrix, is the state quantity update value;

[0096] Update of the state estimation covariance matrix:

[0097]

[0098] Where is the state estimation covariance matrix at the updated current moment;

[0099] Kalman gain matrix Update:

[0100]

[0101] Where is the covariance matrix of the observation error;

[0102] (10)-(14) describe the calculation process of the first-order Kalman filter applied in the present invention, and its output is the mean value of the DC bus voltage fluctuation .

[0103] 6) Monitor the calculated in each cycle, judge whether it is zero. If it is not zero, continue the above processing. If , stop the loop and directly input the line resistance estimation value at this time as the final line resistance estimation result; then compensate the line resistance estimation value to the droop coefficient according to (15) .

[0104]

[0105] where is the compensated droop coefficient.

[0106] 7) If it is set to run repeatedly at a fixed time interval, keep the counter counting and run the above steps again at the set time point.

[0107] The present invention will be further described below in conjunction with the principle

[0108] Through the estimation and compensation of the transmission line resistance, the traditional droop control shown in formula (1) can be further expressed as (16):

[0109]

[0110] At this time, analyzing again the power distribution constraint between parallel converters considering the influence of transmission line resistance, we can obtain (17):

[0111]

[0112] It can be seen that on the premise of accurate line resistance estimation, after compensation, the power constraint between parallel converters can be achieved according to the set value of the droop coefficient in the ideal state.

[0113] The description of the above embodiments is only used to help understand the method and its core idea of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A method for estimating the resistance of a non - communication transmission line and compensating the sag control coefficient, characterized in that, It includes the following steps: 1) Under steady-state conditions, inject a pulse disturbance into the reference value of the current loop in the voltage-current double-loop control of the DC-DC converter, sample its output voltage and inductor current through analog-to-digital conversion within the DC-DC converter, calculate the real-time output current based on the relationship between the inductor current and output current of the Boost converter, and calculate their respective fluctuation amounts; 2) Based on the mathematical model of the system from the output port of the DC-DC converter to the DC bus in the DC microgrid, construct a PI link, use the output of the PI link as the estimated value of the line resistance of the system transmission line, substitute it into the mathematical model, and obtain the bus voltage fluctuation amount; 3) Send the calculated bus voltage fluctuation amount into a first-order Kalman filter to obtain the mean value of the fluctuation amount; 4) Feed back the estimated value and the mean value of the fluctuation amount to the input side of the PI link to complete one closed-loop operation; after the PI link reaches a steady state, output the estimated value of the line resistance and compensate it to the droop coefficient; 5) Repeat enabling this algorithm, and the repeat enabling time interval is greater than 10,000 times the system switching period, and select according to the application scenario as an integer multiple of 10 times 10,000 times the system switching period.

2. A method for estimating the resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 1, characterized in that, Before step 1), it is first determined whether the target system is a single-bus DC microgrid system or a multi-bus DC microgrid system. If it is a single-bus DC microgrid system, it directly enters step 1). If it is a multi-bus DC microgrid system, then the network side and the load side are first transformed to equivalent the multi-bus DC microgrid system into a single-bus DC microgrid system, and then step 1) is entered.

3. A method for estimating the resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 1 or 2, characterized in that, Step 1) specifically includes: After the coefficients enter the steady state, a pulse disturbance with a fixed frequency and a dynamic amplitude is injected into the current-loop reference value in the DC-DC converter, where is the switching frequency of the Boost converter, and is the real-time inductor current of the Boost converter. The specific implementation is shown in Equation (4): and a dynamic amplitude of the pulse disturbance, where , is the switching frequency of the Boost converter, , is the real-time inductor current of the Boost converter. The specific implementation is shown in Equation (4): ； wherein is the reference value of the current loop after perturbation, designed according to the controller parameters, and Equation (4) is executed every 20 switching cycles; sample the voltage at the output port and the inductor current at the input port of the Boost converter after perturbation, and calculate the real-time output current based on the relationship between the inductor current and the output current of the Boost converter, as shown in Equation (5): ； Among them is the duty cycle, is the inductor current of the Boost converter; calculate the real-time output voltage and the output current the fluctuation value of the intersection steady-state quantity 、 , as shown in (6): ； where and are the steady-state values of the voltage and current at the output port of the converter.

4. A method for estimating the line resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 3, characterized in that, Step 2) specifically includes: The calculated , and the initialized parameters , are substituted into Equation (7) together, and assuming , calculate the input quantity of the PI link; ； The calculation of the PI link is shown in Equation (8): ； wherein and are the proportional link parameter and the integral link parameter respectively, is the output of the PI link, i.e., the estimated value of the transmission line resistance; Substitute the calculated by the PI link into Equation (9) to calculate the corresponding to the current ; 。 5. A method for estimating the resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 4, characterized in that, Step 3) specifically includes: Send into a first-order Kalman filter for data processing. The calculation process is divided into two parts. First, the state quantity is predicted, and then the state quantity is updated based on Bayesian probability; 3-1) State quantity prediction: ； As shown in Equation (10), use a first-order identity matrix as the state transition matrix, and the predicted value at the next moment directly inherits the output at the current moment; State estimation covariance matrix prediction: ； As shown in (11), use a first-order identity matrix as the state transition matrix, and the predicted value of the covariance matrix at the next moment directly inherits the covariance prediction matrix at the current moment; 3-2) State quantity update: ； wherein is the observed value of the state quantity, is the Kalman gain matrix, is the updated value of the state quantity; State estimation covariance matrix update: ； Among them is the covariance matrix of the current state estimate after update; 3-3) Kalman gain matrix Update: ； wherein is the covariance matrix of the observation error; (10)-(14) are the calculation process of the first-order Kalman filter, and its output is the mean value of the DC bus voltage fluctuation .

6. A method for estimating the line resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 5, characterized in that, Mean value The specific calculation process function is: the state quantity is the DC bus voltage fluctuation value ; in the first calculation period, it is calculated by (9) , and is sent to the Kalman filter as the observed quantity of the state quantity, that is , combined with the initialization parameters in the filter, the first output quantity is calculated by (12) , that is the initial value; according to the update and prediction formulas of (10)-(14), the Kalman filter continuously outputs new state quantity estimated values, and finally reaches a steady state. At this time, the output value is the mean value of the DC bus voltage fluctuation quantity .

7. A method for estimating the line resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 6, characterized in that, Step 4) specifically includes: After each cycle of calculation is completed, update through sampling and , update through a first-order Kalman filter , then send it into Equation (7) to calculate the input of the PI link again , and start the calculation of the next cycle; until the input of the PI link , stop the loop and output the corresponding output of the PI link at this moment, that is, the line resistance estimation value ; compensate the output to the droop control coefficient as shown in Equation (15); ； Among them is the compensated droop coefficient.

8. A method for estimating the resistance of a non - communication transmission line and compensating the sag control coefficient according to claim 7, characterized in that, The DC-DC converter selects a model that supports floating-point operations.

Images