Anti-countercurrent and demand control system and method for power system

Abstract

The invention discloses a countercurrent prevention and demand control system and method for a power system, and relates to the technical field of power system protection and control. When an existing power system executes load removal operation of demand control, anti-countercurrent protection may be misjudged as an island event, so that the system is disconnected from the network by mistake, and the power supply reliability of a power grid and the effectiveness of demand control are reduced. After the cooperative control decision module receives a demand control request, a disturbance feedforward model representing the load removal operation is generated through transient detection; and then compensating the dynamic impedance spectrum measured in real time by using the disturbance feed-forward model in a duration window of executing a demand control instruction to obtain a corrected dynamic impedance spectrum. Through a feed-forward compensation mechanism, the influence of internal operation disturbance on anti-reflux judgment is actively counteracted, the accuracy of anti-reflux protection in the demand control execution period is ensured, protection misoperation is avoided, and the reliability and cooperative operation capacity of the system are improved.

Description

Technical Field

[0001] This invention relates to the field of power system protection and control technology, specifically a power system anti-reverse current and demand control system. Background Technology

[0002] With the increasing prevalence and large-scale grid connection of distributed generation (DG), it plays a vital role in improving the reliability and economy of the power grid. However, the integration of DG also presents new challenges to the operation and protection of the power system. Among these challenges, backflow prevention protection is a crucial link in ensuring the safety of the power grid and personnel. When a grid fault occurs, such as a line disconnection, causing the DG to operate as an "island" with the local loads it serves, failure of the DG to disconnect from the grid in time may pose a risk of electric shock to maintenance personnel and may also cause equipment damage and system instability when power is restored. Currently, mainstream backflow prevention technologies include active and passive methods, as well as combinations of both. Among these, the backflow prevention method based on impedance spectrum measurement and analysis has attracted attention due to its high sensitivity, low dead zone, and fast response characteristics. This method determines whether islanding has occurred by continuously monitoring changes in the impedance spectrum at the grid connection point.

[0003] On the other hand, with the continuous growth of electricity load, the peak-to-valley difference in the power grid is becoming increasingly significant, making demand control an important means to optimize the power grid load curve and improve power supply reliability and economy. Demand control is usually achieved by cutting off some non-critical loads on the user side to reduce electricity demand during peak hours.

[0004] However, a significant technical conflict exists when combining impedance spectrum-based reverse current protection with demand control systems. User-side load shedding operations driven by demand control commands essentially alter the equivalent impedance of the distributed generation's grid connection point. This impedance change is characteristically similar to the impedance change caused by islanding events resulting from grid disconnection. When the impedance spectrum-based reverse current protection system detects this impedance spectrum change caused by internal demand control operations, its decision logic may interpret it as an islanding event, issuing unnecessary disconnection commands and causing the distributed generation to erroneously disconnect from the grid. This erroneous disconnection not only affects the normal operation of the distributed generation but also reduces the power supply reliability of the grid and the effectiveness of demand control. Current technologies lack a comprehensive and efficient solution for effectively distinguishing impedance spectrum changes caused by internal system operations and external grid faults, and for achieving seamless coordination between reverse current protection and demand control. Therefore, a power system reverse current and demand control system and method that can resolve the aforementioned technical conflict, ensure the accuracy of reverse current protection, and improve system operational reliability is urgently needed. Summary of the Invention

[0005] The purpose of this invention is to solve the problem that when existing power systems perform load shedding operations for demand control, the anti-reverse current protection may misjudge the event as an islanding event, leading to the system being disconnected from the grid erroneously, which reduces the power supply reliability of the grid and the effectiveness of demand control. This invention proposes a power system anti-reverse current and demand control system and method.

[0006] A power system anti-reverse current and demand control system, the system comprising a dynamic impedance spectrum measurement module, a collaborative control decision module, a demand monitoring and execution module, and a distributed power grid connection interface module;

[0007] The dynamic impedance spectrum measurement module is used to inject broadband perturbation detection signals into the grid connection point of the power system, and simultaneously acquire the voltage and current response signals of the grid connection point, calculate the dynamic impedance spectrum, and transmit it to the collaborative control decision module.

[0008] The collaborative control decision module is used to establish and update the grid-connected impedance baseline spectrum. Based on the grid-connected impedance baseline spectrum and the grid-connected point dynamic impedance spectrum measured in real time by the dynamic impedance spectrum measurement module, the spectral difference is calculated to determine islanding events.

[0009] Upon receiving a demand control request from the demand monitoring and execution module, a disturbance feedforward model is generated, and when executing the demand control command, the disturbance feedforward model is used to correct the islanding event judgment logic.

[0010] When an islanding event is detected, a disconnect command is sent to the distributed power grid interface module.

[0011] The demand monitoring and execution module is used to monitor the total power consumption on the user side, and when the total power consumption exceeds a preset threshold, send a demand control request to the collaborative control decision module; and control the target load and provide feedback on the status according to the demand control instructions sent by the collaborative control decision module.

[0012] The distributed power grid connection interface module is used to control the connection status between the distributed power source and the power grid after receiving a disconnection command.

[0013] Preferably, the broadband perturbation detection signal is a pseudo-random binary sequence signal or a linear frequency modulated signal.

[0014] Preferably, the dynamic impedance spectrum measurement module includes a signal generation unit, a digital-to-analog converter, a signal conditioning and injection circuit, and a data acquisition circuit;

[0015] A signal generation unit is used to generate a digital sequence of the broadband perturbation detection signal and transmit it to a digital-to-analog converter.

[0016] A digital-to-analog converter is used to convert the digital sequence into a continuous analog electrical signal and transmit it to a signal conditioning and injection circuit.

[0017] The signal conditioning and injection circuit is used to convert the analog electrical signal into a current signal and inject it into the power system grid connection point through a coupling circuit;

[0018] The data acquisition circuit is used to synchronously acquire the voltage and current response signals of the grid connection point, convert them into a digital signal sequence, and calculate the dynamic impedance spectrum based on the digital signal sequence.

[0019] Preferably, the data acquisition circuit includes a voltage sensing channel, a current sensing channel, a signal conditioning circuit, and an analog-to-digital conversion unit;

[0020] The voltage sensing channel is used to measure the grid connection point voltage and transmit it to the signal conditioning circuit.

[0021] The current sensing channel is used to measure the grid connection point current and transmit it to the signal conditioning circuit.

[0022] The signal conditioning circuit is used to filter and amplify the voltage and current signals at the grid connection point to obtain the processed voltage and current response signals, which are then transmitted to the analog-to-digital conversion unit.

[0023] The analog-to-digital conversion unit is used to synchronously convert voltage and current response signals into digital signal sequences under the same sampling clock, and to calculate the dynamic impedance spectrum based on the digital signal sequence.

[0024] Preferably, the process for obtaining the grid-connected impedance baseline spectrum is as follows:

[0025] Within a preset baseline learning time period, the dynamic impedance spectrum measurement module is controlled to continuously perform M independent network impedance spectrum measurements, and the M impedance spectrum samples are statistically processed at each frequency point to obtain the grid-connected impedance baseline spectrum.

[0026] Preferably, the process of the collaborative control decision module updating the grid-connected impedance baseline spectrum is as follows: when the network state is determined to be stable, the grid-connected impedance baseline spectrum is updated by performing an exponentially weighted moving average using the current real-time impedance spectrum.

[0027] Preferably, the collaborative control decision module is configured to determine the process of isolated events:

[0028] Calculate the dynamic impedance spectrum of the grid connection point in real time;

[0029] Calculate the spectral difference between the real-time dynamic impedance spectrum and the grid-connected impedance baseline spectrum;

[0030] The spectral difference is compared with a preset islanding threshold. When the spectral difference continuously exceeds the islanding threshold, an islanding event is determined to have occurred.

[0031] Preferably, the process of generating the perturbation feedforward model is as follows:

[0032] A transient characteristic detection pulse is injected into the grid connection point, and the transient voltage response excited by the detection pulse is acquired simultaneously;

[0033] Using the transient detection pulse and the transient voltage response, combined with the current real-time dynamic impedance spectrum, the effective admittance of the target load to be removed is calculated;

[0034] Based on the effective admittance, the real-time dynamic impedance spectrum after load removal is predicted, and then subtracted from the real-time dynamic impedance spectrum before load removal to obtain the disturbance feedforward model.

[0035] Preferably, the specific process of modifying the islanding event determination logic using the perturbation feedforward model is as follows:

[0036] Within the duration window of executing the demand control command, the disturbance feedforward model is compensated with the real-time measured dynamic impedance spectrum to obtain the corrected dynamic impedance spectrum.

[0037] Based on the corrected dynamic impedance spectrum, the corrected spectral difference degree is calculated;

[0038] The corrected spectral difference is compared with the islanding determination threshold to determine islanding events.

[0039] A method for preventing reverse current and controlling demand in a power system, the method being based on a power system anti-reverse current and demand control system, the method comprising the following:

[0040] Step S1: Inject a broadband perturbation detection signal into the power system grid connection point, and simultaneously collect the voltage and current response signals of the grid connection point to calculate the dynamic impedance spectrum;

[0041] Step S2: Establish and update the grid-connected impedance baseline spectrum. Based on the grid-connected impedance baseline spectrum and the grid connection point dynamic impedance spectrum, calculate the spectral difference to determine islanding events.

[0042] Step S3: Monitor the total power consumption on the user side, and when the total power consumption exceeds a preset threshold, send a demand control request;

[0043] Step S4: After receiving the demand control request, send the demand control command to control the target load. At the same time, generate a disturbance feedforward model based on the current real-time dynamic impedance spectrum, and use the disturbance feedforward model to correct the dynamic impedance spectrum. Calculate the new spectral difference based on the corrected dynamic impedance spectrum and the grid-connected impedance baseline spectrum, and correct the islanding event judgment logic. If an islanding event occurs, send a disconnect command to the distributed power grid interface and control the connection status between the distributed power source and the grid.

[0044] The beneficial effects of this invention are:

[0045] This invention achieves coordinated operation of backflow prevention protection and demand control, resolving the inherent conflict between the two. Through a coordinated control decision module with a feedforward compensation mechanism, within a specific duration window of the demand control command execution, predictable impedance changes caused by load shedding actions are subtracted from real-time measurements. This enables the backflow prevention judgment logic to distinguish between internal system operations and actual disconnection events from the external power grid, thereby avoiding protection malfunctions caused by demand control and improving the stability and reliability of system operation.

[0046] This invention introduces a backflow prevention protection correction method based on a disturbance feedforward model, significantly improving the accuracy of protection. Traditional backflow prevention protection often struggles to distinguish between impedance changes and islanding effects caused by internal system operations. The collaborative control decision module of this invention uses transient detection technology to pre-model the impedance spectrum changes caused by target load disconnection and uses this model for compensation during actual disconnection. This allows the protection system to more accurately reflect the true state of the power grid, reduces the false judgment rate, and ensures that disconnection operations are only performed when a real islanding event occurs.

[0047] This invention possesses adaptive capabilities, enabling it to adapt to changes in the power grid operating environment. Through the establishment of the grid-connected impedance baseline spectrum and the exponentially weighted moving average update mechanism, the collaborative control decision module can continuously track the gradual changes in the power grid topology or background load. This ensures that during long-term operation, the benchmark for anti-reverse current protection remains consistent with the current normal state of the power grid, avoiding protection failure or maloperation caused by power grid characteristic drift, thereby improving the long-term applicability and robustness of the system. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of a power system anti-reverse current and demand control system.

[0049] Figure 2 This is a flowchart of a power system anti-reverse current and demand control method. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0052] Example:

[0053] A power system anti-reverse current and demand control system, the system comprising a dynamic impedance spectrum measurement module 1, a collaborative control decision module 2, a demand monitoring and execution module 3, and a distributed power source grid connection interface module 4;

[0054] The dynamic impedance spectrum measurement module 1 is used to inject broadband micro-disturbance detection signals into the power system grid connection point, and simultaneously collect the voltage and current response signals of the grid connection point, calculate the dynamic impedance spectrum, and transmit it to the collaborative control decision module 2.

[0055] The collaborative control decision module 2 is used to establish and update the grid-connected impedance baseline spectrum. Based on the grid-connected impedance baseline spectrum and the grid-connected point dynamic impedance spectrum measured in real time by the dynamic impedance spectrum measurement module 1, the spectral difference is calculated to determine the islanding event.

[0056] After receiving the demand control request sent by the demand monitoring and execution module 3, a disturbance feedforward model is generated, and when executing the demand control command, the disturbance feedforward model is used to correct the islanding event judgment logic.

[0057] When an islanding event is determined to have occurred, a disconnect command is sent to the distributed power grid interface module 4.

[0058] The demand monitoring and execution module 3 is used to monitor the total power consumption on the user side, and when the total power consumption exceeds a preset threshold, send a demand control request to the collaborative control decision module 2; and control the target load and provide feedback on the status according to the demand control command sent by the collaborative control decision module 2.

[0059] The distributed power grid connection interface module 4 is used to control the connection status between the distributed power source and the power grid after receiving the disconnection command.

[0060] Specifically, the demand monitoring and execution unit is used to monitor the total power consumption on the user side in real time, and control the on / off state of one or more preset loads when it receives instructions from the collaborative control decision center.

[0061] The distributed power grid connection interface is used to execute grid connection or disconnection operations of distributed power sources according to instructions issued by the collaborative control decision center.

[0062] Further specifying, the broadband perturbation detection signal is a pseudo-random binary sequence signal or a linear frequency modulated signal.

[0063] Specifically, one function of the dynamic impedance spectroscopy measurement module is to generate and inject a broadband perturbation signal into the grid connection point for network characteristic detection. This signal is designed to have spectral components covering a sufficiently wide frequency range to excite the network response in different frequency bands; simultaneously, its injected power needs to be sufficiently low to ensure no perceptible interference with the normal operation of the grid or power quality.

[0064] Broadband perturbation signals can be pseudo-random binary sequence signals. PRBS signals are periodic, deterministic signals with white noise statistical characteristics; their power spectrum is approximately uniformly distributed in the frequency domain, enabling effective excitation of networks over a wide bandwidth. This PRBS sequence can be generated by a linear feedback shift register. Output bits at time 1 It can be defined by the following formula:

[0065] ;

[0066] in, The register's state bits at a previous time step; It is a set of feedback tap positions; This is the feedback coefficient, which has a value of 0 or 1; This represents the XOR operation. By selecting appropriate feedback coefficients, PRBS sequences with desired lengths and characteristics can be generated. The specific hardware or software implementation of LFSRs is a well-known technique in the field of digital circuits and signal processing, and will not be elaborated upon here.

[0067] Broadband perturbation signals can also be other signals with wide-spectrum characteristics, such as linear frequency modulated (LFM) signals. A linear frequency modulated (LFM) signal... It can be represented as:

[0068] ;

[0069] in, It is the signal amplitude. It is the starting frequency. It refers to the frequency scanning rate. This signal can linearly change its frequency from the starting frequency to the ending frequency within a set time period, thereby achieving sequential excitation of different frequency points of the network.

[0070] The generation and injection process of broadband perturbation detection signals may further include the following steps:

[0071] Inside the collaborative control decision center, a digital signal processor or microcontroller generates a digitized broadband signal sequence according to a preset algorithm.

[0072] A digital signal sequence is input to a digital-to-analog converter, which converts it into a continuous-time low-voltage analog electrical signal.

[0073] A low-voltage analog electrical signal is fed into a signal conditioning and injection circuit. One specific implementation of this circuit could be a voltage-controlled current source, used to linearly convert the input voltage signal into a proportional current signal with sufficient driving capability. Using a current source injection method can reduce the impact of local impedance fluctuations at the grid connection point on the injected signal.

[0074] Current signal The signal is injected into the phase line of the power line through a coupling circuit. This coupling circuit includes at least one series coupling capacitor, which isolates the power frequency voltage to prevent it from entering in reverse and damaging the signal generation circuit at the front end, while allowing high-frequency probe signal components to pass through. In some applications requiring higher safety, the coupling circuit may also include a high-frequency isolation transformer to achieve electrical isolation between the injection circuit and the main power grid.

[0075] Another function of the dynamic impedance spectrum measurement module is to acquire the voltage and current response signals of the grid connection point with high precision while injecting the probe signal, and to ensure that the two are strictly synchronized in time, which is the basis for the accuracy of subsequent impedance spectrum calculation.

[0076] Further specifying, the dynamic impedance spectrum measurement module 1 includes a signal generation unit, a digital-to-analog converter, a signal conditioning and injection circuit, and a data acquisition circuit;

[0077] A signal generation unit is used to generate a digital sequence of the broadband perturbation detection signal and transmit it to a digital-to-analog converter.

[0078] A digital-to-analog converter is used to convert the digital sequence into a continuous analog electrical signal and transmit it to a signal conditioning and injection circuit.

[0079] The signal conditioning and injection circuit is used to convert the analog electrical signal into a current signal and inject it into the power system grid connection point through a coupling circuit;

[0080] The data acquisition circuit is used to synchronously acquire the voltage and current response signals of the grid connection point, convert them into a digital signal sequence, and calculate the dynamic impedance spectrum based on the digital signal sequence.

[0081] Further specifying, the data acquisition circuit includes a voltage sensing channel, a current sensing channel, a signal conditioning circuit, and an analog-to-digital conversion unit;

[0082] The voltage sensing channel is used to measure the grid connection point voltage and transmit it to the signal conditioning circuit.

[0083] The current sensing channel is used to measure the grid connection point current and transmit it to the signal conditioning circuit.

[0084] The signal conditioning circuit is used to filter and amplify the voltage and current signals at the grid connection point to obtain the processed voltage and current response signals, which are then transmitted to the analog-to-digital conversion unit.

[0085] The analog-to-digital conversion unit is used to synchronously convert voltage and current response signals into digital signal sequences under the same sampling clock, and to calculate the dynamic impedance spectrum based on the digital signal sequence.

[0086] Specifically, the voltage sensing channel is used to measure the real-time voltage at the grid connection point. Considering the high amplitude of the power frequency voltage on the power line, this channel can employ a high-voltage differential probe or an opto-isolated amplifier with a precision resistor voltage divider network. Its function is to proportionally attenuate and isolate the high-amplitude power frequency voltage to a low-voltage range suitable for processing by the back-end circuitry, thus obtaining the voltage measurement signal. .

[0087] The current sensing channel is used to measure the total current flowing through the grid connection point, which is the sum of the mains current at power frequency and the injected perturbation probe current. The superposition of current and signal strength. This channel can employ a high-frequency Hall effect current sensor or a Rogowski coil, which provides good electrical isolation and sufficient bandwidth to respond to high-frequency detection signal components, ultimately outputting a voltage signal proportional to the total current. .

[0088] The signal conditioning circuit receives raw measurement signals from the voltage sensing channel and the current sensing channel. and The circuit includes at least an anti-aliasing filter to filter out noise signals above half the Nyquist sampling frequency, preventing spectral aliasing. The circuit may also include a programmable gain amplifier to dynamically adjust the signal amplitude to maximize the utilization of the analog-to-digital converter's range.

[0089] The analog-to-digital converter (ADC) unit is the core of synchronous sampling. In a preferred embodiment, this unit employs a dual-channel or multi-channel synchronous sampling ADC. This type of ADC internally includes a single clock source and sample-and-hold circuitry, ensuring that analog signals from multiple channels are sampled and quantized at exactly the same time, thereby eliminating phase delay between channels. The ADC's sampling frequency... It needs to be set according to the Nyquist sampling theorem, that is It must be greater than the highest frequency in the spectrum range of the probe signal. Twice the size of the signal to reconstruct it without distortion.

[0090] The synchronous acquisition process of the grid connection point voltage and current response may further include the following steps:

[0091] The generation of detection signals and the sampling clock of data acquisition units are synchronously triggered by the same master clock signal within the collaborative control decision center. This is one way to achieve system-level synchronization.

[0092] The voltage sensing channel and the current sensing channel work in parallel and continuously to convert the voltage and current at the grid connection point into low-voltage analog signals.

[0093] Low-voltage analog signals are filtered and amplified by signal conditioning circuitry.

[0094] Driven by each sampling clock cycle, the analog-to-digital converter simultaneously samples and quantizes the conditioned voltage and current signals, generating two discrete digital signal sequences, namely the voltage response sequences. and total current response sequence .

[0095] Due to the collected total current response sequence It includes the power frequency current component, while impedance spectrum calculation requires the injected signal. In response, the collaborative control decision center will use digital filtering algorithms (such as high-pass or band-pass filters) to... The injected probe signal component was extracted to obtain... Or, due to the injected digital signal It is known that signal processing techniques such as cross-correlation can be used to... The response components are separated from the data. Finally, the collaborative control decision center obtains synchronized data pairs for calculating the impedance spectrum. .

[0096] Further defining the process for obtaining the grid-connected impedance baseline spectrum:

[0097] Within a preset baseline learning time period, the dynamic impedance spectrum measurement module (1) is controlled to continuously perform M independent network impedance spectrum measurements, and the M impedance spectrum samples are statistically processed at each frequency point to obtain the grid-connected impedance baseline spectrum.

[0098] Specifically, one function of the collaborative control decision center is to establish and maintain a grid-connected impedance baseline spectrum that can accurately characterize the electrical characteristics of the power grid under normal grid-connected conditions. This baseline spectrum is the core benchmark for all subsequent state determinations, especially for islanding event determinations.

[0099] In one specific implementation, the grid-connected impedance baseline spectrum establishment process is initiated when the system is first put into operation or when an external reset command is received, and it is confirmed that the distributed power source is in a stable and reliable grid-connected state. This process may include the following steps:

[0100] Within a preset baseline learning time period, the collaborative control decision center controls the dynamic impedance spectrum measurement module to continuously execute... Sub-independent network impedance spectrum measurements, in which A positive integer. Each measurement yields an independent impedance spectrum sample, denoted as . ,in .

[0101] To reduce the impact of transient network disturbances and measurement noise on the accuracy of the baseline spectrum, the collaborative control decision center performs [further measures] on the acquired [data / data]. Statistical processing is performed on each impedance spectrum sample. One specific processing method is to calculate this... Each sample at each frequency point The arithmetic mean of the values ​​is used to obtain the final grid-connected impedance baseline spectrum. :

[0102] ;

[0103] In another implementation, to further improve the stability of the baseline, an outlier removal algorithm, such as Mahalanobis distance or local anomaly factor algorithm, can be used before calculating the average value to identify and exclude distorted samples caused by accidental strong disturbances during the measurement.

[0104] The calculated grid-connected impedance baseline spectrum It is stored in the form of data files in a non-volatile memory within the collaborative control decision center, such as flash memory or electrically erasable programmable read-only memory.

[0105] Considering that the topology or background load of the public power grid may change slowly over time, leading to mismatch in the stored baseline spectrum, this embodiment also provides an adaptive update mechanism for the baseline spectrum. This mechanism allows the baseline spectrum to be fine-tuned during long-term system operation to adapt to the slowly changing characteristics of the power grid.

[0106] This update mechanism is triggered by the collaborative control decision center when it confirms that the power grid is in a long-term stable operating state. Its specific implementation can be a smooth update algorithm based on an exponentially weighted moving average. When the system determines that the current network state is stable and not an island, the system will use the current real-time impedance spectrum. For the old baseline spectrum Update and generate a new baseline spectrum. :

[0107] ;

[0108] in, It is a small constant with a value between 0 and 1, and is called the smoothing coefficient. The value of determines the update rate; a smaller value indicates a lower update rate. This value makes the baseline spectrum less sensitive to transient changes, thus ensuring its long-term stability. The update process can be performed periodically, provided stable operating conditions are met.

[0109] Another function of the collaborative control decision center is to perform highly reliable backflow prevention judgments during the system's routine monitoring phase. This method is based on pattern comparison and quantitative analysis of the real-time measured network dynamic impedance spectrum and the pre-stored grid-connected impedance baseline spectrum.

[0110] Further defining the process by which the collaborative control decision module 2 updates the grid-connected impedance baseline spectrum: when the network state is determined to be stable, the grid-connected impedance baseline spectrum is updated by performing an exponentially weighted moving average using the current real-time impedance spectrum.

[0111] Further specifying the process for configuring and determining isolated events in the collaborative control decision module 2:

[0112] Calculate the dynamic impedance spectrum of the grid connection point in real time;

[0113] Calculate the spectral difference between the real-time dynamic impedance spectrum and the grid-connected impedance baseline spectrum;

[0114] The spectral difference is compared with a preset islanding threshold. When the spectral difference continuously exceeds the islanding threshold, an islanding event is determined to have occurred.

[0115] Specifically, the standard workflow of this islanding determination method may include the following steps:

[0116] The collaborative control decision center, at a preset period, adjusts the synchronous data collected in real time by the dynamic impedance spectrum measurement module. The discrete real-time dynamic impedance spectrum covering the preset frequency band is calculated. ,in , representing discrete frequency points.

[0117] To quantify the real-time impedance spectrum and the grid-connected impedance baseline spectrum The degree of deviation between them is calculated by the collaborative control decision center using a spectral difference. One specific calculation method is to use Euclidean distance, that is, to calculate the distance between two spectra across all... The square root of the sum of the squares of the differences at each frequency point:

[0118] ;

[0119] in, Indicates the current moment. Is it at the current frequency point? The complex impedance value, It is the complex impedance value of the baseline spectrum at the corresponding frequency point. Represents the modulus of a complex number.

[0120] In another embodiment, this degree of difference It could be a weighted norm distance, which assigns different weights to the differences at different frequencies. To highlight certain frequency bands that are more sensitive to changes in the state of isolated areas:

[0121] ;

[0122] In yet another implementation, this degree of difference It can also be the maximum absolute difference between the two spectra at all frequency points, which is more sensitive to drastic local changes.

[0123] The collaborative control decision center pre-sets a fixed islanding threshold. This threshold is set based on the normal fluctuation range of the impedance spectrum during normal grid operation, while retaining a sufficient safety margin. The system will calculate the difference in real time. Compare with this threshold.

[0124] When the power grid disconnects and forms an island, the physical structure of the network undergoes a fundamental change. Its equivalent impedance transforms from the low impedance characteristics of a large power grid to the high impedance and resonant characteristics of a local load network. This structural abrupt change will lead to… Throughout the entire frequency band This produces a huge deviation, thus making the calculated... The value increases sharply and significantly exceeds the threshold. .

[0125] To avoid misjudgments caused by transient strong disturbances, in a preferred embodiment, the determination logic further includes a continuous verification mechanism. That is, only when a strong disturbance is detected... The condition continues for more than a preset number of judgment periods. It was only then that the occurrence of the isolated island incident was finally confirmed.

[0126] Once an islanding event is confirmed according to the above logic, the collaborative control decision center immediately generates and sends a disconnect command to the distributed power grid interface to execute protection actions.

[0127] After receiving a demand control request, the collaborative control decision center performs a proactive, feedforward disturbance calibration and modeling process to ensure that the load shedding action is not misjudged as an islanding event by the anti-reverse current protection logic. This method can accurately predict the specific impact of the operation on the network impedance spectrum of the grid connection point before the load is actually shedding.

[0128] The transient calibration and generation method of the perturbation feedforward model determines the target load to be removed in the collaborative control decision center. It is subsequently triggered and may include the following steps:

[0129] The collaborative control decision center instructs the dynamic impedance spectrum measurement module to inject a transient characteristic detection pulse into the grid connection point. The waveform and spectral characteristics of the probe pulse can be determined based on the target load to be removed. The design can be optimized based on the type of load. For example, if the target load is an inductive motor, the probe pulse can include more energy components in its characteristic resonant frequency band to obtain a higher signal-to-noise ratio response.

[0130] The dynamic impedance spectroscopy measurement module synchronously acquires data from the probe pulse. The stimulated, which includes the target load Transient voltage of fine response information .

[0131] The collaborative control decision center uses the acquired detection and response data to... And the real-time dynamic impedance spectrum known just before detection. Calculations are performed to generate a feedforward perturbation model. .

[0132] The model is generated as follows: First, the system generates the currently measured real-time dynamic impedance spectrum. Convert to admittance spectrum form:

[0133] ;

[0134] in, It is the total network admittance seen from the grid connection point before load shedding.

[0135] Then, the system uses transient detection and response data pairs to calculate the target load to be removed through a system identification algorithm. Effective Admittance Since the total admittance in a parallel circuit is the sum of the admittances of each branch, disconnecting the load... The new network general guide It can be accurately predicted as:

[0136] ;

[0137] The predicted new network total adduct Converting back to impedance form yields the predicted, new real-time dynamic impedance spectrum after load removal. :

[0138] ;

[0139] Finally, the system subtracts the predicted new impedance spectrum from the impedance spectrum before the cut-off, and the result is the feedforward perturbation model that can accurately characterize the change in impedance spectrum caused by this demand control operation. :

[0140] ;

[0141] The collaborative control decision center will generate a feedforward disturbance model. The data is temporarily stored and prepared for distribution to its internal anti-backflow judgment unit for subsequent collaborative control execution.

[0142] The collaborative control decision center generates a feedforward disturbance model. Subsequently, its third core function is to execute a time-precisely synchronized, feedforward-compensated collaborative control process. This process ensures that load shedding actions for demand control are performed without affecting the accuracy of backflow protection.

[0143] In one specific implementation, the collaborative control execution logic may include the following steps:

[0144] The collaborative control decision center issues synchronous commands. Specifically, it uses its internally calculated feedforward perturbation model. The data is transmitted to the internal anti-backflow determination unit; and within the same system clock cycle, the target load cut-off command is sent to the demand monitoring and execution unit 30 through the communication interface.

[0145] In the anti-backflow determination unit, a modified determination algorithm based on feedforward compensation is temporarily activated. The activation of this modified algorithm is synchronized with the issuance of demand control commands, and only occurs within a preset duration window that covers the load shedding transient process. Effective internally.

[0146] Within the effective time window of this corrected algorithm, the anti-backflow determination unit still calculates the real-time dynamic impedance spectrum in each measurement cycle. Subsequently, the algorithm compensates for the real-time impedance spectrum using the received feedforward perturbation model to calculate a corrected real-time impedance spectrum. The calculation formula is as follows:

[0147] ;

[0148] in, It is the impedance spectrum actually measured at the current moment. It is the pre-calibrated expected change in impedance spectrum caused by the load shedding operation. This subtraction operation is equivalent to actively filtering out known internal operational disturbances from the actual measurement results.

[0149] The corrected determination algorithm then uses the corrected impedance spectrum. Calculate the corrected spectral difference. :

[0150] ;

[0151] in, It is the stored grid-connected impedance baseline spectrum.

[0152] The calculated corrected spectral difference Compared with the preset, unchanged island determination threshold A comparison was made. Since the feedforward perturbation model has accurately offset the effects of the load shedding operation, Its shape will be very close to ,therefore The value will remain well below The level is such that it will not trigger a grid disconnection command. If a real grid disconnection event occurs during this period, its resulting... The enormous changes will be far from The compensation that can be made leads to The value will still surge and exceed This ensures that the reliability of the protection is not affected.

[0153] During the preset duration After completion, the decision-making algorithm in the collaborative control decision center automatically exits the correction mode and reverts to the standard operating mode, which directly uses the uncompensated real-time impedance spectrum. To calculate spectral difference And make a judgment.

[0154] One function of the demand monitoring and execution unit is to monitor the total power consumption on the user side in real time, and when the power value exceeds a certain limit, to initiate a demand control request to the collaborative control decision center, thereby triggering the subsequent collaborative control process.

[0155] This function is implemented by the power monitoring module within the demand monitoring and execution unit. This power monitoring module can be an integrated energy metering chip, or it can consist of independent voltage and current sensors and a dedicated computing unit.

[0156] The monitoring and event triggering process for total power consumption may further include the following steps:

[0157] The power monitoring module synchronously acquires the real-time instantaneous voltage value at the main incoming line through voltage transformers (PTs) and current transformers, or equivalent voltage and current sensors, installed at the user's main incoming line. and instantaneous current value .

[0158] The computing unit processes the collected voltage and current signals to calculate the current total active power on the user side. One specific calculation method is to integrate the product of the instantaneous values ​​of voltage and current within one power frequency cycle T, and then calculate its average value:

[0159] ;

[0160] For discrete sampling systems, this calculation can be performed by the following formula:

[0161] ;

[0162] in, It is the active power value at the current sampling time. and These are historical voltage and current sampling values. The number of sampling points within a power frequency cycle is given. For the specific circuit implementation of this power calculation, those skilled in the art can use commercially available power metering chips, which is a well-known technology in the field and will not be described in detail here.

[0163] One or more demand control alarm thresholds are pre-set within the demand monitoring and execution unit. This threshold is typically set based on the maximum demand stipulated in the power supply contract between the user and the power grid.

[0164] The computing unit will calculate the total active power in real time. With alarm threshold Conduct continuous comparisons.

[0165] When detected When the conditions are met, the demand monitoring and execution unit immediately sends a demand control request signal to the collaborative control decision center through its communication interface. This signal may contain information such as the current excess power value and a priority list of loads to be cut off, for the collaborative control decision center to make a decision. To prevent unnecessary frequent triggering caused by instantaneous power fluctuations, a delayed confirmation mechanism can be added, that is, only when... The conditions lasted for a period of time Only then is the formal request signal sent.

[0166] Another function of the demand monitoring and execution unit is to accurately execute the on / off control of one or more predetermined target loads according to the instructions issued by the collaborative control decision center, and to report the operation results back to the decision center.

[0167] In one specific implementation, this control function is achieved through a load control relay array and a communication interface. The user side pre-connects the power supply circuits for a portion of non-critical, interruptible loads via this relay array. Each relay controls an independent load or load group.

[0168] The control and status feedback process for the target load may further include the following steps:

[0169] The demand monitoring and execution unit receives load control commands from the collaborative control decision center through its communication interface. These commands explicitly specify the identifier of the target load that needs to be cut off or restored.

[0170] The microcontroller inside the unit parses the instruction and locates the relay drive channel corresponding to the target load identifier.

[0171] The microcontroller outputs a control level signal through this drive channel to drive the corresponding relay. In one embodiment, the relay can be an electromagnetic relay, where the control signal energizes or de-energizes its coil, thereby driving its main contacts to open or close. In another embodiment, the relay can be a solid-state relay, where the control signal turns its internal optocoupler circuit on or off, thereby controlling the on / off state of the main circuit. The specific design of the relay drive circuit is well-known in the field of electronic technology and will not be elaborated upon here.

[0172] After executing the control action, the demand monitoring and execution unit verifies the actual result of the operation. One specific verification method is to read the status of the auxiliary contacts on the relay used for status indication. Another, more reliable method is to monitor whether the current in the target load circuit has dropped to zero or returned to the operating current using a miniature current transformer installed on the target load circuit.

[0173] Based on the verification results, the demand monitoring and execution unit generates a status feedback message and sends it to the collaborative control decision center via the communication interface. This message contains at least the identifier of the operated load and status information indicating success or failure of the operation, for the collaborative control decision center to perform closed-loop confirmation and event logging.

[0174] The distributed generation grid connection interface is the final executor of the system. Its core function is to receive and unconditionally execute protection commands from the collaborative control decision center, so as to quickly and reliably disconnect the distributed generation from the grid when an islanding event is confirmed.

[0175] In one specific implementation, the distributed power grid connection interface can physically be a device integrating a communication unit, a local controller, and a fast circuit breaker. This device is directly connected in series between the AC output of the distributed power source and the grid connection point.

[0176] The process of receiving and executing protection commands may further include the following steps:

[0177] The communication unit of the distributed power grid interface continuously listens for instructions from the collaborative control and decision-making center through a reliable communication link. When the collaborative control and decision-making center determines that an islanding event has occurred, it will immediately broadcast or send a clear disconnection instruction message point-to-point.

[0178] Upon receiving the disconnect command message, the local controller of the interface first performs integrity and validity checks on it.

[0179] In one embodiment, the verification may include checking the cyclic redundancy check (CRC) code of the message to ensure that no errors occur during data transmission. In applications with higher security requirements, the instruction message may be accompanied by a digital signature or message authentication code generated by the collaborative control decision center. The local controller must use a pre-shared key for verification to prevent forged instructions or replay attacks.

[0180] Once the command passes verification, the local controller immediately outputs a drive signal to the trip coil or actuator of the fast-acting circuit breaker. This signal causes the main contacts of the circuit breaker to open within a very short time, thereby physically disconnecting the distributed power source from the power grid.

[0181] The specific implementation of the circuit breaker tripping circuit is a well-known technology in the field of power system protection, and will not be elaborated here.

[0182] After the circuit breaker completes its operation, the local controller reads the status of the circuit breaker's auxiliary contacts to confirm whether it is indeed in the open position.

[0183] Based on the confirmed results, the local controller generates a status feedback message and sends it to the collaborative control decision center via the communication link, informing it that the grid disconnection command has been successfully executed. This interface will remain in a disconnected, locked state until it receives a reset or grid reconnection permission command from the collaborative control decision center after confirming that the power grid has returned to normal.

[0184] Further defining the process of generating the perturbation feedforward model:

[0185] A transient characteristic detection pulse is injected into the grid connection point, and the transient voltage response excited by the detection pulse is acquired simultaneously;

[0186] Using the transient detection pulse and the transient voltage response, combined with the current real-time dynamic impedance spectrum, the effective admittance of the target load to be removed is calculated;

[0187] Based on the effective admittance, the real-time dynamic impedance spectrum after load removal is predicted, and then subtracted from the real-time dynamic impedance spectrum before load removal to obtain the disturbance feedforward model.

[0188] Further specifying the process of modifying the islanding event determination logic using the perturbation feedforward model:

[0189] Within the duration window of executing the demand control command, the disturbance feedforward model is compensated with the real-time measured dynamic impedance spectrum to obtain the corrected dynamic impedance spectrum.

[0190] Based on the corrected dynamic impedance spectrum, the corrected spectral difference degree is calculated;

[0191] The corrected spectral difference is compared with the islanding determination threshold to determine islanding events.

[0192] A method for preventing reverse current and controlling demand in a power system, the method being based on a power system anti-reverse current and demand control system, the method comprising the following:

[0193] Step S1: Inject a broadband perturbation detection signal into the power system grid connection point, and simultaneously collect the voltage and current response signals of the grid connection point to calculate the dynamic impedance spectrum;

[0194] Step S2: Establish and update the grid-connected impedance baseline spectrum. Based on the grid-connected impedance baseline spectrum and the grid connection point dynamic impedance spectrum, calculate the spectral difference to determine islanding events.

[0195] Step S3: Monitor the total power consumption on the user side, and when the total power consumption exceeds a preset threshold, send a demand control request;

[0196] Step S4: After receiving the demand control request, send the demand control command to control the target load. At the same time, generate a disturbance feedforward model based on the current real-time dynamic impedance spectrum, and use the disturbance feedforward model to correct the dynamic impedance spectrum. Calculate the new spectral difference based on the corrected dynamic impedance spectrum and the grid-connected impedance baseline spectrum, and correct the islanding event judgment logic. If an islanding event occurs, send a disconnect command to the distributed power grid interface and control the connection status between the distributed power source and the grid.

[0197] Specifically, during the system initialization phase, the dynamic impedance spectrum measurement module probes the power grid under stable grid-connected conditions. The collaborative control decision center calculates and stores a grid-connected impedance baseline spectrum based on the collected signals. This baseline spectrum serves as the benchmark for all subsequent decisions.

[0198] In the routine monitoring phase, the dynamic impedance spectrum measurement module performs periodic measurements, and the collaborative control decision center calculates the dynamic impedance spectrum in real time. By comparing it with the grid-connected impedance baseline spectrum, continuous monitoring of the anti-reverse flow status is carried out.

[0199] When the demand monitoring and execution unit detects that the total power consumption exceeds a preset threshold, the unit sends a demand control request to the collaborative control decision center. After determining the target load to be disconnected, the collaborative control decision center does not immediately execute the disconnection action.

[0200] The collaborative control decision center first enters the feedforward calibration stage, which instructs the dynamic impedance spectrum measurement module to perform a transient detection of the target load and generate a feedforward perturbation model that can accurately characterize the impedance spectrum change caused by the load shedding action based on the detection results.

[0201] Subsequently, the collaborative control decision center temporarily modified its internal anti-backflow judgment algorithm based on the feedforward disturbance model and simultaneously issued a load shedding command to the demand monitoring and execution unit. The modified anti-backflow judgment algorithm can identify and eliminate disturbances caused by this load shedding operation, thereby avoiding misjudgment. At the same time, the collaborative control decision center sent a command to the distributed power grid connection interface to maintain grid connection status.

[0202] After the load shedding action is completed and the disturbance transient ends, the anti-backflow judgment algorithm of the collaborative control decision center returns to the normal monitoring mode, and the system continues to perform periodic monitoring until the next event is triggered.

[0203] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A power system reverse flow and demand control system, comprising: The system comprises a dynamic impedance spectrum measurement module (1), a collaborative control decision module (2), a demand monitoring and execution module (3) and a distributed power grid-connected interface module (4); The dynamic impedance spectrum measurement module (1) is used for injecting a broadband perturbation detection signal into a power system grid-connected point, synchronously collecting voltage and current response signals of the grid-connected point, calculating a dynamic impedance spectrum and transmitting the dynamic impedance spectrum to the collaborative control decision module (2); The collaborative control decision module (2) is used for establishing and updating a grid-connected impedance baseline spectrum, calculating a frequency spectrum difference degree based on the grid-connected impedance baseline spectrum and a dynamic impedance spectrum measured by the dynamic impedance spectrum measurement module (1) in real time, and determining an islanding event; After receiving a demand control request sent by the demand monitoring and execution module (3), a perturbation feedforward model is generated, and the perturbation feedforward model is used to correct the islanding event determination logic when the demand control instruction is executed; When the islanding event is determined to occur, a grid disconnection instruction is sent to the distributed power grid-connected interface module (4); The demand monitoring and execution module (3) is used for monitoring total power consumption on a user side, sending a demand control request to the collaborative control decision module (2) when the total power consumption exceeds a preset threshold, and controlling a target load according to a demand control instruction sent by the collaborative control decision module (2) and feeding back a state; The distributed power grid-connected interface module (4) is used for controlling a connection state of a distributed power and a power grid after receiving the grid disconnection instruction.

2. A power system anti-ffow and demand control system according to claim 1 wherein, The broadband perturbation detection signal is a pseudo-random binary sequence signal or a linear frequency modulation signal.

3. A power system anti-ffow and demand control system according to claim 1 wherein, The dynamic impedance spectrum measurement module (1) comprises a signal generation unit, a digital-to-analog converter, a signal conditioning and injection circuit and a data acquisition circuit; The signal generation unit is used for generating a digital sequence of the broadband perturbation detection signal and transmitting the digital sequence to the digital-to-analog converter; The digital-to-analog converter is used for converting the digital sequence into a continuous analog electric signal and transmitting the continuous analog electric signal to the signal conditioning and injection circuit; The signal conditioning and injection circuit is used for converting the analog electric signal into a current signal and injecting the current signal into the power system grid-connected point through a coupling circuit; The data acquisition circuit is used for synchronously collecting voltage and current response signals of the grid-connected point, converting the voltage and current response signals into digital signal sequences, and calculating a dynamic impedance spectrum according to the digital signal sequences.

4. A power system anti-ffow and demand control system according to claim 3 wherein, The data acquisition circuit comprises a voltage sensing channel, a current sensing channel, a signal conditioning circuit and an analog-to-digital conversion unit; The voltage sensing channel is used for measuring a grid-connected point voltage and transmitting the grid-connected point voltage to the signal conditioning circuit; The current sensing channel is used for measuring a grid-connected point current and transmitting the grid-connected point current to the signal conditioning circuit; The signal conditioning circuit is used for filtering and amplifying the grid-connected point voltage and current signals to obtain processed voltage and current response signals and transmitting the processed voltage and current response signals to the analog-to-digital conversion unit; The analog-to-digital conversion unit is used for synchronously converting the voltage and current response signals into digital signal sequences under the same sampling clock and calculating a dynamic impedance spectrum according to the digital signal sequences.

5. A power system anti-ffow and demand control system according to claim 1 wherein, The process of establishing the grid-connected impedance baseline spectrum comprises the following steps: In a preset baseline learning time period, the dynamic impedance spectrum measurement module (1) is controlled to continuously perform M independent network impedance spectrum measurements, and statistical processing is performed on the M impedance spectrum samples at each frequency point to obtain the grid-connected impedance baseline spectrum.

6. A power system anti-ffow and demand control system according to claim 1 wherein, The process of updating the grid-connected impedance baseline spectrum by the cooperative control decision module (2): when it is judged that the network state is stable, the current real-time impedance spectrum is used to update the grid-connected impedance baseline spectrum by exponential weighted moving average.

7. A power system anti-ffow and demand control system according to claim 1 wherein, The process of configuring the islanding event determination by the cooperative control decision module (2): Real-time calculation of the dynamic impedance spectrum of the grid-connected point; Calculation of the frequency spectrum difference degree between the real-time dynamic impedance spectrum and the grid-connected impedance baseline spectrum; Comparison of the frequency spectrum difference degree with a preset islanding determination threshold value, and determination of an islanding event when the frequency spectrum difference degree continuously exceeds the islanding determination threshold value.

8. A power system anti-ffow and demand control system according to claim 1 wherein, The process of generating a disturbance feedforward model: Injection of a transient characteristic detection pulse into the grid-connected point, and synchronous acquisition of the transient voltage response excited by the detection pulse; Calculation of the effective admittance of the target load to be cut off by using the transient detection pulse and the transient voltage response, and combining the current real-time dynamic impedance spectrum; Prediction of the real-time dynamic impedance spectrum after load cutting off according to the effective admittance, and subtraction of the real-time dynamic impedance spectrum before load cutting off to obtain the disturbance feedforward model.

9. A power system anti-ffow and demand control system according to claim 1 wherein, The specific process of using the disturbance feedforward model to correct the islanding event determination logic: In a duration window of executing the demand control instruction, the disturbance feedforward model is compensated with the real-time measured dynamic impedance spectrum to obtain a corrected dynamic impedance spectrum; Calculation of a corrected frequency spectrum difference degree based on the corrected dynamic impedance spectrum; Comparison of the corrected frequency spectrum difference degree with the islanding determination threshold value to determine an islanding event.

10. A method for reverse flow prevention and demand control of a power system, the method being implemented based on a reverse flow prevention and demand control system according to any one of claims 1-9, characterized in that, The method comprises the following contents: Step S1: injection of a broadband perturbation detection signal into a grid-connected point of a power system, synchronous acquisition of voltage and current response signals of the grid-connected point, and calculation of a dynamic impedance spectrum; Step S2: establishment and update of a grid-connected impedance baseline spectrum, calculation of a frequency spectrum difference degree based on the grid-connected impedance baseline spectrum and the dynamic impedance spectrum of the grid-connected point, and determination of an islanding event; Step S3: monitoring of total power consumption on the user side, and sending of a demand control request when the total power consumption exceeds a preset threshold value; Step S4: after receiving the demand control request, sending of a demand control instruction to control a target load, generating a disturbance feedforward model according to the current real-time dynamic impedance spectrum, using the disturbance feedforward model to correct the dynamic impedance spectrum, calculating a new frequency spectrum difference degree according to the corrected dynamic impedance spectrum and the grid-connected impedance baseline spectrum, correcting the islanding event determination logic, and sending a disconnection instruction to a distributed power grid-connected interface and controlling the connection state of the distributed power grid when an islanding event occurs.

Images