CN107065519B - PMU feedback control signal preprocessing method - Google Patents

Abstract

A PMU feedback control signal preprocessing method is designed to solve the problem that the PMU signal in the feedback control of the wide-area power system has a time lag and the time lag may vary randomly over a large range. A preprocessing module is designed before the PMU feedback control signal enters the stability controller, and a preprocessing method of the random time-lag PMU feedback control signal based on the idea of optimal time-lag matching is proposed. The method includes the key steps of determining the optimal time lag of the controller, reordering random PMU signals, filtering, data interpolation and sampling, total time lag calculation, optimal time-lag feedback control signal selection, and feedback control. The present invention can enable various wide-area feedback stability controllers to still have the preset stable control capability when the time lag of the feedback control signal fluctuates greatly.

Description

A PMU feedback control signal preprocessing method

Technical field:

The invention relates to a large-scale power system wide-area time delay stability control method, and belongs to the technical field of power system stability control.

Background technique:

With the interconnection of regional power grids, the scale of the power system continues to expand, and the problem of stability is becoming more and more serious. How to ensure that the power system under the UHV interconnection of large-scale power grids has strong damping capacity and can maintain stable operation under complex and severe disturbances has become an urgent control problem for wide-area smart grids.

After the interconnection of regional power grids, disturbances or faults may spread to multiple regional power grids, which are widely distributed and have a great impact. The traditional power system stability controller (PSS) that uses local signals as feedback signals is limited by the observability of control signals, and has very limited effect in suppressing low-frequency oscillations in the interval. Phasor Measurement Unit (PMU) based on Global Positioning System (GPS) or Beidou Satellite Navigation System (BDS) timing technology makes the synchronous measurement of power system operating status no longer a problem. At present, the PMU-based power system wide area measurement system ( WAMS) is also constantly improving. The use of PMU synchrophasor data for wide-area power system stability control has significant advantages over the traditional PSS control based on local signals, and clear conclusions have been drawn in a large number of literatures. However, the wide-area and long-distance transmission of the PMU synchrophasor also brings about the time delay problem of the control signal. If the influence of the large fluctuation of random time delay is not properly handled, the stabilization controller will not only work, but may even further deteriorate the power system. stability.

The time delay of the PMU signal fluctuates randomly in a large range, so that the signals sent in sequence by the PMU end become chaotic and disordered when they reach the controller end; in addition, the controllers designed according to different principles often correspond to a specific time when the optimal damping effect is exerted. lag value, that is, the optimal time lag. If the traditional control signal preprocessing method is used, when the time lag fluctuates randomly and in a large range, the maximum stable control ability of the controller cannot be exerted. Therefore, it is necessary to design an appropriate PMU feedback control signal preprocessing method for the large fluctuation of time delay in wide-area stability control.

Invention content:

Aiming at the preprocessing problem of the PMU feedback control signal caused by the large fluctuation of time delay in the stability control of the wide-area power system, the present invention designs a preprocessing method when the PMU signal reaches the stable controller end, so as to ensure that the controller reaches the preset value. stabilizing effect.

The PMU control signal preprocessing method of the present invention includes two stages and seven steps:

(1) Controller design stage:

(1) Determination of the optimal time delay of the controller. Add the designed controller to the simulation analysis model of the power system, set the time delay of the feedback control signal, analyze the dynamic characteristics of the power system after being disturbed under different time delays, calculate the corresponding damping index, and select the maximum damping index corresponding to The delay of τ is taken as the optimal delay of the controller, denoted as τ _m .

(2) Real-time feedback control stage:

(2) Random PMU signal reordering. When the random PMU signal reaches the controller, the signal is marked as Y1, and its corresponding measurement time series is marked as T1. According to the time stamp marked by each PMU signal, it is reordered. The sorted signal is marked as Y2, and the corresponding The time series is marked as T2.

(3) PMU signal filtering. A non-causal filter without delay is used to filter out the high-frequency random noise mixed in the PMU signal Y2. The denoised PMU signal is denoted as Y3, and the corresponding time series is denoted as T3.

(4) PMU signal data interpolation and sampling. According to the requirements of the power system stability controller for the feedback signal input, the PMU signal Y3 is interpolated and sampled, so that the sampling time of the PMU signal is consistent with the operating frequency of the controller, the PMU signal after interpolation and sampling is recorded as Y4, and the corresponding time series is recorded as T4.

(5) Calculate the total time delay of the PMU feedback control signal. Read the real-time GPS timing data _TG , subtract the time stamp in the time series T4 from _TG , and then calculate the total time delay of each sampling point in the PMU signal Y4, and the time delay data sequence is marked as ΔT.

(6) Selection of optimal time-delay feedback control signal. Determine the time delay point ΔT _i that is closest to the optimal time delay τ _m in step (1) in the time delay data sequence ΔT, and select the point Y4 _i in the PMU signal sequence Y4 corresponding to ΔT _i as the feedback at the current _TG time control signal y.

(7) Feedback control. Input the optimal time-delay control signal y screened in step (6) into the power system stability controller.

The advantages of the invention are: it solves the technical problem of time-delay PMU signal preprocessing encountered in the time-delay stability control of the wide-area power system, and innovatively proposes a signal preprocessing method based on the idea of optimal time-delay matching, which can not only be applied The generator excitation device controller shown in FIG. 1 can also be analogously applied to the controllers of other control devices in the power system.

Description of drawings:

Fig. 1 Control structure diagram of wide area power system

2 is a flowchart of the PMU control signal preprocessing method of the present invention

Figure 3 Four-machine two-area power system diagram

Figure 4. Comparison of stability control effects under different fixed time delays

Figure 5 Variation of random time delay with time

Figure 6. Comparison of the effects of signal preprocessing methods on time-delay PMU feedback control signals

Fig.7 Comparison of the influence of time-delay PMU feedback control signal preprocessing on control effect

Detailed ways:

The technical solutions of the present invention are further described below with reference to the accompanying drawings.

Figure 1 shows the control structure diagram of the wide area power system. The power grid connects various power plants and electrical loads that are geographically separated by hundreds or even thousands of kilometers into a network through AC or DC transmission lines and substations. Such a huge power and energy network, which is related to the national economy and people's livelihood, brings convenience and benefits to the overall planning and complementary optimization of energy. After long-term research and application promotion, a large number of controllable and adjustable devices have been used in the power system to enhance the anti-interference and stable operation capabilities of the system. These control devices include: traditional generator excitation devices, valve adjustment devices, high-voltage DC Transmission control devices, emerging flexible AC transmission devices, etc. The control effect of these control devices is not only closely related to the design of hardware technology and control strategy, but also has a great relationship with the support ability of the power system measurement data platform WAMS and the data analysis technology. At present, power plants, 500kV and above substations, and some 220kV substations in my country's power system have been installed with PMUs. PMUs can accurately measure the static and dynamic data of the power system in real time, laying a solid data foundation for system stability analysis and control. The PMU feedback control signal preprocessing method of the present invention is applied to the power system control device end in FIG. 1 as an intermediate module between the PMU signal transmitted by WAMS and the stabilization controller of the control device, that is, the “PMU signal preprocessing” in FIG. 1 . Processing module", to provide guarantee for the normal operation of the control device to stabilize the controller.

The flowchart of the PMU control signal preprocessing method of the present invention is shown in Figure 2, which mainly includes two stages and seven steps:

(1) Controller design stage:

(1) Determination of the optimal time delay of the controller. Add the designed controller to the simulation analysis model of the power system, set the time delay of the feedback control signal, analyze the dynamic characteristics of the power system after being disturbed under different time delays, calculate the corresponding damping index, and select the maximum damping index corresponding to The delay of τ is taken as the optimal delay of the controller, denoted as τ _m .

(2) Real-time feedback control stage:

(2) Random PMU signal reordering. When the random PMU signal reaches the controller, the signal is marked as Y1, and its corresponding measurement time series is marked as T1. According to the time stamp marked by each PMU signal, it is reordered. The sorted signal is marked as Y2, and the corresponding The time series is marked as T2.

(3) PMU signal filtering. A non-causal filter without delay is used to filter out the high-frequency random noise mixed in the PMU signal Y2. The denoised PMU signal is denoted as Y3, and the corresponding time series is denoted as T3.

(4) PMU signal data interpolation and sampling. According to the requirements of the power system stability controller for the feedback signal input, the PMU signal Y3 is interpolated and sampled, so that the sampling time of the PMU signal is consistent with the operating frequency of the controller, the PMU signal after interpolation and sampling is recorded as Y4, and the corresponding time series is recorded as T4.

(5) Calculate the total time delay of the PMU feedback control signal. Read the real-time GPS timing data _TG , subtract the time stamp in the time series T4 from _TG , and then calculate the total time delay of each sampling point in the PMU signal Y4, and the time delay data sequence is marked as ΔT.

(6) Selection of optimal time-delay feedback control signal. Determine the time delay point ΔT _i that is closest to the optimal time delay τ _m in step (1) in the time delay data sequence ΔT, and select the point Y4 _i in the PMU signal sequence Y4 corresponding to ΔT _i as the feedback at the current _TG time control signal y.

(7) Feedback control. Input the optimal time-delay control signal y screened in step (6) into the power system stability controller.

The four-machine, two-area power system shown in Figure 3 is a benchmark test system to study interval low-frequency oscillations. Generators G1 and G2 are located in area 1, generators G3 and G4 are located in area 2, and the two areas are interconnected by long tie lines. In the simulation, it is assumed that the output power of the four generators is 700MW under normal operation. Since the load in area 1 is lighter and the load in area 2 is heavier, the double-circuit tie line 7-9 needs to transmit about 300MW of active power from area 1 at this time. Power goes to zone 2, and node 8 is an intermediate substation for long-distance transmission lines. The analysis of small disturbance stability eigenvalues shows that there are three low-frequency oscillation modes in the system. G1 and G2 in region 1 have interval low-frequency power oscillations relative to G3 and G4 in region 2. The oscillation frequency is 0.5370 Hz and the damping value is less than 0. Since the damping value is negative, once the system is disturbed, the relative power angle of the generator and the power of the tie line will oscillate violently for a long time, as shown in the interval relative power angle curve when there is no controller in Figure 4.

Assuming that a stable controller K is installed on the excitation control device of the generator G2, the dynamic simulation analysis of the system is carried out by setting different fixed time delays. It can be seen from Figure 4 that the stable controller K has the best performance when the time delay is 500ms. Therefore, the optimal time delay is set as τ _m =500ms.

Then, according to the normal distribution characteristics of the communication delay of the PMU signal under the actual operating conditions of the WAMS, the mean value of the two-way signal transmission delay in the feedback control in the simulation test is 200ms, and the standard deviation is 200ms. Figure 5 shows the random delay with time. In the case of changes, it is assumed that the lower limit of the delay is 50ms, and the upper limit of the delay is 1000ms. In order to simulate the influence of noise in the PMU signal, Gaussian white noise is added to the system dynamic simulation measurement data. The simulated random time-delay PMU feedback control signal is transmitted to the excitation control device end on the generator G2, and processed through the following steps:

(1) Random PMU signal reordering. When the random PMU signal reaches the controller, the signal is marked as Y1, and its corresponding measurement time series is marked as T1. According to the time stamp marked by each PMU signal, it is reordered. The sorted signal is marked as Y2, and the corresponding The time series is marked as T2.

(2) PMU signal filtering. A non-causal filter without delay is used to filter out the high-frequency random noise mixed in the PMU signal Y2. The denoised PMU signal is denoted as Y3, and the corresponding time series is denoted as T3.

(3) PMU signal data interpolation and sampling. According to the input requirements of the feedback signal of the power system stability controller, the PMU signal Y3 is interpolated and sampled, so that the sampling time of the PMU signal is consistent with the operating frequency of the controller. The PMU signal after interpolation and sampling is recorded as Y4, and the corresponding time series is recorded as T4.

(4) Calculate the total time delay of the PMU feedback control signal. Reading the GPS timing data _TG , subtracting the time stamp in the time series T4 from _TG , the total time delay of each sampling point in the PMU signal Y4 can be calculated, and the time delay data sequence is marked as ΔT.

(5) Selection of optimal time-delay feedback control signal. Determine the time delay point ΔT _i that is closest to the optimal time delay τ _m in step (1) in the time delay data sequence ΔT, and select the midpoint Y4 _i of the PMU signal sequence Y4 corresponding to ΔT _i as the feedback control at the current _TG time signal y.

Finally, the optimal time-delay control signal screened out in real time is continuously input to the stable controller in the generator G2 excitation control device. After the PMU feedback control signal is preprocessed by this method, the stabilization controller can normally play the damping effect (equivalent to the effect when the fixed time delay is 500ms in Figure 4), even if the random time delay fluctuates randomly in a large range as shown in Figure 5.

In order to further illustrate the importance of this method in wide-area power system stability control, this case also compares it with other three possible PMU signal selection strategies, as follows:

Replacement strategy 1 - receiving sequence: feedback control is performed according to the sequence in which the PMU signal arrives at the stable controller, without re-queuing;

Replacement strategy 2 - sending order: re-queue the received out-of-order PMU signals, and perform feedback control according to the sending order of the PMU end;

Replacement strategy 3 - latest principle: select the latest data for feedback control;

If the time delay is small or remains unchanged, the PMU feedback control signals selected by the three replacement strategies are basically consistent with the method of the present invention. When the time lag is large and fluctuates greatly, the PMU feedback control signals selected by different strategies may be quite different. The effect comparison in this case is shown in Figure 6. It can be seen from Fig. 6 that the time delay of the feedback control signal under the strategy of the present invention "time delay matching" is closer to the optimal time delay of 500ms, and the time delay under the replacement strategy 1 "reception order" and the replacement strategy 3 "latest principle" are relatively It is close to the lower limit of the delay time of 50ms, among which the feedback control signal has more abrupt changes under the "receive order" strategy, and the delay under the replacement strategy 2 "send order" is very close to the upper limit of the delay time of 1000ms.

Figure 7 is the dynamic response diagram of the relative power angle in the system interval corresponding to Figure 6. For the same stable controller, the damping effect under different PMU feedback control signal selection strategies is very different: the damping effect under the "delay matching" strategy is the most The "receive order" and "latest principle" strategies are the next best, and the "send order" strategy makes the system unstable. It can be seen that, in the case of using the PMU signal for feedback control in a random time delay environment with large fluctuations, attention must be paid to the selection strategy of the feedback control signal in the preprocessing link of the random time delay signal. The PMU feedback based on the idea of time delay matching proposed in the present invention The control signal preprocessing method can achieve satisfactory results.

Claims (1)

1. A PMU feedback control signal preprocessing method, comprising the following steps: (1) Determination of the optimal time delay of the controller; the designed controller is added to the power system simulation analysis model, the time delay of the feedback control signal is set, and the dynamic characteristics of the power system after being disturbed under different time delays are analyzed by simulation. Calculate the corresponding damping index, and select the time delay corresponding to the maximum damping index as the optimal time delay of the controller, denoted as τ _m ; (2) Reordering of random PMU signals; when the random PMU signal reaches the controller, the signal is marked as Y1, and its corresponding measurement time series is marked as T1, and according to the time stamp marked by each PMU signal, it is re-sorted and sorted. The latter signal is denoted as Y2, and the corresponding time series is denoted as T2; (3) PMU signal filtering; a non-causal filter without delay is used to filter out the high-frequency random noise mixed in the PMU signal Y2, the denoised PMU signal is recorded as Y3, and the corresponding time series is recorded as T3; (4) Interpolation and sampling of PMU signal data; according to the requirements of the power system stability controller for feedback signal input, the PMU signal Y3 is interpolated and sampled, so that the sampling time of the PMU signal is consistent with the operating frequency of the controller, and the PMU signal after interpolation and sampling is interpolated and sampled. It is recorded as Y4, and the corresponding time series is recorded as T4; (5) Calculate the total time delay of the PMU feedback control signal; read the real-time GPS timing data _TG , subtract the time scale in the time series T4 from _TG , and then calculate the total time of each sampling point in the PMU signal Y4 delay, the time delay data sequence is recorded as ΔT; (6) Selection of the optimal time-delay feedback control signal; determine the time-delay point ΔT _i that is closest to the optimal time-delay τ _m in step (1) in the time-delay data sequence ΔT, and select the PMU signal sequence corresponding to ΔT _i The point Y4 _i in Y4 is used as the optimal time-delay feedback control signal y at the current _TG moment; (7) Feedback control; input the optimal time-delay feedback control signal y screened in step (6) into the power system stability controller.

Images