A maximum power point tracking control system and method for a pump turbine power generation system

Abstract

This invention discloses a maximum power point tracking (MPPT) control system and method for a pump-driven turbine power generation system. It calculates output power by acquiring DC-side voltage and current, and also acquires rotational speed. Within each MPPT cycle, the operating state is identified by the power-speed slope feature G: when operating conditions change rapidly, the weight of the disturbance observation branch is increased; when approaching the maximum power point, the weight of the adaptive scale fuzzy logic control branch is enhanced. The two branches achieve branch current fusion according to the weights constructed using (|G|), outputting the q-axis stator current increment. A dq current closed loop is used to output a PWM drive signal to control the machine-side converter, adjusting the electromagnetic torque of the permanent magnet synchronous generator, thereby achieving efficient MPPT under time-varying head / flow conditions. This invention relies only on measurable power and speed signals, requiring no precise hydraulic model or data-intensive training, resulting in low computational load, ease of embedded implementation, improved energy recovery efficiency, and suppression of power fluctuations and torque pulsations.

Description

Technical Field

[0001] This invention relates to the field of distributed hydropower recovery and power electronic control technology, specifically to a maximum power point tracking (MPPT) control system and method applicable to a pump-as-turbine (PAT) coupled power generation system and a permanent magnet synchronous generator (PMSG). Background Technology

[0002] In urban water supply networks, mountain gravity water supply facilities, and agricultural irrigation systems, there exists a large amount of unused hydraulic pressure. Pump-acted turbines (PATs) are widely considered an economical solution for distributed hydropower recovery due to their low cost, simple structure, and convenient maintenance. However, in actual operation, the flow rate, head, and downstream boundary conditions in the pipeline network exhibit significant time-varying characteristics. The optimal operating point of the PAT will shift with changes in operating conditions. Therefore, the energy recovery efficiency of the PAT-PMSG power generation system is highly dependent on the dynamic and steady-state performance of the MPPT control strategy.

[0003] Among existing MPPT strategies, the Perturbation and Observation (P&O) method is simple in structure and easy to implement in embedded systems, but it is prone to steady-state oscillations near the maximum power point. Fuzzy logic control (FLC) is robust to noise and parameter disturbances, but its rules and range design often rely on experience, and its dynamic response capability is insufficient when operating conditions change rapidly. More complex methods such as model predictive control and sliding mode control have superior performance, but they require high model accuracy and computational resources, which are not compatible with the engineering economy and implementation cost of small PAT systems. Therefore, there is an urgent need for an MPPT control scheme that does not require an accurate hydraulic model, a large amount of training data, or computational resources, and can adaptively balance "rapid optimization" and "low steady-state oscillations". Summary of the Invention

[0004] To address the problems of "difficulty in balancing dynamic response and steady-state fluctuations" and "poor adaptability across operating conditions due to fixed FLC range" in the existing PAT–PMSG system MPPT control, this invention proposes a maximum power point tracking control system and method for pump-driven turbine power generation systems. By identifying slope operating conditions and adaptive weight allocation, the power generation system automatically prioritizes either speed or stability at different operating stages, thereby improving energy recovery efficiency and operational stability.

[0005] The present invention achieves the above-mentioned technical objectives through the following technical means.

[0006] A maximum power point tracking control method for pump-driven turbine power generation systems:

[0007] The signal acquisition unit collects the power P(k) and rotational speed ω(k) within period k, and calculates the power increment respectively. Speed ​​increment Then, the local power-speed slope characteristics are calculated. ;

[0008] The direction of the disturbance is determined by the sign of the power increment. The P&O module outputs a q-axis current increment consistent with the direction of the disturbance, based on the relationship between the applied disturbance and the actual operation. ; Fuzzy controller with ( , Taking a given number as input, the output is obtained through reasoning using a preset membership function and rule base. ;based on Constructing Adaptive Weight Fusion The final output is the q-axis stator current increment. ;in, , These represent the normalized errors for power and speed, respectively.

[0009] The generator-side current control unit uses a dq current closed-loop output PWM drive signal to control the generator-side converter, thereby adjusting the electromagnetic torque of the permanent magnet synchronous generator, enabling the PAT-PMSG system to achieve maximum power point tracking when the head / flow rate changes; the dq current closed loop outputs a q-axis current reference value based on the incremental output of the q-axis stator current.

[0010] Furthermore, when the speed increment |Δω| is less than the threshold εω, the local power-speed slope characteristic G(k)=G(k-1) is maintained.

[0011] Furthermore, the q-axis current increment output by the P&O module ,in, Indicates the effective current step size, and , This indicates the current step size during rapid adjustment. This indicates the current step size during fine-tuning. This indicates the step order.

[0012] Furthermore, the normalization errors of power and speed are obtained through adaptive scaling.

[0013] Furthermore, the process for obtaining the power normalization error is as follows:

[0014] Slow average of maintenance power Slow average speed Construct error signal , Maintain maximum leakage value Minimum leakage ;in, Indicates the virtual optimal rotational speed;

[0015] The center of power leakage extrema and half span :

[0016]

[0017] ,

[0018] in, This indicates the minimum value set.

[0019] Constructing normalized error .

[0020] Furthermore, maintain the maximum leakage value. Minimum leakage Specifically:

[0021]

[0022]

[0023] in, This indicates the leakage factor.

[0024] Furthermore, the virtual optimal rotational speed is recursively updated based on the perturbation direction: , ,in, This refers to the preset or adaptively updated speed disturbance. Indicates the effective speed step size. Indicates the step order. This indicates the speed step size during rapid adjustment. This indicates the speed step size during fine-tuning.

[0025] Furthermore, based on Constructing Adaptive Weight Fusion Specifically: when hour, =0; when hour, =1; when hour, ,in, and All are preset thresholds.

[0026] Furthermore, the dq current closed loop specifically includes:

[0027] (1) PI regulation is used for the d-axis and q-axis currents respectively;

[0028] (2) Introduce cross-coupling compensation and back EMF feedforward to improve bandwidth and suppress coupling;

[0029] (3) When the voltage is limited or the modulation is saturated, the limiting residual is fed back to the integral stage to achieve anti-saturation;

[0030] (4) The reference voltage vector is input into SVPWM after amplitude constraint to generate a three-phase duty cycle to drive the machine-side converter;

[0031] (5) The q-axis current reference value is determined by "the q-axis current reference value of the previous cycle iq* + "It is obtained through recursion."

[0032] A maximum power point tracking control system for a pump-driven turbine power generation system includes:

[0033] The hydroelectric generator set and power conversion unit includes a pump-driven turbine (PAT), a permanent magnet synchronous generator (PMSG) coaxially connected to the PAT, a generator-side converter, and a DC load. The three-phase output of the PMSG is rectified / converted by the generator-side converter and then connected to the DC load to realize the conversion of water energy into DC electrical energy.

[0034] The signal acquisition unit obtains the DC side voltage and current of the permanent magnet synchronous generator in real time and calculates the output power; at the same time, it obtains the rotational speed of the permanent magnet synchronous generator.

[0035] The hybrid MPPT control unit outputs the q-axis stator current increment during each MPPT update cycle;

[0036] The machine-side current control unit adopts a dq current closed loop and outputs a PWM drive signal to control the machine-side converter.

[0037] Compared with existing MPPT strategies based on a single P&O or a single FLC, the present invention has at least the following advantages:

[0038] (1) Dynamic-steady-state adaptive consideration: The degree of change of operating conditions is identified by the power-speed slope characteristics. The P&O weight is increased in the rapid change stage to improve the response speed, and the FLC weight is increased in the stage close to the maximum power point to reduce steady-state fluctuations.

[0039] (2) Adaptive range across operating conditions: Through adaptive scale normalization, the input domain of FLC can be automatically scaled according to the operating conditions, reducing the dependence on fixed range and empirical parameter tuning;

[0040] (3) Improved energy utilization efficiency: In the simulation comparison of typical working conditions, the MPPT tracking efficiency can be improved from about 96.8% of the traditional adaptive P&O to about 98.5%, the steady-state power fluctuation is reduced by about 35%, and the robustness under noise and parameter disturbances is improved; under the estimation of typical daily flow curves, the daily power generation can be increased by about 3-6%. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the overall structure of the PAT–PMSG power generation system of the present invention;

[0042] Figure 2 This is the overall flowchart of the P&O–FLC adaptive weighted hybrid MPPT control unit of the present invention;

[0043] Figure 3 This is a schematic diagram of the output surface of the present invention. Detailed Implementation

[0044] The technical solution of the present invention will be further described below with reference to the accompanying drawings. It should be understood that the following embodiments are used to explain the present invention and not to limit the scope of protection of the present invention.

[0045] like Figure 1 As shown, a maximum power point tracking control system for a pump-driven turbine power generation system includes a hydroelectric generator set and power conversion unit, a signal acquisition unit, a hybrid MPPT control unit, and a turbine-side current control unit (MSC).

[0046] The hydroelectric generator set and power conversion unit includes a pump-driven turbine (PAT), a permanent magnet synchronous generator (PMSG) coaxially connected to the PAT, a generator-side converter, and a DC load. The three-phase output of the PMSG is rectified / converted by the generator-side converter and then connected to the DC load to realize the conversion of water energy into DC electrical energy.

[0047] The signal acquisition unit obtains the DC-side voltage U of the permanent magnet synchronous generator in real time. dc Current I dc And calculate the output power P=U dc ·I dc (The effect of switching ripple can be suppressed by low-pass filtering); at the same time, the rotational speed ω of the permanent magnet synchronous generator is obtained.

[0048] like Figure 2 As shown, the hybrid MPPT control unit is used to complete the following within each MPPT update cycle:

[0049] (1) Calculation of slope characteristic G

[0050] The signal acquisition unit obtains the power P(k) and rotational speed ω(k) within period k, and calculates the power increment. Speed ​​increment Then, the local power-speed slope characteristics are calculated. When |Δω| is less than the threshold εω, in order to suppress the influence of rotational speed noise, the local power-speed slope characteristic G(k)=G(k-1) is maintained.

[0051] (2) Increment of q-axis current output from P&O branch (Quick optimization)

[0052] The perturbation direction dir is determined based on the sign (positive or negative) of ΔP. The P&O module outputs the q-axis current increment consistent with the perturbation direction dir, according to the relationship between the applied perturbation and the actual operation. , ,in, Indicates the effective current step size, and , This indicates the current step size during rapid adjustment. The value indicates the current step size during fine adjustment. In an optional embodiment, the degree of proximity to the maximum power is determined based on G (the larger the value of G, the farther away from the maximum power; the smaller the value of G, the closer to the maximum power). Thus, two current step sizes (Stage-flag(k)) can be set, either "fast" or "fine" or an effective current step size that decays with the degree of saturation, to accelerate the search when far from the optimal point and reduce disturbances when approaching the optimal point.

[0053] The P&O module recursively updates the virtual optimal rotational speed based on the disturbance direction dir. , ,in The preset or adaptively updated speed disturbance is used to characterize the speed near the currently estimated maximum power point and to assist in the search of the P&O branch; Indicates the effective speed step size. Indicates the step order. This indicates the speed step size during rapid adjustment. This indicates the speed step size during fine-tuning.

[0054] (3) Adaptive scaling normalization and FLC branch output q-axis current increment (Fine-tuning)

[0055] To reduce the limitations of fixed range on FLC, adaptive scaling normalization is introduced:

[0056] Slow average of maintenance power Slow average speed (The slow mean is obtained by inputting power or speed into a first-order discrete low-pass filter), constructing the error signal. , , Indicates the virtual optimal rotational speed;

[0057] Maintain maximum leakage value Minimum leakage The update rules are as follows:

[0058]

[0059]

[0060] in, Indicates the leakage factor;

[0061] Then the center of the power leakage extremum is generated. and half span :

[0062]

[0063] ,

[0064] in, This indicates the minimum value set.

[0065] Similarly, the center of the extreme value of speed leakage is obtained. and half span ;

[0066] Constructing normalized error:

[0067] ,

[0068] Fuzzy controller with ( , Taking a given number as input, the output is obtained through reasoning using a preset membership function and rule base. It can be scaled using a scaling factor to match the current regulation level.

[0069] In a preferred embodiment, the two input variables of the fuzzy controller are and One output variable is The normalized quantity, input / output universe of discourse are both set to [-1, 1], membership functions can be triangular or trapezoidal, and fuzzy subsets can be {NB, NM, NS, ZO, PS, PM, PB}; Mamdani inference and centroid method are used to resolve fuzziness and obtain a definite output, such as... Figure 3 As shown, the rule table can be constructed based on the "joint trend of power deviation and speed deviation", so that the system outputs a smaller adjustment amount when approaching the maximum power point, thereby reducing steady-state fluctuations.

[0070] Table 1 Rules Table

[0071]

[0072] (4) Based on Adaptive weight fusion

[0073] Constructing adaptive weights :when hour, =0; when hour, =1; when hour, ,in, and For the preset threshold, take an empirical value; in Greater than the threshold At that time, Proportional decay, Reaching the threshold Freeze The disturbance is mitigated to suppress steady-state power oscillations near the maximum power point; the final output is the q-axis stator current increment.

[0074]

[0075] like Figure 2 As shown, the generator-side current control unit (MSC) employs a dq current closed loop (including cross-coupling compensation, back EMF feedforward, and anti-saturation backflow) to output a PWM drive signal to control the generator-side converter, thereby adjusting the electromagnetic torque of the permanent magnet synchronous generator. This enables the PAT–PMSG system to achieve maximum power point tracking when the head / flow rate changes. The dq current closed loop specifically includes:

[0076] (1) PI regulation is used for the d-axis and q-axis currents respectively;

[0077] (2) Introduce cross-coupling compensation and back EMF feedforward to improve bandwidth and suppress coupling;

[0078] (3) When the voltage is limited or the modulation is saturated, the limiting residual is fed back to the integral stage to achieve anti-saturation;

[0079] (4) The reference voltage vector is input into SVPWM after amplitude constraint to generate a three-phase duty cycle to drive the machine-side converter;

[0080] (5) The q-axis current reference value iq* is obtained from "the previous cycle iq* + "It is obtained recursively, and the slope / amplitude constraints of the stator current can be superimposed to avoid the impact of sudden current changes on the machinery and busbars."

[0081] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A maximum power point tracking control method for a pump-driven turbine power generation system, characterized in that: The signal acquisition unit collects the power P(k) and rotational speed ω(k) within period k, and calculates the power increment respectively. Speed ​​increment Then, the local power-speed slope characteristics are calculated. ; The direction of the disturbance is determined by the sign of the power increment. The P&O module outputs a q-axis current increment consistent with the direction of the disturbance, based on the relationship between the applied disturbance and the actual operation. ; Fuzzy controller with ( , Taking a given number as input, the output is obtained through reasoning using a preset membership function and rule base. ;based on Constructing Adaptive Weight Fusion The final output is the q-axis stator current increment. ;in, , These represent the normalized errors for power and speed, respectively. The generator-side current control unit uses a dq current closed-loop output PWM drive signal to control the generator-side converter, thereby adjusting the electromagnetic torque of the permanent magnet synchronous generator, enabling the power generation system to achieve maximum power point tracking when the head / flow rate changes; the dq current closed loop outputs a q-axis current reference value based on the incremental output of the q-axis stator current.

2. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 1, characterized in that, When the speed increment |Δω| is less than the threshold εω, the local power-speed slope characteristic G(k)=G(k-1) is maintained.

3. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 1, characterized in that, The q-axis current increment output by the P&O module ,in, Indicates the effective current step size, and , This indicates the current step size during rapid adjustment. This indicates the current step size during fine-tuning. This indicates the step order.

4. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 1, characterized in that, The normalization errors of power and speed are obtained through adaptive scaling.

5. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 4, characterized in that, The process for obtaining the power normalization error is as follows: Slow average of maintenance power Slow average speed Construct error signal , Maintain maximum leakage value Minimum leakage ;in, Indicates the virtual optimal rotational speed; The center of power leakage extrema and half span : ， in, This indicates the minimum value set. Constructing normalized error .

6. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 5, characterized in that, Maintain maximum leakage value Minimum leakage Specifically: in, This indicates the leakage factor.

7. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 5, characterized in that, The virtual optimal rotational speed is updated recursively based on the perturbation direction: , ,in, This refers to the preset or adaptively updated speed disturbance. Indicates the effective speed step size. Indicates the step order. This indicates the speed step size during rapid adjustment. This indicates the speed step size during fine-tuning.

8. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 1, characterized in that, based on Constructing Adaptive Weight Fusion Specifically: when hour, =0; when hour, =1; when hour, ,in, and All are preset thresholds.

9. The maximum power point tracking control method for a pump-driven turbine power generation system according to claim 1, characterized in that, The dq current closed loop specifically includes: (1) PI regulation is used for the d-axis and q-axis currents respectively; (2) Introduce cross-coupling compensation and back EMF feedforward to improve bandwidth and suppress coupling; (3) When the voltage is limited or the modulation is saturated, the limiting residual is fed back to the integral stage to achieve anti-saturation; (4) The reference voltage vector is input into SVPWM after amplitude constraint to generate a three-phase duty cycle to drive the machine-side converter; (5) The q-axis current reference value is derived from "the q-axis current reference value of the previous cycle iq* + "It is obtained through recursion." 10. A system based on the maximum power point tracking control method for a pump-driven turbine power generation system according to any one of claims 1-9, characterized in that, include: The hydroelectric generator set and power conversion unit includes a pump-driven turbine (PAT), a permanent magnet synchronous generator (PMSG) coaxially connected to the PAT, a generator-side converter, and a DC load. The three-phase output of the PMSG is rectified / converted by the generator-side converter and then connected to the DC load to realize the conversion of water energy into DC electrical energy. The signal acquisition unit obtains the DC side voltage and current of the permanent magnet synchronous generator in real time and calculates the output power; at the same time, it obtains the rotational speed of the permanent magnet synchronous generator. The hybrid MPPT control unit outputs the q-axis stator current increment during each MPPT update cycle; The machine-side current control unit adopts a dq current closed loop and outputs a PWM drive signal to control the machine-side converter.

Images