Flywheel energy storage charging control method

Abstract

The invention is suitable for the technical field of energy storage, and provides a flywheel energy storage charging control method, which comprises the following steps of: 1, modeling a high-speed permanent magnet synchronous motor, establishing a flywheel energy storage system type, simplifying a PMSM mathematical model, constructing a PMSM model under an ABC coordinate system, and converting to a PMSM model under a d-q coordinate system; and 2, performing model prediction current control on the MRAS, performing double-vector model prediction current control, performing model reference adaptive method, performing parameter identification precision verification based on the MRAS, and adopting a parameter step-by-step identification method based on the MRAS to solve the problem that the model prediction current control height depends on prediction model accuracy, so that the accuracy of the model prediction current control is not influenced. The problem that prediction current control performance of a double-vector model is reduced due to mismatching of model parameters is solved, the identification precision of rotor flux linkage, stator resistance and stator inductance is high, meanwhile, the speed regulation range is widened, the dynamic response characteristic is improved, and the motor model can reflect the actual situation more accurately.

Description

Technical Field

[0001] The present invention relates to the field of energy storage technology, and more specifically, to a flywheel energy storage charging control method. Background Art

[0002] The controller used in model predictive control requires a discrete model of the motor. When the speed of the driven object is high, if the carrier ratio of the converter driving the motor is too low, the nonlinear change of the motor will be obvious, which will reduce the dynamic and steady-state performance of the motor. The current ripple and current amplitude fluctuation will be large, affecting the control effect of the motor and increasing the loss of the system. In addition, long-term use will cause changes in parameters such as motor flux and moment of inertia, while model predictive control is highly dependent on the accuracy of the mathematical model of the controlled object, especially the dual-vector model predictive current control. Model parameter mismatch will reduce the accuracy of future state prediction, thereby affecting the accuracy of the optimal voltage vector selected to minimize the cost function.

[0003] In the PMSM system, the speed outer loop control is very important, and the proportional integral (PI) control strategy is traditionally used. However, the PI speed loop controller is extremely sensitive to changes in load and speed. This method not only has insufficient dynamic response capabilities, but also has poor adaptability when facing internal and external disturbances in the system. The sliding mode control technology has gradually replaced the traditional PI control due to its simple structure, rapid response, and strong robustness, and is widely used in PMSM speed control. Summary of the invention

[0004] In view of the deficiencies in the prior art, an object of the present invention is to provide a flywheel energy storage charging control method.

[0005] To achieve the above object, the present invention provides the following technical solution: a flywheel energy storage charging control method, comprising the following steps:

[0006] Step 1: Modeling of high-speed permanent magnet synchronous motor, establishing flywheel energy storage system type, simplifying PMSM mathematical model, constructing PMSM model in ABC coordinate system, and converting to PMSM model in dq coordinate system;

[0007] Step 2: Model predictive current control steps for MRAS, dual-vector model predictive current control, model reference adaptive method, and verification of MRAS parameter identification accuracy;

[0008] Step 3: Design steps of the speed outer loop based on sliding mode control, design of the sliding mode control switching surface, selection of the sliding mode control reaching law, and analysis of the speed outer loop speed regulation performance.

[0009] Preferably, the first step: establishing a flywheel energy storage system type, establishing a flywheel energy storage system type flywheel energy storage device stores kinetic energy by rotating the flywheel at a high speed, and its storage energy can be represented by a flywheel kinetic energy equation.

[0010] Preferably, the first step is to simplify the PMSM mathematical model. Accurately controlling the speed and torque of the permanent magnet synchronous motor is crucial to the charging and discharging of the flywheel energy storage system, and its mathematical model can accurately reflect the dynamic and static characteristics of the motor. Now, the PMSM is mathematically modeled. Based on its nonlinear and strong coupling characteristics, the following simplified assumptions are usually made:

[0011] (1) Assume that the PMSM magnetic circuit is approximately linear, ignoring core saturation, hysteresis and eddy current losses;

[0012] (2) The three-phase structure of the PMSM stator is the same, the current is sinusoidally distributed, and high-order harmonics are ignored;

[0013] (3) The magnetic field of a permanent magnet is sinusoidally distributed.

[0014] Preferably, the second step: the dual vector model predicts current control to output two voltage vectors in a single cycle and reasonably allocate the duty cycle, which has a larger output voltage amplitude, a wider angle range and better performance than single vector control. Based on FCS-MPC, the current prediction value under the switching state of the inverter is traversed, the current prediction error is evaluated with the cost function, and the switching state that minimizes the dq axis current prediction error is selected for output in the next cycle.

[0015] Preferably, the second step: MRAS includes three parts: adjustable model, reference model and adaptive law. Determining the parameter adaptive law is the key to its high-performance system. The system designed based on Popov's hyperstability theory can be regarded as a nonlinear feedback system, which consists of linear time-invariant and nonlinear time-varying links. Therefore, this theory is used to design the adaptive law;

[0016] The stator resistance and inductance are first measured to identify the rotor flux, and then re-measured after the flux is stable. Finally, these parameters are used to optimize the deadbeat current predictive controller.

[0017] Preferably, the second step: according to the parameter step-by-step identification method, the identification waveforms of the rotor flux, stator resistance and stator inductance can be obtained, and the motor is first started at a given speed of 3000 rpm.

[0018] Preferably, the second step: based on the MRAS parameter identification accuracy verification, the dynamic quality of the three-phase permanent magnet synchronous motor mainly depends on the selected approach rate. When the exponential approach rate is selected, the control system can obtain good dynamic performance.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] (1) A flywheel energy storage charging control method adopts a parameter step-by-step identification method based on MRAS for the current inner loop, which solves the problem that the model prediction current control is highly dependent on the accuracy of the prediction model and the model parameter mismatch leads to the degradation of the dual vector model prediction current control performance. The rotor flux, stator resistance and stator inductance are identified with high accuracy, while the speed regulation range is increased and the dynamic response characteristics are improved, so that the motor model can more accurately reflect the actual situation.

[0021] (2) A flywheel energy storage charging control method is proposed. For the speed outer loop, a speed controller based on sliding mode theory is designed. That is, a sliding mode speed controller based on variable exponential reaching law is used to replace the traditional PI speed loop. Compared with the PI controller, the sliding mode controller is more applicable in nonlinear control environments, and can make the actual speed of the motor track the given speed more smoothly, and can improve the robustness and anti-interference ability of the system, so that the system has better stability and dynamic quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

[0023] Figure 1 It is the structural diagram of MRAS of the present invention;

[0024] Figure 2 This is a control block diagram of the MRAS current inner loop and the SMC speed outer loop of the present invention;

[0025] Figure 3 It is the parameter identification waveform diagram of the present invention;

[0026] Figure 4 This is a curve diagram of the motor speed change of the present invention. DETAILED DESCRIPTION

[0027] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but it cannot be understood as a limitation on the scope of protection of the present invention.

[0028] See also Figure 1-4 The present invention provides a technical solution: a flywheel energy storage charging control method, comprising the following steps:

[0029] High-speed permanent magnet synchronous motor modeling steps

[0030] Step 1: Establish the flywheel energy storage system model

[0031] The flywheel energy storage device stores kinetic energy by rotating the flywheel at high speed. Its storage energy can be expressed by the flywheel kinetic energy equation:

[0032] In the formula, is the moment of inertia of the rotor of the flywheel energy storage system, is the angular velocity of the flywheel rotor. From formula (1), it can be seen that the flywheel energy storage is proportional to the moment of inertia and the square of the rotation speed, but the rotation speed is limited by the strength of the flywheel structure and should not be too high. Therefore, in order to ensure fast charging, the flywheel speed is usually set within a specific range. At this time, the energy that the system can store or release is:

[0033] In the formula, the maximum and minimum operating speeds of the system are defined, and the concepts of discharge depth and flywheel charge state are introduced.

[0034] The flywheel energy storage system usually selects a maximum speed of 2-3 times the minimum speed. At that time, the system's discharge depth can be close to 90%, and at that time, the discharge depth can reach 75%, far exceeding that of chemical batteries. Formula (4) represents the speed of the flywheel rotor at time t. At that time, the flywheel energy storage system SOC = 1; at that time, the system SOC = 0.

[0035] Step 2: Simplify the PMSM mathematical model

[0036] Precise control of the speed and torque of a permanent magnet synchronous motor is crucial to the charging and discharging of a flywheel energy storage system. Its mathematical model can accurately reflect the dynamic and static characteristics of the motor. Now, based on its nonlinear and strongly coupled characteristics, the following simplified assumptions are usually made for the mathematical modeling of PMSM:

[0037] (1) Assume that the PMSM magnetic circuit is approximately linear, ignoring core saturation, hysteresis and eddy current losses;

[0038] (2) The three-phase structure of the PMSM stator is the same, the current is sinusoidally distributed, and high-order harmonics are ignored;

[0039] (3) The magnetic field of a permanent magnet is sinusoidally distributed.

[0040] Step 3: Construct PMSM model in ABC coordinate system

[0041] Motor stator three-phase voltage equation:

[0042]

[0043] The magnetic flux equation is:

[0044]

[0045] In formula (5) and (6), u a ,u b ,u c are the voltages of stator ABC, i a ,i b ,i c are the three-phase currents of the stator, ψ a , ψ b , ψc They are the stator three-phase flux, L aa , L bb , L cc They are the stator three-phase winding self-inductance, M ab =M ba =M ca =M ac =M bc =M cb is the mutual inductance of each phase winding, ψ f is the permanent magnet flux.

[0046] The voltage and flux equations of PMSM in the ABC coordinate system are complex and are often simplified to the d and q coordinate systems through coordinate transformation.

[0047] Step 4: Convert the PMSM model to the dq coordinate system

[0048] Voltage equation:

[0049]

[0050] In the formula, u d and u q are the components of the stator voltage on the d-axis and q-axis respectively; i d ,i q , ψ d , ψ q are the stator current and flux respectively. d and L q is the equivalent inductance of the stator winding. e is the rotor electrical angular velocity.

[0051] The magnetic flux equation is:

[0052]

[0053] The electromagnetic torque equation is:

[0054]

[0055] Where, T e is the electromagnetic torque of the motor, and p is the number of motor pole pairs.

[0056] The mechanical motion equation is:

[0057]

[0058] In the formula, ω m is the mechanical angular velocity of the motor, J is the moment of inertia, B is the damping coefficient, T L is the load torque.

[0059] When the flywheel energy storage system is charged, ignoring loads such as friction, the flywheel dynamic equation can be established as follows:

[0060]

[0061] 2. Model prediction current control steps based on MRAS

[0062] Step 1: Dual-vector model predictive current control

[0063] The dual-vector model predictive current control outputs two voltage vectors in a single cycle and reasonably allocates the duty cycle. Compared with single-vector control, the output voltage amplitude is larger, the angle range is wider, and the performance is better. Based on FCS-MPC, the current prediction value under the switching state of the inverter is traversed, and the current prediction error is evaluated by the cost function. The switching state that minimizes the dq axis current prediction error is selected for output in the next cycle. The cost function is defined as:

[0064]

[0065] and are the dq axis current reference values ​​respectively. d (k+1) and i q (k+1) are the predicted values ​​of dq axis current at time k+1 respectively.

[0066] In the dual-vector model predictive current control, eight voltage vectors are substituted into the cost function, and the one with the minimum value is selected as the first optimal voltage vector. To simplify the selection of the second voltage vector, only the two adjacent vectors and the zero vector of the first voltage vector are considered, for a total of three combinations. When the first voltage vector is V3, the possible choices for the second voltage vector are V2, V4 and V0 (or V7). The optimal combination is selected based on the principle of minimizing the cost function, and the dq axis voltage is expressed as:

[0067]

[0068] Where, t represents the action time of the first voltage vector, T s is the sampling time, so the second voltage vector action time is (T s -t). The current prediction model is expressed by the current change rate as:

[0069]

[0070] In the formula, s d1 ,s q1 ,s d2 ,s q2 are the change rates of the dq axis current components under the first and second voltage vectors, respectively. The change rate of the dq axis current under the zero vector is:

[0071]

[0072] Then, the dq axis current change rate under the action of the first and second voltage vectors is expressed as:

[0073]

[0074] Based on the q-axis current deadbeat tracking, the expression for the first voltage vector action time t is:

[0075]

[0076] Step 2: Model Reference Adaptive Method

[0077] MRAS consists of three parts: adjustable model, reference model and adaptive law. Determining the parameter adaptive law is the key to its high-performance system. The system designed based on Popov's hyperstability theory can be regarded as a nonlinear feedback system, which consists of linear time-invariant and nonlinear time-varying links. Therefore, this theory is used to design the adaptive law.

[0078] First, the stator resistance and inductance are measured to identify the rotor flux. After the flux is stable, the stator resistance and inductance are re-measured based on this. Finally, these parameters are used to optimize the deadbeat current prediction controller. The MRAS parameter identification block diagram is shown in the figure below. Figure 1 shown.

[0079] (1) Design of reference model and adjustable model of rotor flux

[0080] The motor current is selected as the state variable. A surface-mounted permanent magnet synchronous motor is used, and the dq axis inductance is equal, that is, L d =L q =L, rewrite the above equation (7) into the current state equation as shown in the following equation (17):

[0081]

[0082] The rotor flux is identified according to the stator inductance and stator resistance values ​​provided in the motor parameter manual, and the current state equation without the parameters to be identified is used as the reference model. Since the rotor flux is only related to the q-axis current, according to the above formula (17), the state equation for the q-axis current can be derived as shown in the following formula (18):

[0083]

[0084] The above formula (18) is used as the reference model, and the adjustable model is shown in the following formula (19):

[0085]

[0086] (2) Adaptive law design of rotor flux

[0087] Subtracting equation (19) from equation (18) yields the following equation (20):

[0088]

[0089] In the formula, To identify the q-axis current, To identify the flux linkage of permanent magnets.

[0090] The adaptive law of the rotor flux is specifically shown in the following equation (21):

[0091]

[0092] In the formula, k′ p and k i ′ are the proportional and integral parameters of the PI controller, and both are greater than 0. yes The initial value of .

[0093] When the rotor flux adaptive law is stable, the final rotor flux identification value can be obtained.

[0094] (3) Design of reference model and adjustable model of stator inductance and stator resistance

[0095] The rotor flux identification value obtained above is used to identify the stator inductance and stator resistance. The current state equation of the stator inductance and stator resistance can be obtained through the above formula (17), which is the reference model, as shown in the following formula (22):

[0096]

[0097] make Then the above formula (22) can be simplified as:

[0098] pI=AI+BU+C (24)

[0099] In the formula,

[0100] The adjustable model can be obtained by the above formula (22), as shown in the following formula (24):

[0101]

[0102] Then the above formula (24) can be simplified as:

[0103]

[0104] In the formula,

[0105] (4) Adaptive law design of stator inductance and stator resistance

[0106] Subtracting equation (25) from equation (23) yields:

[0107]

[0108] In the formula,

[0109] make The above formula (27) can be simplified as shown in the following formula (28):

[0110] pe=Ae-W (28)

[0111] The adaptive law of stator resistance and stator inductance is shown in the following equation (29):

[0112]

[0113] In the formula, k″ p , k″ i , k″′ p , k″′ i Both are greater than 0.

[0114] When the adaptive laws of stator inductance and stator resistance are stable, the final identification values ​​of stator inductance and stator resistance can be obtained.

[0115] Step 3: Verification of MRAS parameter identification accuracy

[0116] The specific simulation motor parameters are shown in Table 1. Unless otherwise specified, the experimental motor also has these parameters.

[0117] Table 1 Motor parameters

[0118]

[0119] According to the parameter step-by-step identification method, the identification waveforms of the rotor flux, stator resistance and stator inductance can be obtained. First, the motor is started at a speed of 3000rpm. Figure 3 The rotor flux, stator resistance and stator inductance are shown from the beginning of identification through the adjustment time to the final stable time and convergence value. As shown in Table 2, in general, the MRAS step-by-step identification method can efficiently and accurately identify the motor parameters.

[0120] Table 2 Parameter identification accuracy

[0121]

[0122] 3. Design steps of speed outer loop based on sliding mode control

[0123] For PMSM, id = 0 vector control method, the system control effect is better. Equations (7), (9), and (10) can be simplified as:

[0124]

[0125] Define the following system state variables:

[0126]

[0127] In the formula, is the reference speed of the motor, which is generally a constant; ω m is the actual motor speed.

[0128] According to equations (30) and (31):

[0129]

[0130] Define u d =i q , formula (32) is expressed as:

[0131]

[0132] Step 1: Design of sliding mode control switching surface

[0133] Define the sliding surface function:

[0134] s1=cx1+x2 (34)

[0135] In the formula, c>0 is the sliding mode gain coefficient to be designed.

[0136] Taking the derivative of equation (34), we can get:

[0137]

[0138] Step 2: Selection of the reaching law for sliding mode control

[0139] The dynamic quality of the three-phase permanent magnet synchronous motor mainly depends on the selected approach rate. When the exponential approach rate is selected, the control system can obtain good dynamic performance.

[0140] The exponential reaching law is defined as follows:

[0141]

[0142] When s(x)>0, Conclusion

[0143]

[0144] When s(x)<0, Conclusion

[0145]

[0146] Where s0 is the value of the switching function s(s(t)) of the system at the initial time (t=0).

[0147] At this time, the expression of the sliding mode controller is:

[0148]

[0149] The reference current of the q axis is obtained as:

[0150]

[0151] Select the Lyapunov function shown in equation (41):

[0152]

[0153] After taking the derivative of the above formula:

[0154]

[0155] If and only if x1=0, The sliding mode reachability condition is met.

[0156] Step 3: Speed ​​outer ring speed regulation performance analysis

[0157] In order to prove that the speed regulation performance and ability to suppress chattering of the designed sliding mode controller are better than those of the traditional PI controller, a PMSM speed tracking experiment is carried out. The speed is set to 2000r / min and the simulation time is set to 0.4s. Figure 4 The motor speed response diagram under traditional PI control and sliding mode control is shown in Figure 2. By comparing the two figures, it can be seen that both traditional PI control and sliding mode control can start normally under no-load starting. Compared with PI control, the rise time and adjustment time of SMC control are both shorter than those of PI control. This shows that SMC control has faster response speed, better control accuracy and stronger robustness.

[0158] Figure 4 The results also include the response curves of the motor speed change under PI control and SMC control after the sudden load at 0.2s. It can be seen that the SMC control method has a stronger speed regulation ability under load conditions. Compared with PI control, the adjustment time of SMC control during the recovery process after the load drops is shorter than that of PI control, and the overshoot generated during the recovery process is much smaller than that of PI control. In summary, the SMC control method has a stronger speed regulation ability under load conditions, can better cope with sudden load disturbances in actual control applications, and has higher control performance.

[0159] 1. For the current inner loop, the dual-vector model predictive current control is first selected, and on this basis, the parameter identification technology based on MRAS is added to design the adjustable model, reference model, and adaptive law of the rotor flux, stator inductance, and stator resistance to form a step-by-step identification. The problem of performance degradation of the dual-vector model current predictive control when the parameters are mismatched is solved, which not only maintains the fast response characteristics of FCS-MPC, but also enhances the transient and steady-state performance of the system under parameter mismatch conditions.

[0160] 2. For the outer speed loop, a speed controller based on sliding mode theory is designed to replace the traditional PI speed loop. First, the state variables of the system are determined, and then the sliding mode control plane is designed, and the variable exponential reaching law is adopted. The sliding mode variable structure speed controller can make the actual speed of the motor track the given speed more smoothly, improve the system performance and convergence effect, and is more suitable for the nonlinear control environment of the flywheel energy storage system, and can improve the robustness of the system.

[0161] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the knowledge scope of ordinary technicians in the technical field without departing from the purpose of the present invention.

Claims

1. A flywheel energy storage charging control method, characterized in that: The following steps are involved: Step 1: Modeling of high-speed permanent magnet synchronous motor, establishing flywheel energy storage system type, simplifying PMSM mathematical model, constructing PMSM model in ABC coordinate system, and converting to PMSM model in dq coordinate system; Step 2: Model predictive current control steps for MRAS, dual-vector model predictive current control, model reference adaptive method, and verification of MRAS parameter identification accuracy; Step 3: Design steps of the speed outer loop based on sliding mode control, design of the sliding mode control switching surface, selection of the sliding mode control reaching law, and analysis of the speed outer loop speed regulation performance.

2. A flywheel energy storage charging control method according to claim 1, characterized in that: The first step is to establish a flywheel energy storage system. The flywheel energy storage device stores kinetic energy by rotating the flywheel at high speed, and the storage energy can be represented by the flywheel kinetic energy equation.

3. A flywheel energy storage charging control method according to claim 2, characterized in that: The first step is to simplify the PMSM mathematical model. Accurately controlling the speed and torque of the permanent magnet synchronous motor is crucial to the charging and discharging of the flywheel energy storage system. Its mathematical model can accurately reflect the dynamic and static characteristics of the motor. Now, based on the nonlinear and strong coupling characteristics of PMSM, the following simplified assumptions are usually made: (1) Assume that the PMSM magnetic circuit is approximately linear, ignoring core saturation, hysteresis and eddy current losses; (2) The three-phase structure of the PMSM stator is the same, the current is sinusoidally distributed, and high-order harmonics are ignored; (3) The magnetic field of a permanent magnet is sinusoidally distributed.

4. A flywheel energy storage charging control method according to claim 3, characterized in that: The second step: dual vector model predictive current control outputs two voltage vectors in a single cycle and reasonably allocates the duty cycle. Compared with single vector control, the output voltage amplitude is larger, the angle range is wider, and the performance is better. Based on FCS-MPC, the current prediction value under the switching state of the inverter is traversed, the current prediction error is evaluated with the cost function, and the switching state that minimizes the dq axis current prediction error is selected for output in the next cycle.

5. A flywheel energy storage charging control method according to claim 4, characterized in that: The second step: MRAS includes three parts: adjustable model, reference model and adaptive law. Determining the parameter adaptive law is the key to becoming a high-performance system. The system designed based on Popov's hyperstability theory can be regarded as a nonlinear feedback system, which consists of linear time-invariant and nonlinear time-varying links. Therefore, this theory is used to design the adaptive law; The stator resistance and inductance are first measured to identify the rotor flux, and then re-measured after the flux is stable. Finally, these parameters are used to optimize the deadbeat current predictive controller.

6. A flywheel energy storage charging control method according to claim 5, characterized in that: The second step: according to the parameter step-by-step identification method, the identification waveforms of the rotor flux, stator resistance and stator inductance can be obtained. First, the motor is started at a speed of 3000 rpm.

7. A flywheel energy storage charging control method according to claim 6, characterized in that: The second step: Based on the MRAS parameter identification accuracy verification, the dynamic quality of the three-phase permanent magnet synchronous motor mainly depends on the selected approach rate. When the exponential approach rate is selected, the control system can obtain good dynamic performance.

Images