A reliability evaluation method for inverter based on semi-physical simulation platform

Abstract

The present invention belongs to the technical field of electrical fault detection, and specifically relates to an inverter reliability evaluation method based on a semi-physical simulation platform. The present invention builds a semi-physical simulation platform for a photovoltaic grid-connected system based on the inverter topology, physical structure, and power device model. The working conditions of the inverter power devices are generated according to the actual working environment to evaluate the inverter reliability. The evaluation method proposed by the present invention is more accurate than the evaluation method using average value calculation, and can also provide a reference for the manufacturer's inverter redundancy design.

Description

Technical Field

[0001] The present invention belongs to the technical field of electrical fault detection, and in particular relates to an inverter reliability evaluation method based on a semi-physical simulation platform. Background Art

[0002] The inverter can convert direct current into alternating current. During operation, the main power device IGBT (insulated gate bipolar transistor) in the inverter has turn-on loss and turn-off loss, which will cause the temperature of the IGBT chip to change. Long-term use will cause the IGBT to age and fail.

[0003] The life of a photovoltaic power station, including construction and operation, is usually guaranteed to be 25 years. The relevant tests for photovoltaic inverters are designed and tested in a laboratory simulation environment. In actual applications, the reliability and life of power devices are greatly affected by operating conditions. Inverters with the same design parameters show different lifespans under different operating environments.

[0004] At present, most evaluation methods proposed by researchers are aimed at the mean model calculation and evaluation under long time scales. Their disadvantage is that they cannot accurately reflect the working state of power devices. Since temperature is the main factor affecting its reliability, the transient and drastic current and voltage changes of power tubes have a more obvious impact on reliability. Summary of the invention

[0005] The present invention is made to solve the above technical problems, and aims to provide an inverter reliability assessment method based on semi-physical simulation with higher precision and more accurate assessment.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The present invention proposes an inverter reliability evaluation method based on a semi-physical simulation platform. The evaluation method performs loss calculation according to real-time simulation data of the semi-physical simulation platform, and builds a corresponding semi-physical simulation platform according to the working environment, topology and device model of the inverter. The inverter is a power switch tube IGBT, a DC bus capacitor or a filter. The inverter works in a photovoltaic grid-connected system. The photovoltaic grid-connected system includes a photovoltaic component model, an inverter model and an AC power grid model that are sequentially connected to each other. The specific steps of the evaluation method are as follows:

[0008] (1) Generate a mission profile based on the inverter working environment to be evaluated;

[0009] (2) Build a photovoltaic grid-connected system model in Matlab / Simulink;

[0010] (3) Build a semi-physical simulation platform for photovoltaic grid-connected systems;

[0011] (4) Evaluate the inverter reliability based on the simulation results of the semi-physical simulation platform.

[0012] In the present invention, step (1) is specifically as follows: based on the statistical irradiance and temperature data in the historical meteorological data, the average values ​​of the past three years are taken to generate a one-year mission profile, that is, a one-year irradiance and temperature time series.

[0013] In the present invention, the photovoltaic module model, the inverter model and the AC power grid model are modeled in step (2), specifically as follows:

[0014] (2.1) Modeling the photovoltaic module model in the photovoltaic grid-connected system, wherein the photovoltaic module model is a current source model, whose function is to input environmental parameters and output corresponding DC current and voltage; the environmental condition input of the current source model is temperature and irradiance; the photovoltaic module parameter input is one or more of the open circuit voltage Voc, maximum power point voltage Vmpp, short circuit current Isc or maximum power point current Impp of the photovoltaic module, and the specific value is determined by the type and process of the photovoltaic module;

[0015] (2.2) Modeling the inverter model in the photovoltaic grid-connected system. The inverter model is determined by the actual topology of the inverter, and its function is to input direct current and output alternating current; the power switch tube model is a Matlab / Simulink ready-made library module, and its parameters are switch resistance, stray capacitance and stray impedance, which are determined according to the power switch tube parameters actually used in the inverter; the DC bus capacitor model is a Matlab / Simulink ready-made library module, and its parameters are capacitance value, which is determined according to the DC bus capacitor parameters actually used in the inverter; the filter model is set according to the DC bus capacitor parameters actually used in the inverter;

[0016] (2.3) Modeling the AC power grid model in the photovoltaic grid-connected system, wherein the AC power grid model functions as a three-phase voltage source; the three-phase voltage source is a three-phase independent voltage source, and the parameters are the root mean square value of the phase voltage, the initial phase angle and the voltage frequency, and the parameters are set according to the actual situation of the AC power grid where the inverter is located.

[0017] In the present invention, step (3) is specifically as follows:

[0018] The semi-physical simulation platform comprises a simulation machine, a control board, an adapter board and a host computer. The host computer is bidirectionally connected to the simulation machine and is used to control the simulation machine. The simulation machine runs a simulation model. The simulation model comprises a photovoltaic component model, an inverter model and an AC power grid model. The simulation machine is bidirectionally connected to the IO board, the IO board is bidirectionally connected to the adapter board, the adapter board is bidirectionally connected to the control board, the output sampling signal of the simulation machine is sequentially transmitted to the control board via the IO board and the adapter board, the control board is an inverter control board, and the output control signal of the control board is transmitted back to the simulation machine via the adapter board to control the simulation model. The task profile generated in step (1) is input into the semi-physical simulation platform to obtain the power tube current and voltage of the inverter under this task profile.

[0019] In the present invention, step (4) is specifically as follows:

[0020] (4.1): Calculate the inverter power device loss; According to the current and voltage of the power device obtained in step (3) under this task profile, calculate the power device switching loss and conduction loss: P IGBT =P cond，IGBT +P sw，IGBT , total loss P l =P sw，IGBT +P cond，IGBT +P sw，Diode +P cond，Diode , the calculation formula is as follows:

[0021] IGBT conduction loss:

[0022]

[0023] Among them, V CE is the conduction voltage drop, i C is the on-state current, τ is the duty cycle of the control signal;

[0024] IGBT switching loss:

[0025]

[0026] Among them, f sw is the switching frequency, E on (I nom , V nom ) and E off (I nom , V nom ) is the switching loss under rated conditions, is the ratio of current to rated current, is the ratio of DC side voltage to rated voltage;

[0027] Diode conduction loss:

[0028]

[0029] Among them, v F is the diode conduction voltage drop, i C is the on-state current, τ is the duty cycle of the control signal;

[0030] Diode switching losses:

[0031]

[0032] Among them, f sw is the switching frequency, E Diode(off) is the turn-off loss under rated conditions, is the ratio of current to rated current, is the ratio of DC side voltage to rated voltage;

[0033] (4.2): Calculate the junction temperature T of the inverter power device j ; Generate a thermal resistance network based on the inverter heat dissipation design, and calculate the junction temperature T based on the thermal resistance under this thermal resistance network j ;

[0034] T j =T air +P l ×(R h-a (t)+R c-h (t))+P igbt ×R c-j (i)

[0035] Among them, T air is the ambient temperature, P l is the total power loss in step S4, P igbt is the IGBT loss in step S4, R h-a is the equivalent thermal resistance from the heat sink to the environment, R c-h is the equivalent thermal resistance from the heat sink to the power module, R c-j is the equivalent thermal resistance from the power module housing to the IGBT junction temperature;

[0036] (4.3): Use rain flow counting to count power cycles; perform power cycle calculation based on the junction temperature of the power device calculated in step (4.2), and use the rain flow counting method to calculate the cycle count matrix N(ΔT) of the inverter under different average junction temperatures and junction temperature fluctuations based on the time series of the junction temperature. jm , ΔT jm );

[0037] (4.4): Evaluate the reliability change of power devices under the current mission profile; calculate the number of power device cycles N corresponding to each power cycle based on the life model of the inverter power device f (ΔT jm , ΔTjm ), the reciprocal is the loss D(ΔT jm , ΔT jm ), the losses are superimposed to obtain the inverter power device loss under the current mission profile, that is, the loss D of one year of use; when the loss accumulates to 1, it is considered to be a failure;

[0038]

[0039] Among them, A, α, β1, β0, C, γ, E a is a constant, related to the IGBT model, k b is the Boltzmann constant, T jm is the average junction temperature during the power cycle, ΔT jm is the difference between the maximum junction temperature and the minimum junction temperature of the power cycle, t on is the IGBT conduction time, f d is the switching frequency;

[0040] d=1 / N f

[0041] D = ∑d.

[0042] The beneficial effects of the present invention are:

[0043] 1) The evaluation method adopts an evaluation method based on semi-physical simulation, which can be closer to the actual working environment of the inverter and avoid using the average value model for loss calculation, which can improve the accuracy of the inverter reliability evaluation.

[0044] 2) It can provide reference for manufacturers’ inverter redundancy design and save costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a block diagram of the semi-physical simulation platform of the photovoltaic grid-connected system built by the present invention.

[0046] Figure 2 It is the historical meteorological data of the photovoltaic inverter installation environment for one year in the embodiment of the present invention, wherein: (a) is the temperature curve for one year, and (b) is the irradiance curve for one year.

[0047] Figure 3 The photovoltaic inverter in the embodiment of the present invention is Figure 1 Working conditions of power devices under working environment. Where: (a) is the voltage that the power device withstands, and (b) is the current flowing through the power device.

[0048] Figure 4 The photovoltaic inverter in the embodiment of the present invention is Figure 1 Power loss of power devices in working environment.

[0049] Figure 5 It is the thermal resistance network of the photovoltaic inverter power device in the example of the present invention.

[0050] Figure 6 The photovoltaic inverter in the embodiment of the present invention is Figure 1 Junction temperature of power devices under operating conditions.

[0051] Figure 7 The photovoltaic inverter in the embodiment of the present invention is Figure 1 The loss per power cycle of a power device under operating conditions. DETAILED DESCRIPTION

[0052] The present invention is further described below by way of embodiments in conjunction with the accompanying drawings.

[0053] Embodiment 1:

[0054] The purpose of the present invention is to propose a long-life power control method for an active photovoltaic inverter based on reliability evaluation, which can correct the maximum power limit of the inverter according to the working environment and the configuration of the photovoltaic power station, is easy to modify, and improves user benefits.

[0055] Taking a 6kW grid-connected inverter operating in Lhasa, Tibet as an example, the ratio of the rated power of the photovoltaic panel to the rated power of the inverter is n=1.25.

[0056] Step 1: Generate a mission profile based on the inverter working environment. Based on the irradiance and temperature data in the statistical historical meteorological data, take the average value of the past three years to generate a one-year mission profile, that is, a one-year irradiance and temperature time series, such as Figure 2 As shown in the figure, (a) is the temperature curve for one year, and (b) is the irradiance curve for one year as the environmental input for the temperature of the photovoltaic array and inverter in the simulation model. At the same time, the grid factor task profile is generated according to the working characteristics of the local power grid as the setting input of the AC power grid in the simulation model.

[0057] Step 2: Construct the photovoltaic array, inverter and AC power grid models respectively in Matlab / Simulink. The photovoltaic array model includes a photovoltaic cell model with input quantities of irradiance and temperature and output quantities of DC current and voltage; the inverter model includes a power switch tube, a DC bus capacitor, and a filter; and the AC power grid model is a three-phase voltage source including a transformer model.

[0058] (2.1) Model the photovoltaic modules in the photovoltaic grid-connected system. The photovoltaic module model is a current source model, which has the function of inputting environmental parameters and outputting corresponding DC current and voltage. The environmental condition inputs of the model are temperature and irradiance; the photovoltaic module parameter inputs are the open circuit voltage Voc, maximum power point voltage Vmpp, short circuit current Isc, and maximum power point current Impp of the photovoltaic module. The specific values ​​are determined by the module type and process. In the embodiment of the present invention, Voc = 38.68V, Vmpp = 31.35V, Isc = 8.54A, and Impp = 7.97A.

[0059] (2.2) Model the inverter in the photovoltaic grid-connected system. The inverter model is determined by the actual topology of the inverter, and its function is to input direct current and output alternating current. The model generally includes a power switch tube, a DC bus capacitor, and a filter. The power switch tube model is a Matlab / Simulink ready-made library module, and its parameters are switch resistance, stray capacitance, and stray impedance, which are determined according to the power switch tube parameters actually used in the inverter. The DC bus capacitor model is a Matlab / Simulink ready-made library module, and its parameters are capacitance values, which are determined according to the DC bus capacitor parameters actually used in the inverter. The filter model is set according to the DC bus capacitor parameters actually used in the inverter.

[0060] (2.3) Model the AC power grid in the photovoltaic grid-connected system, and the AC power grid model function is a three-phase voltage source. The model is a three-phase independent voltage source, and the parameters are the root mean square value of the phase voltage, the initial phase angle, and the voltage frequency. The parameters are set according to the actual situation of the power grid where the inverter is located. In the embodiment of the present invention, the root mean square value of the phase voltage is 220V, the initial phase angle is 0, and the voltage frequency is 50Hz.

[0061] Step 3: Build the corresponding semi-physical simulation platform of photovoltaic grid-connected system, including simulation machine, control board, interface board and host computer. The structural block diagram is as follows Figure 1 As shown. The simulator runs the simulation model, including the photovoltaic module model, the inverter model, and the AC power grid model. The output sampling signal is transmitted to the control board through the interface board. The control board is the inverter control board. The output control signal is transmitted back to the simulator through the interface board to control the simulation model. The task profile generated in step (1) is input into the semi-physical simulation platform to obtain the power tube current and voltage of the inverter under this task profile, as shown in Figure 3 shown.

[0062] Step 4: Evaluate the inverter reliability based on the simulation results of the semi-physical simulation platform.

[0063] (4.1): Calculate the inverter power device loss. According to the current and voltage of the power device under this task profile obtained in step (3), calculate the power device switching loss and conduction loss: P IGBT =Pcond，IGBT +P sw，IGBT , total loss P l =P sw，IGBT +P cond，IGBT +P sw，Diode +P cond，Diode , the calculation formula is as follows:

[0064] IGBT conduction loss:

[0065]

[0066] Among them, V CE is the conduction voltage drop, i C is the on-state current, and τ is the duty cycle of the control signal.

[0067] IGBT switching loss:

[0068]

[0069] Among them, f sw is the switching frequency, E on (I nom , V nom ) and E off (I nom , V nom ) is the switching loss under rated conditions, is the ratio of current to rated current, It is the ratio of DC side voltage to rated voltage.

[0070] Diode conduction loss:

[0071]

[0072] Among them, v F is the diode conduction voltage drop, i C is the on-state current, and τ is the duty cycle of the control signal.

[0073] Diode switching losses:

[0074]

[0075] Among them, f sw is the switching frequency, E Diode(off) is the turn-off loss under rated conditions, is the ratio of current to rated current, is the ratio of the DC side voltage to the rated voltage. The calculation result is as follows Figure 4 shown.

[0076] (4.2): Calculate the junction temperature T of the inverter power device j Generate a thermal resistance network based on the inverter heat dissipation design, such as Figure 5 The thermal resistance network model is a mathematical model. The parameters are the thermal resistance of the power device and the thermal resistance of the heat sink. It is determined according to the heat sink material, shape, cooling type and power device parameters actually used in the inverter. Under this thermal resistance network, the junction temperature T is calculated based on the thermal resistance. j .

[0077] T j =T air +P l ×(R h-a (t)+R c-h (t))+P igbt ×R c-j (t)

[0078] Among them, T air is the ambient temperature, P l is the total power loss in step S4, P igbt is the IGBT loss in step S4, R h-a is the equivalent thermal resistance from the heat sink to the environment, R c-h is the equivalent thermal resistance from the heat sink to the power module, R c-j is the equivalent thermal resistance from the power module housing to the IGBT junction temperature;

[0079] The calculation results are as follows Figure 6 shown.

[0080] (4.3): Use rain flow counting to count power cycles. Perform power cycle calculation based on the junction temperature of the power device calculated in step S5, and use the rain flow counting method to calculate the cycle count matrix N(ΔT) of the inverter under different average junction temperatures and junction temperature fluctuations according to the time series of the junction temperature. jm , ΔT jm ).

[0081] (4.4): Evaluate the reliability change of power devices under the current mission profile. Calculate the number of power device cycles N corresponding to each power cycle according to the life model of the inverter power device. f (ΔT jm , ΔT jm ), the reciprocal is the loss D(ΔT jm , ΔT jm ), the calculation results are as follows Figure 7 As shown. The losses are superimposed to obtain the inverter power device loss under the current mission profile, that is, the loss D of one year of use. When the loss accumulates to 1, it is considered to be failed. In this embodiment, the loss of one year of use is: 2.951×10 -4 .

[0082]

[0083] Among them, A, α, β1, β0, C, γ, E a is a constant, related to the IGBT model, k b is the Boltzmann constant, T jm is the average junction temperature during the power cycle, ΔT jm is the difference between the maximum junction temperature and the minimum junction temperature of the power cycle, t on is the IGBT conduction time, f d is the switching frequency.

[0084] d=1 / N f

[0085] D = ∑d.

Claims

1. An inverter reliability assessment method based on a semi-physical simulation platform, characterized in that: The evaluation method calculates the loss according to the real-time simulation data of the semi-physical simulation platform, and builds the corresponding semi-physical simulation platform according to the working environment, topology and device model of the inverter. The inverter is a power switch tube IGBT, a DC bus capacitor or a filter. The inverter works in a photovoltaic grid-connected system, and the photovoltaic grid-connected system includes a photovoltaic component model, an inverter model and an AC power grid model that are sequentially connected to each other. The specific steps of the evaluation method are as follows: (1) Generate a mission profile based on the inverter working environment to be evaluated; (2) Build a photovoltaic grid-connected system model in Matlab / Simulink; (3) Build a semi-physical simulation platform for photovoltaic grid-connected systems; (4) Evaluate the inverter reliability based on the simulation results of the semi-physical simulation platform; (4.1): Calculate the inverter power device loss; According to the current and voltage of the power device obtained in step (3) under this task profile, calculate the power device switching loss and conduction loss: P IGBT =P cond，IGBT +P sw，IGBT , total loss P l =P sw，IGBT +P cond,IGBT +P sw，Diode +P cond,Diode , the calculation formula is as follows: IGBT conduction loss: Among them, V CE is the conduction voltage drop, i C is the on-state current, τ is the duty cycle of the control signal; IGBT switching loss: Among them, f sw is the switching frequency, E on (I nom , V nom ) and E off (I nom , V nom ) is the switching loss under rated conditions, is the ratio of current to rated current, is the ratio of DC side voltage to rated voltage; Diode conduction loss: Among them, v F is the diode conduction voltage drop, i C is the on-state current, τ is the duty cycle of the control signal; Diode switching losses: Among them, f sw is the switching frequency, E Diode(off) is the turn-off loss under rated conditions, is the ratio of current to rated current, is the ratio of DC side voltage to rated voltage; (4.2): Calculate the junction temperature T of the inverter power device j ; Generate a thermal resistance network based on the inverter heat dissipation design, and calculate the junction temperature T based on the thermal resistance under this thermal resistance network j ; Among them, T air is the ambient temperature, P l is the total power loss in step S4, P igbt is the IGBT loss in step S4, R h-a is the equivalent thermal resistance from the heat sink to the environment, R c-h is the equivalent thermal resistance from the heat sink to the power module, R c-j is the equivalent thermal resistance from the power module housing to the IGBT junction temperature; (4.3): Use rain flow counting to count power cycles; perform power cycle calculation based on the junction temperature of the power device calculated in step (4.2), and use the rain flow counting method to calculate the cycle count matrix N(ΔT) of the inverter under different average junction temperatures and junction temperature fluctuations based on the time series of the junction temperature. jm , ΔT jm ); (4.4): Evaluate the reliability change of power devices under the current mission profile; calculate the number of power device cycles N corresponding to each power cycle based on the life model of the inverter power device f (ΔT jm , ΔT jm ), the reciprocal is the loss D(ΔT jm , ΔT jm ), the losses are superimposed to obtain the inverter power device loss under the current mission profile, that is, the loss D of one year of use; when the loss accumulates to 1, it is considered to be a failure; Among them, A, α, β1, β0, C, γ, E a is a constant, related to the IGBT model, k b is the Boltzmann constant, T jm is the average junction temperature during the power cycle, ΔT jm is the difference between the maximum junction temperature and the minimum junction temperature of the power cycle, t on is the IGBT conduction time, f d is the switching frequency; d=1 / N f D = ∑d.

2. The inverter reliability evaluation method of a semi-physical simulation platform according to claim 1 is characterized in that Step (1) is specifically as follows: based on the irradiance and temperature data in the statistical historical meteorological data, the average values ​​of the past three years are taken to generate a one-year mission profile, that is, a one-year irradiance and temperature time series.

3. The inverter reliability evaluation method of a semi-physical simulation platform according to claim 1 is characterized in that In step (2), the photovoltaic module model, inverter model and AC power grid model are modeled as follows: (2.1) Modeling the photovoltaic module model in the photovoltaic grid-connected system, wherein the photovoltaic module model is a current source model, whose function is to input environmental parameters and output corresponding DC current and voltage; the environmental condition input of the current source model is temperature and irradiance; the photovoltaic module parameter input is one or more of the open circuit voltage Voc, maximum power point voltage Vmpp, short circuit current Isc or maximum power point current Impp of the photovoltaic module, and the specific value is determined by the type and process of the photovoltaic module; (2.2) Modeling the inverter model in the photovoltaic grid-connected system. The inverter model is determined by the actual topology of the inverter, and its function is to input direct current and output alternating current; the power switch tube model is a Matlab / Simulink ready-made library module, and its parameters are switch resistance, stray capacitance and stray impedance, which are determined according to the power switch tube parameters actually used in the inverter; the DC bus capacitor model is a Matlab / Simulink ready-made library module, and its parameters are capacitance value, which is determined according to the DC bus capacitor parameters actually used in the inverter; the filter model is set according to the DC bus capacitor parameters actually used in the inverter; (2.3) Modeling the AC power grid model in the photovoltaic grid-connected system, wherein the AC power grid model functions as a three-phase voltage source; the three-phase voltage source is a three-phase independent voltage source, and the parameters are the root mean square value of the phase voltage, the initial phase angle and the voltage frequency, and the parameters are set according to the actual situation of the AC power grid where the inverter is located.

4. The inverter reliability evaluation method of a semi-physical simulation platform according to claim 1 is characterized in that Step (3) is as follows: The semi-physical simulation platform comprises a simulation machine, a control board, an adapter board and a host computer. The host computer is bidirectionally connected to the simulation machine and is used to control the simulation machine. The simulation machine runs a simulation model. The simulation model comprises a photovoltaic component model, an inverter model and an AC power grid model. The simulation machine is bidirectionally connected to the IO board, the IO board is bidirectionally connected to the adapter board, the adapter board is bidirectionally connected to the control board, the output sampling signal of the simulation machine is sequentially transmitted to the control board via the IO board and the adapter board, the control board is an inverter control board, and the output control signal of the control board is transmitted back to the simulation machine via the adapter board to control the simulation model. The task profile generated in step (1) is input into the semi-physical simulation platform to obtain the power tube current and voltage of the inverter under this task profile.

Images