Deterioration state estimation device for power storage system, deterioration state estimation method, and operation optimization method for power storage system

Abstract

To estimate a deterioration state of an entire power storage system including a secondary battery and an orthogonal transformer.SOLUTION: A deterioration state estimation device estimates a deterioration state of a power storage system including a secondary battery and an orthogonal transformer. The deterioration state estimation device comprises an estimation arithmetic processing section which estimates an open-circuit voltage and internal resistance with respect to a charging rate of the secondary battery based on a charge / discharge voltage and a charge / discharge current of the secondary battery and output / input power of the orthogonal transformer, estimates efficiency of the orthogonal transformer and auxiliary unit power consumption of the orthogonal transformer and estimates charge / discharge energy efficiency with respect to the charging rate of the secondary battery and a rating output / input ratio at an AC side of the orthogonal transformer based on the open-circuit voltage, the internal resistance, the efficiency of the orthogonal transformer and the auxiliary unit power consumption of the orthogonal transformer estimated.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a degradation state estimating device and a degradation state estimating method for a power storage system that estimates the degradation state of a power storage system including a secondary battery and an orthogonal converter, and a method for optimizing operation of the power storage system. [Background technology]

[0002] In recent years, battery energy storage systems (BESS) have been expected to be used as temporary power storage devices for time shifting and countering output fluctuations of variable renewable energy (VRE). Lithium-ion batteries (LIBs), which have excellent energy density and power density, have been attracting attention as storage batteries, i.e., secondary batteries.

[0003] LIBs have issues in terms of cost and resources (cobalt, nickel, lithium, copper, etc.), and to solve these issues, efforts are being made to reuse LIBs that have reached the end of their lifespan as in-vehicle LIBs, which require high performance, by repurposing them as stationary BESSs, which are operated relatively gently. However, LIBs are subject to a phenomenon called degradation, in which their performance decreases over time, and when they are reused, it is necessary to classify them based on their intended use and remaining performance at the start of the reuse, and to perform degradation diagnosis to understand their condition during operation. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2016-145795 A Summary of the Invention [Problem to be solved by the invention]

[0005] For example, Patent Document 1 discloses an energy storage system that includes a lithium-ion battery (LIB) and a power conditioning system (PCS) that has an orthogonal conversion function and a step-up / step-down function. For the LIB, a method for estimating the degradation state by estimating the energy efficiency has been proposed.

[0006] However, not only Patent Document 1 but also other conventional techniques do not show a method for estimating the degradation state of the entire power storage system including an orthogonal converter such as a PCS. Therefore, conventional techniques cannot perform the above-mentioned degradation diagnosis with sufficient accuracy for practical use.

[0007] Therefore, an object of the present invention is to provide a degradation state estimation device for a power storage system, a degradation state estimation method, and a storage system operation optimization method that make it possible to estimate the degradation state of the entire power storage system including a secondary battery and an orthogonal converter. [Means for solving the problem]

[0008] In order to achieve the above-mentioned object, a degradation state estimation device according to an embodiment of the present invention is a degradation state estimation device that estimates the degradation state of a power storage system including a secondary battery and an orthogonal converter, and is equipped with an estimation calculation processing unit that estimates the open circuit voltage and internal resistance relative to the charging rate of the secondary battery based on the charging / discharging voltage and charging / discharging current of the secondary battery and the input / output power of the orthogonal converter, and estimates the efficiency of the orthogonal converter and the auxiliary power consumption of the orthogonal converter, and estimates the charging / discharging energy efficiency relative to the charging rate of the secondary battery and the rated input / output ratio on the AC side of the orthogonal converter based on the estimated open circuit voltage, internal resistance, efficiency of the orthogonal converter, and auxiliary power consumption of the orthogonal converter.

[0009] Furthermore, a degradation state estimation method according to an embodiment of the present invention is a degradation state estimation method for a power storage system including a secondary battery and an orthogonal converter, and includes performing an estimation calculation process to estimate an open circuit voltage for a charging rate of the secondary battery, estimating an internal resistance, estimating the efficiency of the orthogonal converter, estimating the auxiliary power consumption of the orthogonal converter, and estimating a degradation state of the power storage system by estimating a charging / discharging energy efficiency for the charging rate of the secondary battery and a rated input / output ratio of the orthogonal converter based on the estimated open circuit voltage curve, internal resistance curve, efficiency curve of the orthogonal converter, and a history of the auxiliary power consumption of the orthogonal converter. Effect of the Invention

[0010] According to the present invention, it is possible to estimate the degradation state of the entire power storage system including the secondary battery and the orthogonal converter. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a power storage system according to an embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of the configuration of the degradation state estimating device 6. As shown in FIG. [Diagram 3] FIG. 3 is a diagram showing an example of the specifications of the LIB module 3 in FIG. [Figure 4] FIG. 4 is a diagram conceptually showing an approximation curve for correcting the open circuit voltage. [Diagram 5] FIG. 5 is a diagram conceptually showing an equivalent circuit of a lithium ion battery. [Figure 6] FIG. 6 is a diagram showing the estimation error of the analysis result obtained using the Kalman filter. [Figure 7] FIG. 7 is a diagram illustrating an example of a Gaussian function. [Figure 8] FIG. 8 is a conceptual diagram showing an example of a model used for estimating the efficiency curve of a PCS. [Figure 9] FIG. 9 is a graph showing a simulated operation pattern of power input / output to / from the PCS AC side of the BESS used in the verification. [Figure 10] FIG. 10 is a graph showing a histogram of power input and output on the PCS AC side. [Figure 11] FIG. 11 is a graph showing the results of the change in the profile of the open circuit voltage and internal resistance learned and estimated by BT-MGFFD over a given period of time. [Figure 12] FIG. 12 is a graph showing the change in the estimation error of the charge / discharge voltage over time. [Figure 13]FIG. 13 is a graph showing the results of determining trends in the PCS efficiency curve and the estimated value of PCS auxiliary power consumption estimated by learning estimation using PCS-MGFFD from the operational time-series data of the LIB module and PCS. [Figure 14] FIG. 14 is a graph showing the results of estimating the charge / discharge power on the DC side of the LIB module from the input / output power on the AC side of the PCS. [Figure 15] FIG. 15 is a graph showing the estimated RTE of the entire BESS versus the charging rate of the LIB and the rated input / output ratio on the AC side of the PCS. [Figure 16] FIG. 16 is a flowchart showing an example of the operation of the degradation state estimating device 6 of the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments will be described with reference to the drawings.

[0013] Before describing the embodiment in detail, the significance of estimating the degradation state of the entire power storage system including the secondary battery and the orthogonal converter will be described.

[0014] Lithium-ion batteries (LIBs), a typical secondary battery, have a tendency to lose charge / discharge energy efficiency due to degradation. This is mainly caused by an increase in Joule heat due to an increase in internal resistance (electrical resistance and polarization resistance), which leads to an increase in charge / discharge energy loss. Here, the charge / discharge energy loss is defined as E lib,loss Let I be the charge / discharge current of the LIB and R be the internal resistance. Then, E lib,loss can be expressed by the following equation (1).

[0015]

number

[0016] R is a LIB parameter that changes with deterioration and cannot be intentionally changed in operation, but I is a parameter related to charge / discharge operation control and can be adjusted. Therefore, by keeping I low, Elib,loss As a result, it is possible to improve the charge / discharge energy efficiency.

[0017] On the other hand, BESS includes not only LIB but also an orthogonal converter like a power conditioning system (PCS), through which the DC LIB and AC transmission and distribution system (power system) are connected. Therefore, the final charge / discharge energy efficiency (RTE: Round Trip Efficiency) of the BESS in the temporary storage of VRE mentioned at the beginning is the product of the efficiency of the LIB and the PCS (hereinafter, the parts in the BESS that have the orthogonal conversion and step-up / step-down functions are collectively referred to as PCS). The energy efficiency of the BESS is often treated as a single fixed value in simulations, but in reality, the energy efficiency changes depending on the ratio of input / output power to the rated value (hereinafter, referred to as the "rated input / output ratio"). In addition, the RTE of the BESS is generally treated as a single value, but as described above, the energy efficiency of both the LIB and the PCS changes depending on the input / output, so it is easy to imagine that the RTE of the entire BESS fluctuates during operation. Above all, the sensitivity of LIB and PCS to the energy efficiency relative to the rated input / output ratio is positive and negative in the low input / output range, i.e., they are in a trade-off relationship; therefore, understanding the energy efficiency of each will lead to improvements in the RTE and operational economy of the BESS.

[0018] Therefore, the inventors verified a method for estimating the energy efficiency characteristics of the LIB and the PCS during the operation of the BESS and for comprehensively diagnosing the RTE of the BESS. In this embodiment, the method for comprehensively diagnosing the RTE of the BESS is shown together with the verification results.

[0019] <Configuration of Energy Storage System (BESS)> FIG. 1 is a diagram illustrating an example of the configuration of a power storage system according to an embodiment.

[0020] The power storage system 1 shown in Fig. 1 corresponds to the above-mentioned BESS. This power storage system 1 mainly includes a power conditioning system 2, a LIB module 3, a battery management unit (BMU: Battery Management Unit) 4, an energy management unit (EMU: Energy Management Unit or EMS: Energy Management system) 5, a degradation state estimation device 6, and a battery temperature measurement unit 7.

[0021] The LIB module 3 corresponds to the above-mentioned module having the LIB. The power conditioner 2 corresponds to the above-mentioned PCS. The degradation state estimation device 6 realizes a function of estimating the degradation state of the entire BESS including the LIB and the PCS, and further a function of estimating the charge / discharge energy efficiency and the operational economy of the BESS based on the result of the degradation state estimation.

[0022] The function of the degradation state estimation device 6 can be realized as a program executed by one or more computers. Moreover, the degradation state estimation device 6 does not necessarily have to be provided in the power storage system 1. The degradation state estimation device 6 may be provided outside the power storage system 1. When the degradation state estimation device 6 is provided outside the power storage system 1, the degradation state estimation device 6 may be provided at any place on a network (cloud) capable of communicating with the power storage system 1. In this case, the power storage system 1 side is provided with a function of transmitting various information required for estimating the degradation state of the BESS to the degradation state estimation device 6 side. Moreover, the degradation state estimation device 6 side is provided with a function of receiving the above-mentioned various information transmitted from the power storage system 1 side. In this embodiment, a case where the degradation state estimation device 6 is provided in the power storage system 1 will be described.

[0023] The deterioration state estimating device 6 will be described in detail later.

[0024] The power conditioner 2 is an orthogonal converter that performs power conversion between DC and AC. The power conditioner 2 converts, for example, power supplied from a power system 9 such as an external power company, power supplied from a solar power generation system, or power supplied from the LIB module 3 so that it can be used by an electrically driven device including a specific load 8. Furthermore, the power conditioner 2 may have a function of a charger that charges a storage battery. Furthermore, when the specific load 8 is an electrical device driven by AC power, the power conditioner 2 converts the DC power supplied from the LIB module 3 into an AC power form. Furthermore, depending on the electrical device of the specific load 8, the power conditioner 2 may increase the voltage value of the power. Furthermore, the power conditioner 2 can not only supply power to the specific load 8, but also reduce the consumption of the power supplied from the power system 9 by releasing the energy stored in the LIB module 3 when the consumption of the power supplied from the power system 9 is at its maximum. In this case, the LIB module 3 after discharge can be charged to full charge through the power conditioner 2 when the demand for power decreases, such as late at night.

[0025] The specific loads 8, which are the recipients of power supply from the energy storage system 1, are assumed to be devices to which power should be supplied when the power supply from the power grid 9 is stopped (e.g., during a power outage), and include, for example, electronic devices such as computers and communication devices, to which power is supplied for power backup.

[0026] The LIB module 3 includes a secondary battery (storage battery) 11 that outputs a direct current voltage, a cell monitor unit (CMU: Cell Monitor Unit) 12, and a protection unit 13. The number of LIB modules 3 is appropriately set according to the design of the power supply amount of a specific load, etc., and when forming a large-capacity secondary battery, a plurality of LIB modules 3 may be electrically connected to configure one assembly, that is, one battery pack. In addition, in this embodiment, a lithium-ion storage battery is used as an example of a secondary battery, but the present invention is not limited to this. As with the lithium-ion storage battery, if the battery has a small memory effect and good self-discharge characteristics, it can be easily applied to batteries of different structures. For example, it can be applied to nanowire batteries improved from lithium-ion storage batteries.

[0027] The secondary battery 11 of this embodiment is not limited in terms of the battery internal materials (electrode materials, etc.) or cell structure, and the form of the exterior material may be a cylindrical can type, a square can type, a laminate type, etc. The connection form of the secondary batteries 11 constituting the LIB module 3 may be any known connection form such as a single cell, a series assembled battery, or a parallel assembled battery.

[0028] The battery temperature measuring unit 7 measures the temperature using a temperature sensor (not shown) arranged in contact with each secondary battery 11. The ambient temperature in which the lithium ion battery can be used in the device is in the range of about 5 to 40°C, but depending on the installation environment (cold region or tropical region), if necessary, a battery temperature adjustment mechanism can be installed in the device. This battery temperature adjustment mechanism is composed of a fan and a heater for adjusting the temperature within the range in which the secondary battery 11 can be used (about 5 to 40°C) so that the battery performance does not deteriorate when the upper or lower limit of a preset temperature range is exceeded based on the temperature measured by the battery temperature measuring unit 7. Of course, if the temperature range in which the secondary battery 11 can be used is expanded due to subsequent battery improvements, the battery temperature adjustment mechanism can accommodate all of those temperature ranges.

[0029] The cell monitoring unit 12 has a function of continuously measuring the output voltage, current, and temperature of each secondary battery 11, which is a single battery (or a single cell), and transmitting the measurement results to the battery management unit 4. This cell monitoring unit 12 measures the charge / discharge voltage at regular intervals during the charge / discharge process for estimating the charge / discharge curve, under the control of the calculation control unit 14, which will be described later.

[0030] Furthermore, the cell monitoring unit 12 has a function of transmitting the output voltage, current, and temperature acquired from the secondary battery 11 to the battery management unit 4 as monitor information.

[0031] The battery management unit 4 has a function of determining the occurrence of an abnormality such as overcharging, over-discharging, and temperature rise based on the received monitor information, and controlling the protection unit 13 to stop charging or output (discharging) to the secondary battery 11 to prevent overcharging and over-discharging. In addition, when an emergency abnormality occurs due to a failure of the secondary battery 11, the protection unit 13 stops charging or output (discharging) to the secondary battery 11 by electrically cutting off. The protection unit 13 may have a function of avoiding danger by notifying the battery management unit 4 of an abnormality. The abnormality occurrence determination function may be installed in either the cell monitoring unit 12 on the LIB module 3 side or the battery management unit 4 on the power storage system 1 side, but it may also be installed in each to improve safety by double determination. In the double determination, the order of determination is determined in advance, for example, the cell monitoring unit 12 first determines the occurrence of an abnormality, and then the battery management unit 4 determines the occurrence of an abnormality for the second time. As a determination procedure at this time, normally, when either one of the two determination units determines an abnormality, a protection operation by the protection unit 13 is executed. Depending on the design concept, it is also possible to configure the protection unit 13 to execute a protective operation only when both are judged to be abnormal, and to issue a warning if only one is abnormal.

[0032] Furthermore, the battery management unit 4 has a function of centrally aggregating monitor information transmitted from the cell monitoring units 12 of each LIB module 3 and transmitting it to a higher-level energy management unit 5. Based on this monitor information, the energy management unit 5 instructs the power conditioner 2 to charge and discharge the LIB module 3. The power conditioner 2 controls the charging and discharging of the LIB module 3 according to the instruction.

[0033] The energy management unit 5 is composed of an arithmetic and control unit 14, a display unit 15, a server 16, and an interface unit 17.

[0034] The arithmetic control unit 14 has functions equivalent to those of a computer arithmetic processing unit, etc., and instructs the battery management unit 4 or the power conditioner 2 to charge and discharge the LIB module 3. In addition, an upper limit charging voltage value and a lower limit discharging voltage value are set in advance for each LIB module 3, and based on monitor information transmitted from the battery management unit 4, the arithmetic control unit 14 instructs the battery management unit 4 or the power conditioner 2 to stop charging or discharging.

[0035] The display unit 15 is configured, for example, by a liquid crystal display unit, and displays the operating status of the power storage system 1, the remaining capacity of the LIB module 3 (secondary battery 11), and warnings under the control of the arithmetic control unit 14. The display unit 15 may also be configured as a touch panel or the like and used as an input device.

[0036] The server 16 stores the latest information, such as the operating status of the power storage system 1, monitor information on the LIB module 3, etc., and information on the charge and discharge curve, transmitted to the energy management unit 5, in a timely manner. The interface unit 17 communicates with a centralized management system composed of an externally installed server, etc., via a network 18 such as the Internet.

[0037] <Configuration of Deterioration State Estimation Device 6> Fig. 2 is a diagram showing an example of the configuration of the degradation state estimating device 6. Fig. 3 is a diagram showing an example of the specifications of the LIB module 3 in Fig. 2. However, the configuration of the degradation state estimating device 6 and the specifications of the LIB module 3 are not limited to the examples shown in Figs. 2 and 3.

[0038] The method for comprehensively diagnosing the RTE of the BESS described at the beginning is applied to the degradation state estimation device 6 (particularly the estimation calculation processing unit 27) shown in Fig. 2. The details will be described later.

[0039] The degradation state estimation device 6 shown in FIG. 2 includes a measurement unit 21, a charging power supply unit 22, a discharge unit 23, a discharge load unit 24, a measurement unit 25, a time measurement unit 26, an estimation calculation processing unit 27, a SOC calculation unit 28, a function derivation unit 29, and a state estimation unit 30.

[0040] The measuring unit 21 measures the voltage and current on the direct current (DC) side and the voltage and current on the alternating current (AC) side of the power conditioner 2. The measurement results are transmitted to the estimation calculation processing unit 27.

[0041] The charging power supply unit 22 detects the charging state of the secondary battery 11, and outputs a DC current and voltage that matches the rating to the secondary battery 11 within the range up to the upper charging voltage limit of the secondary battery 11, to charge the secondary battery 11. This charging power supply unit 22 is provided as a dedicated power supply for estimating the charge / discharge curve of the secondary battery 11, but a battery charging power supply unit provided in the power storage system or in the power conditioner 2 may be used instead. In this embodiment, the measurement unit 25 and the charging power supply unit 22 constitute the charging unit.

[0042] The discharge unit 23 includes a discharge load unit 24, and is electrically connected to the secondary battery 11 by operating a switch (not shown), causing a predetermined amount of power (assumed to be a constant current or constant voltage here) to be discharged from the secondary battery 11. This discharge load unit 24 may be a resistor or an electronic load, but it may also be possible to simulate a load and regenerate power to the power system without providing such a dedicated load.

[0043] The measurement unit 25 measures the DC voltage and DC current output by the LIB module 3 (secondary battery 11). The measurement is performed periodically at every predetermined time interval, measuring the DC voltage and DC current output from the LIB module 3. Note that, instead of actually measuring the voltage and current, it is also possible to use the voltage and current values ​​contained in the latest monitor information transmitted from the battery management unit 4 and stored in the server 16 of the energy management unit 5.

[0044] The time measurement unit 26 is a timer for measuring the time during which power is charged and discharged from the LIB module 3, and measures the measurement timing.

[0045] The SOC calculation unit 28 calculates the state of charge (SOC) at the timing when the measurement unit 25 measures the voltage. The SOC value is a value obtained by dividing the amount of charging current Qc [Ah] by the full charge capacity FCC [Ah] at that time. The SOC value of the state of charge in this embodiment can be calculated using a Kalman filter described later. The full charge capacity FCC value can be calculated by an appropriate method such as a charge / discharge voltage difference method, an AC impedance method, a discharge curve differentiation method, or an optimization filter. For example, it may be calculated by dividing the difference between the accumulated charge amount at two points in time by the difference between the SOC estimated by the Kalman filter.

[0046] The function derivation unit 29 derives approximate curves of the open circuit voltage function and impedance function to be corrected, and sets them as pre-estimated functions. In determining the approximate curves, various curve fitting methods can be applied, and the approximate curves determined in advance can be stored and set as pre-estimated functions. In addition, the initial open circuit voltage function and initial pre-estimated function can also be derived.

[0047] The estimation calculation processing unit 27 is a calculation processing unit (CPU or the like) that stores a calculation algorithm using a relational expression described later, and estimates a charge / discharge curve based on machine learning and estimated state parameters of the secondary battery 11, an efficiency curve of the secondary battery 11, and an efficiency curve of the power conditioner 2. This estimation calculation processing unit 27 does not need to be provided exclusively in the degradation state estimation device 6, and it is also possible to have the calculation control unit 14 of the energy management unit 5 substitute for the processing function.

[0048] In particular, the estimation calculation processing unit 27 has a function of estimating the open circuit voltage and internal resistance for the charging rate of the secondary battery 11 based on the charging / discharging voltage and charging / discharging current of the secondary battery 11 in each LIB module 3 and the input / output power of the AC converter, estimating the efficiency of the power conditioner 2 and the auxiliary power consumption of the power conditioner 2, and estimating the charging / discharging energy efficiency (RTE) for the charging rate of the secondary battery 11 and the rated input / output ratio on the AC side of the power conditioner 2 based on the estimated open circuit voltage, internal resistance, efficiency of the AC converter, and auxiliary power consumption of the AC converter, thereby estimating the degradation state of the BESS. Furthermore, the estimation calculation processing unit 27 has a function of estimating the charging / discharging energy efficiency and operational economy of the BESS based on the result of the degradation state estimation performed.

[0049] The state estimation unit 30 can estimate state parameters including the charging rate, open circuit voltage, and internal resistance (internal impedance) of the secondary battery 11 using a Kalman filter based on an electrochemical model, which in this embodiment is an extended Kalman filter that supports nonlinearity.

[0050] The technology described in this embodiment can be applied to online estimation using network communication between an external device 19 and a system server 20 that manages multiple other power storage systems 1 through a network communication network 18 such as the Internet. Note that, in Fig. 1, one power storage system 1 is representatively shown for ease of understanding, but the number of power storage systems 1 is not limited to one.

[0051] (Overview of estimation using the Kalman filter) The inventors have studied the estimation of charge / discharge curves used for deterioration diagnosis in charge / discharge cycles of a lithium-ion battery using a Kalman filter and a Gaussian function. Fig. 4 is a conceptual diagram showing an approximation curve corrected for an open circuit voltage. Fig. 5 is a conceptual diagram showing an equivalent circuit of a lithium-ion battery. Fig. 7 is a diagram showing an example of a Gaussian function.

[0052] Here, as an example, a case where charging and discharging are performed in multiple LIB modules connected in series and time series data is obtained is shown. Note that, when performing deterioration diagnosis or state estimation, cycle testing using a single cell of a small lithium-ion storage battery is generally adopted, but here, a LIB module consisting of a large 50 Ah single cell was regarded as one battery pack (assembly) and performed. In addition, during the charge and discharge cycle test, a constant power mode was adopted to simulate the charging and discharging state by the power conditioner 2. Room temperature is desirable as the ambient temperature for the cycle test, but since unavoidable resistance heat occurs during charging and discharging, a range of approximately 20°C to 30°C was set as the module temperature.

[0053] In addition, the Modified Gaussian Free-form Deformation (MGFFD) method was applied to estimate the open circuit voltage and internal impedance. MGFFD is known as an effective method for estimating the open circuit voltage curve versus the charge rate of a LIB. In MGFFD, a Gaussian function is directly added to the open circuit voltage curve versus the charge rate of the LIB, gradually deforming the open circuit voltage curve in a direction perpendicular to the charge rate axis, gradually reducing the error.

[0054] In Fig. 4, D prior estimate represents the prior estimate of the open circuit voltage by graph learning. The deformation amount M represents a Gaussian function. D posterior estimate represents the posterior estimate of the open circuit voltage estimated by the Kalman filter. The accurate estimate is the LIB charging rate S at the time of correction. oc,0 This represents the open circuit voltage estimated from the time series data at

[0055] As shown in Figure 4, the open circuit voltage and internal impedance of a lithium-ion battery are transformed by adding a Gaussian function with adjusted height. That is, the error (estimation error) between the pre-estimated value (scalar) at the current point of the charging rate SOC and the posterior estimated value (scalar) estimated this time by the Kalman filter is multiplied by the learning rate L, and this is used as the height of the Gaussian function, which is added to the pre-estimated value (vector) of graph learning. The graph function transformed in this way is the posterior estimated value (vector) of graph learning.

[0056] Next, the error index I maer Define as the mean absolute error rate.

[0057]

number

[0058] Here, the combination conditions of the learning rate L and the rough range of the correction width σ affect the accuracy of the charge / discharge energy estimation, so we analyzed the relationship between these conditions and the estimation error index. In this analysis, the actual initial V oc (S oc ) to the C k The results are shown in Figure 6. Figure 6 shows the combination of the learning rate L and the rough correction width σ and the error index I maer Each dot is the best-precision point over the entire learning rate L.

[0059] From the results shown in Figure 6, the estimation process is -2 In most cases, the learning rate diverges, and it is found that the convergence approaches when the correction width σ is reduced. This is because a large value of the learning rate L and the correction width σ reduces the E ~ In addition, the trend of the optimal correction width σ is different when the learning rate L is L=10 -3.8 and L=10 -3.2 It was found that there was a boundary change.

[0060] <Method for comprehensively diagnosing the RTE of BESS> As described at the beginning, the inventor estimated the energy efficiency characteristics of each of the LIB and PCS during the operation of the BESS and verified a method for comprehensively diagnosing the RTE of the BESS. The following shows the specific method and the verification results. The method shown here is applied to the function of the estimation calculation processing unit 27 in the deterioration state estimation device 6.

[0061] In this verification, a BESS was constructed using an actual PCS and LIB modules and subjected to experiments. The BESS installation area was managed with an environmental temperature of 23 ± 5°C, and the voltage, current, and charge rate of the LIB module (DC side) were measured by a BMS (Battery Management System) attached to the module. The voltage and current on the AC side of the PCS were measured by a clamp-type power logger.

[0062] (Estimation of the efficiency curve of LIB) Estimation of the charge and discharge energy efficiency of the LIB module requires estimation of the open circuit voltage, internal resistance, and full charge capacity. Let the open circuit voltage be V oc , the internal resistance be R, and the full charge capacity be C fc . Among these, the learning-based estimation of V oc and R was performed using the aforementioned MGFFD method. Specifically, the estimation was performed by learning the characteristic profiles of V oc and R with respect to the charge rate S oc . To distinguish it from the MGFFD for PCS described later, here the MGFFD for the LIB module is denoted as BT-MGFFD.

[0063] In BT-MGFFD, the estimated charge and discharge voltage value V^ is calculated from the estimated open circuit voltage V^ and the estimated internal resistance R^ by the following formula (3), and the estimation error V oc with respect to the measured value V of the charge and discharge voltage c,d is obtained. And this estimation error is the estimation error V of the open circuit voltage and the internal resistance ~ c,d = V c,d - V^ c,d is obtained. And this estimation error is the estimation error of the open circuit voltage and the internal resistance V~ oc ,R ~ V ~ c,d From V ~ oc and R ~ The charge / discharge current I is defined as positive during charging and negative during discharging.

[0064]

number

[0065] V oc and R is the discrete value vector of the charging rate S oc = [0.01, 0.02, …, 0.99] as vectors corresponding to the vectors D^ - , the posterior estimate is D^, and the correction amount at that time is M=[M1, M2,… M 99 M is the charging rate S of the LIB to be estimated at the time of correction. oc,0 The prior estimation error V ~ oc - (S oc,0 ) or R ~- (S oc,0 ) with learning rate L lib Variance σ with height multiplied by lib 2 In other words, we define the target estimation error as D ~- (S oc,0 ), this can be expressed as follows:

[0066]

number

[0067] This correction, i.e., V oc In this case, the learning of BT-MGFFD progresses by sequentially and continuously performing the process described in FIG. 4 using the charge / discharge time-series data.

[0068] Similarly, for R, V~ c,d Based on the relationship in equation (3), BT-MGFFD was performed to learn the profile. Here, the learning parameters of BT-MGFFD are L lib = σ lib = 0.01. Also, at the start of learning, the discrete value vector V oc Since the initial values ​​of V and R must be given, the correct V oc and R are used as references, and V oc An offset is added, and R is set to a discrete vector that is linear and deviates from the correct value, resulting in an initial value that includes error.

[0069] Also, here, C fc The rated capacity of 40 Ah indicated by the LIB module manufacturer was used as a fixed value. fc If you need to estimate, for example, the cumulative charge / discharge electricity quantity ΔQ c,d and the difference in charging rate ΔS oc From the above, C is calculated by the following formula (5): fc may be calculated.

[0070]

number

[0071] (Estimation of PCS efficiency curve) The estimation of the efficiency curve of the PCS was performed based on the model shown in Figure 8, taking into account the power consumption of the PCS's auxiliary devices.

[0072] FIG. 8 is a conceptual diagram showing an example of a model used for estimating the efficiency curve of a PCS.

[0073] FIG. 8 shows the relationship between the LIB module 3, the power conditioner 2, and the power grid 9. The LIB corresponds to the LIB module 3, and the PCS corresponds to the power conditioner 2. The amount of charging power of the LIB is E lib,c The discharge power of the LIB is expressed as E lib,dThe input power of the PCS (AC side) is expressed as E pcs,in The output power of the PCS (AC side) is expressed as E pcs,out It is expressed as:

[0074] The PCS efficiency curve estimation is pcs Let R be the discrete value vector of the rated input / output ratio. typ = [0, 0.01, …, 1] corresponding to vector η pcs (R typ ) and was performed in the same manner as in the BT-MGFFD method described above. Here, to distinguish it from the BT-MGFFD described above, MGFFD for PCS is denoted as PCS-MGFFD.

[0075] In PCS-MGFFD, the power consumption of the PCS auxiliary equipment E au,l Taking this into consideration, the model is divided into charging and discharging, and the estimation error E ~ lib,c =E lib,c -E^ lib,c Or the estimated error E of the PCS output power ~ pcs,out =E pcs,out -E^ pcs,out is the estimated error η ~ pcs and E ~ au,l In other words, in the case of charging, the input power of the PCS is assumed to be generated by E pcs,in , the amount of charging energy of the LIB is E lib,c Focusing on the amount of charging energy in the LIB, it is assumed that the relationship in the following equation (6) holds.

[0076]

number

[0077] In addition, in the case of discharging, the output power of the PCS is E pcs,out , the discharge energy of the LIB is E lib,d Focusing on the amount of output power of the PCS, it is assumed that the relationship in the following equation (7) holds.

[0078]

number

[0079] Based on equations (6) and (7), the rated input / output ratio R of the AC side of the PCS to be estimated at the time of correction is calculated. typ,0 , that is, the rated power of the PCS is P pcs When this is set, R typ,0 =P pcs,in / P pcs , R during discharge typ,0 =P pcs,out / P pcs R pcs The learning parameter of PCS-MGFFD is the learning rate L pcs and variance σ pcs 2 Then, L pcs = 0.01, σ pcs = 0.05, L in equation (4) lib and σ lib It is treated the same as η pcs (R typ ) and E au,l As an example, the initial value of [W] is given as follows, which is different from the true value of the actual PCS characteristics.

[0080]

number

[0081] (BESS simulated operation pattern) The graph in Figure 9 shows the simulated operation pattern (325,170 seconds in total) of the power input and output on the PCS AC side of the BESS used in this verification. The horizontal axis of the graph indicates the elapsed time [×10 s], and the vertical axis indicates the power value [W]. The power value on the vertical axis indicates charging as positive and discharging as negative, and the time series data was sampled every 10 s. The BESS was operated within the range of "0.1≦S ocThe test was performed in the range of "≦0.9". The histogram of the power input and output on the PCS AC side is shown in the graph in Figure 10. The horizontal axis of the graph shows the power value [W], and the vertical axis shows the number of occurrences.

[0082] <Results and Discussion> (Estimation of LIB open circuit voltage and internal resistance) The open circuit voltage V V was estimated by learning using the LIB operation time series data (voltage and current) based on the simulated operation pattern shown in Fig. 9. oc -Profile of internal resistance R, elapsed time (4×10 4 The graphs in Fig. 11(a) and (b) show the change in the charging time per second. The horizontal axis of the graph in Fig. 11(a) is the charging time. oc The vertical axis indicates the open circuit voltage V oc The horizontal axis of the graph in Fig. 11(b) is the charge S oc The vertical axis indicates the internal resistance R [Ω]. The different types of lines in each graph indicate different elapsed times.

[0083] Figures 11(a) and (b) show that as time passes, the learning by MGFFD converges toward the end of the simulated operation, and the profile hardly changes at all.

[0084] (Estimation of charge / discharge voltage of LIB module) The current value of the charge / discharge time series data of the LIB module based on the simulated operation pattern shown in Fig. 9 and the V oc Based on the profile of R, the charge / discharge voltage V of the LIB module is calculated based on the following equation (9): c,d was estimated sequentially.

[0085]

number

[0086] Charge / discharge voltage V c,d The graph in Figure 12 shows the change in the estimation error over time. The horizontal axis of the graph shows the elapsed time [×10 s], and the vertical axis shows the charge / discharge voltage Vc,d The graph shows the estimated error [V] of the sampling data. The solid line in the graph shows the actual measured value, and the dashed line shows the moving average.

[0087] The moving average of the estimated error decreased over time, and at the end, the error was shown to be a maximum of about 3 V compared to the LIB module's rated voltage of 248.4 V.

[0088] (Estimation of PCS efficiency curve) Based on the simulated BESS operation pattern shown in Figure 9, the PCS efficiency curve estimated by learning from the operation time series data of the LIB module and PCS using PCS-MGFFD, and the results of determining the trends in the estimated power consumption of PCS auxiliary devices are shown in the graphs of Figures 13(a) and (b), respectively. The horizontal axis of the graph in Figure 13(a) shows the rated input / output ratio on the AC side of the PCS, and the vertical axis shows the efficiency of the PCS. The different types of lines in the graph in Figure 13(a) represent different elapsed times. In other words, the graph in Figure 13(a) shows the efficiency over time (4 x 10 4 ) seconds. The horizontal axis of the graph in Fig. 13(b) represents the elapsed time [×10 s], and the vertical axis represents the auxiliary power consumption [W].

[0089] Figure 13(a) shows that the learning results of the PCS efficiency curve converged at the end of the simulated operation, and Figure 13(b) shows that the estimated auxiliary power consumption converged within a certain range.

[0090] (Estimation of LIB module input / output from PCS input / output) Based on the learning results of BT-MGFFD and PCS-MGFFD shown in Fig. 11 and Fig. 13, the results of estimating the charge / discharge power on the DC side of the LIB module from the input / output power on the AC side of the PCS are shown in the graphs of Fig. 14(a), (b), and (c). The horizontal axis of the graph in Fig. 14(a) indicates the elapsed time [× 10 s], and the vertical axis indicates the charge / discharge power on the DC side of the LIB module [W]. The solid line in the graph in 14(a) indicates the actual measured value, and the dashed line indicates the moving average. The horizontal axis of the graph in Fig. 14(b) indicates the elapsed time [× 10 s], and the vertical axis indicates the estimation error [W] of the charge / discharge power on the DC side of the LIB module. The solid line in the graph in 14(b) indicates the actual measured value of the time series data, and the dashed line indicates the moving average. The graph in Fig. 14(c) is an enlarged view of a part of the graph in Fig. 14(b). The vertical axis of each graph indicates charging as positive and discharging as negative.

[0091] In Figure 14(b), a peak is seen that is thought to be due to the difference in response time between the AC and DC sides of the PCS. However, as shown in Figure 14(c), the moving average converges to an error of less than 30 W, which is considered to be an acceptable level of estimation accuracy for a certain time period.

[0092] (Development towards operational efficiency and economic optimization) Open circuit voltage of LIB estimated by BT-MGFFD and PCS-MGFFD V oc From the characteristic profile of the internal resistance R (Fig. 11), and the efficiency curve of the PCS and the transition of the auxiliary power consumption (Fig. 13), the graph shown in Fig. 15 is obtained, that is, the charging rate S of the LIB. oc It is possible to obtain information on the estimated RTE of the entire BESS with respect to the rated input / output ratio on the AC side of the PCS and the rated output / input ratio on the AC side of the BESS. It is self-evident that this estimated RTE varies depending on conditions such as the type of BESS and the aging deterioration of the installed LIB and PCS. Therefore, when configuring a cloud battery management system that bundles and controls a large number of BESSs using ICT, the MGFFD method described here is considered to be an effective means of optimizing the overall operational efficiency and economy.

[0093] <Operation of the deterioration state estimating device> Next, an example of the operation of the degradation state estimating device 6 of this embodiment will be described with reference to the flowchart shown in FIG. First, an initial value / initial function derivation process is performed (step S1). Specifically, the initial open circuit voltage function V oc (S oc ) and the initial internal resistance function R(S oc ) to derive an approximate curve.

[0094] Next, the discharge unit 23 electrically connects the secondary battery 11 to the discharge load unit 24 to start discharging, and the timer unit 26 starts measuring time (step S2).

[0095] After the discharge starts, it is determined whether the voltage is at the lower limit voltage (step S3). If the voltage is found to have reached the lower limit voltage (YES) in the determination in step S3, charging is started (step S8). On the other hand, if the voltage is not found to have reached the lower limit voltage (NO), discharging is continued.

[0096] After the discharge starts, it is determined whether a specified time has elapsed (step S4). In the determination in step S4, the discharge continues until the specified time has elapsed (NO), and if the specified time has elapsed (YES), the measurement unit 25 measures the discharge voltage and the charge / discharge current of the secondary battery 11, and the SOC calculation unit 28 and the like calculate the DC voltage and current, the SOC value, and the AC voltage and current (step S5), and machine learning processing is performed (step S6). At this time, the Kalman filter of the state estimation unit 30 is used to repeatedly perform corrections to improve the accuracy of the ex-post estimation approximation curve.

[0097] Next, it is determined whether the number of repeated corrections has reached a predetermined number of repetitions (step S7). If the number of corrections has not reached the number of repetitions (NO), the process returns to step S3 and continues discharging. If discharging is to be continued, it is determined again whether the voltage is at the lower limit of discharge (step S3), and if the lower limit of discharge has not been reached (NO), the process routine from step S3 to step S7 is repeated. If the number of corrections has reached the number of repetitions (YES) in the determination in step S7, the charge / discharge curve of the LIB module, the efficiency curve of the LIB, and the efficiency curve of the PCS are estimated using approximate curves of the open circuit voltage function and the internal resistance function (step S14). The estimated charge / discharge curve of the LIB module and the efficiency curve of the PCS are temporarily stored in the server 16 of the energy management unit 5.

[0098] If it is determined in step S3 that the lower limit discharge voltage has been reached (YES), charging is started (step S8). After charging is started, it is determined whether the voltage has reached the upper limit charge voltage (step S9). If it is determined in step S9 that the upper limit charge voltage has not been reached (NO), it is determined whether a predetermined time has elapsed (step S10). On the other hand, if it is determined that the upper limit charge voltage has been reached (YES), the process returns to step S2.

[0099] In the determination of step S10, after a predetermined time has elapsed (YES), the measurement unit 25 measures the discharge voltage V c,d and the charge / discharge current I are measured, and the SOC calculation unit 28 and the like calculate the DC voltage and current, the SOC value, and the AC voltage and current (step S11), and machine learning processing is performed (step S12). At this time, the Kalman filter of the state estimation unit 30 is used to repeatedly perform corrections to improve the accuracy of the ex-post estimation approximation curve.

[0100] Next, it is determined whether the number of repeated corrections has reached a predetermined number of repetitions (step S13). If it is determined in step S13 that the number of repetitions has not been reached (NO), the process returns to step S9 and charging is continued. If charging is to be continued, it is determined in step S9 whether the charging voltage is at the upper limit voltage, and if the charging voltage has not been reached and it is determined in step S13 that the number of repetitions has not been reached, the processing routine from step S9 to step S12 is repeated.

[0101] In the determination in step S13, if the number of repetitions has been reached (YES), the process proceeds to step S14.

[0102] Next, the estimation calculation processing unit 27 estimates the charge / discharge curve of the LIB module, the efficiency curve of the LIB, and the efficiency curve of the PCS using approximation curves of the open circuit voltage function and the internal resistance function, and stores them, for example, in the server 16 of the energy management unit 5 (step S14).

[0103] Finally, the estimation calculation processing unit 27 estimates the charge / discharge efficiency characteristics of the power storage system 1 based on the efficiency curve of the LIB and the efficiency curve of the PCS, and stores them, for example, in the server 16 of the energy management unit 5 (step S15).

[0104] The stored charge / discharge curves of the LIB module and the efficiency curves of the PCS can be read out from the server 16 upon request and used for other purposes in addition to being displayed on the display unit 15.

[0105] <Summary> In this embodiment, the MGFFD method is applied not only to estimating the open circuit voltage curve and the internal resistance curve of the LIB module, but also to estimating the efficiency curve and auxiliary power consumption of the PCS. That is, a combination of BT-MGFFD and PCS-MGFFD is adopted. This makes it possible to accurately estimate the charge / discharge energy efficiency based on the AC input / output terminal of the PCS, taking into account the charging rate of the LIB and the rated input / output ratio of the PCS. This also makes it possible to estimate the RTE of the entire BESS, as well as the final utilization rate and operational economy in the temporary storage of renewable energy, and enables the optimization of the operation of the cloud battery management system in a microgrid, etc.

[0106] The present invention is not limited to the above-described embodiment, and may be modified in various ways without departing from the spirit and scope of the present invention. Furthermore, various inventions that solve the above problems can be extracted by selecting or combining multiple disclosed constituent elements. [Explanation of symbols]

[0107] 1...energy storage system, 2...power conditioner (PCS), 3...LIB module, 4...battery management unit, 5...energy management unit, 6...deterioration state estimation device, 7...battery temperature measurement unit, 8...specific load, 9...power system, 11...secondary battery, 12...cell monitoring unit, 13...protection unit, 14...arithmetic and control unit, 15...display unit, 16...server, 17...interface unit, 18...network communication network, 19...external device, 20...system server, 21...measurement unit, 22...charging power supply unit, 23...discharging unit, 24...discharging load unit, 25...measurement unit, 26...time measurement unit, 27...estimation calculation processing unit, 28...SOC calculation unit, 29...function derivation unit, 30...state estimation unit.

Claims

1. A degradation state estimation device that estimates a degradation state of a power storage system including a secondary battery and an orthogonal converter, a degradation state estimation device comprising an estimation calculation processing unit that estimates an open circuit voltage and internal resistance relative to the charging rate of the secondary battery based on the charging and discharging voltage and charging and discharging current of the secondary battery and the input and output power of an AC converter, and that estimates the efficiency of the AC converter and the auxiliary power consumption of the AC converter based on the estimated open circuit voltage, the internal resistance, the efficiency of the AC converter, and the auxiliary power consumption of the AC converter.

2. The estimation calculation processing unit: estimating a deterioration state of a power storage system including the secondary battery and the AC converter by estimating a charging / discharging energy efficiency with respect to a charging rate of the secondary battery and a rated input / output ratio of the AC side of the AC converter; The degradation state estimating device according to claim 1 .

3. The estimation calculation processing unit: estimating an efficiency curve of the orthogonal transformer by defining the efficiency of the orthogonal transformer as a vector corresponding to a vector of discrete values ​​of a rated input-output ratio; The degradation state estimating device according to claim 1 .

4. a state estimation unit that estimates state parameters including a charging rate, an open circuit voltage, and an internal resistance of the secondary battery using a Kalman filter corresponding to nonlinearity; The degradation state estimating device according to claim 1 .

5. The degradation state estimation device is provided outside the power storage system. The degradation state estimating device according to claim 1 .

6. A method for estimating a degradation state in a power storage system including a secondary battery and an orthogonal converter, comprising: and performing an estimation calculation process for estimating an open circuit voltage for the charging rate of the secondary battery, estimating an internal resistance, estimating the efficiency of the orthogonal converter, estimating the auxiliary power consumption of the orthogonal converter, and estimating a deterioration state of the power storage system by estimating a charge / discharge energy efficiency for the charging rate of the secondary battery and a rated input / output ratio of the orthogonal converter based on the estimated open circuit voltage curve, the internal resistance curve, the efficiency curve of the orthogonal converter, and a history of the auxiliary power consumption of the orthogonal converter. Deterioration state estimation method.

7. The estimation calculation process includes: estimating a deterioration state of a power storage system including the secondary battery and the AC converter by estimating a charging / discharging energy efficiency with respect to a charging rate of the secondary battery and a rated input / output ratio of an AC side of the AC converter; The degradation state estimating method according to claim 6.

8. The estimation calculation process includes: estimating an efficiency curve of the quadrature converter by defining the efficiency of the quadrature converter as a vector corresponding to a vector of discrete values ​​of a rated input-output ratio; The degradation state estimating method according to claim 6.

9. and estimating state parameters including a charging rate, an open circuit voltage, and an internal resistance of the secondary battery using a Kalman filter corresponding to nonlinearity. The degradation state estimating method according to claim 6.

10. A method for optimizing operation of a power storage system, which estimates the charge / discharge energy efficiency and operational economy of the power storage system based on a result of degradation state estimation performed by the degradation state estimation device according to claim 1 or the degradation state estimation method according to claim 6.

Images