A method for actively isolating power battery formation equipment fault

Abstract

The present invention discloses a method for actively isolating faults of power battery formation equipment, and belongs to the technical field of power battery fault diagnosis. The method obtains a state model of power battery formation equipment containing unknown but bounded noise and faults, designs a space update strategy based on the idea of ​​fully symmetric polytope set membership filtering, and constructs a state observer. At this time, it is not necessary to assume that the prior distribution of model noise is known, thereby enhancing the robustness and practicality of the algorithm; by introducing test points to perform fault detection, the tangency of the state feasible set and the fault set is expressed as the critical condition for the occurrence of the fault, and this critical condition is described as two optimization problems to directly obtain the value of the auxiliary signal, thereby achieving effective fault isolation, and avoiding the complexity of the problem caused by the traditional method of relying on the observer performance index to design the auxiliary signal; simulation examples have proved that compared with the traditional method, the method of the present application is more sensitive to minor faults, can better avoid delays or missed detection of faults, and improves the efficiency of fault detection and isolation.

Description

Technical Field

[0001] The invention relates to a method for actively isolating power battery formation equipment fault, and belongs to the technical field of power battery fault diagnosis. Background Art

[0002] In recent years, driven by the rapid expansion of the new energy vehicle market and other factors, the demand for high-end power battery products in many fields such as scientific research, industrial production and daily life has shown a significant upward trend. The performance of the battery is directly related to the operating efficiency and service life of the equipment, so the quality control of the battery has become a key link in the production process. In the production process of power batteries, the formation process plays a vital role. This link activates the lithium battery to ensure the performance consistency of the battery during the assembly process, and becomes the key to affecting the quality control of power batteries. The technical level and operating accuracy of this process are directly related to the reliability, safety and service life of the battery. Therefore, by taking active fault detection and isolation measures, the technical level and process quality of the formation process can be improved, and the sustainable and healthy development of the new energy vehicle industry can be promoted.

[0003] The battery formation process refers to the process of using a formation power converter to repeatedly charge and discharge the battery during the battery manufacturing process to activate its internal chemical activity. Active fault isolation using auxiliary signals is a widely used fault diagnosis method that uses the differences between system inputs and outputs to actively separate and identify faults in the system by adding specific auxiliary signals to the system.

[0004] Traditional auxiliary signal design methods, such as auxiliary signal design based on frequency domain analysis and auxiliary signal design based on models, may be sensitive to noise and changes in system parameters when designing auxiliary signals, resulting in insufficient robustness in practical applications, that is, after the fault is identified, the fault cannot be effectively isolated in real time due to changes in system parameters. In addition, traditional auxiliary signal design methods often use observer performance indicators as the standard for auxiliary signal design, which not only increases the complexity of the problem, but also may reduce the efficiency of fault detection; in the early stage of the fault, the system performance may only decrease slightly, which may cause the auxiliary signal design method to fail to detect the fault, or detect the fault only after it has developed to a certain extent, thereby causing a delay in fault diagnosis. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a method for actively isolating faults of power battery formation equipment, which is implemented based on polycell filtering and auxiliary signal design. The method includes:

[0006] Step S1: Establishing a power battery formation equipment model including unknown but bounded noise and faults;

[0007] First, based on the similar functions and behaviors of the battery formation power conversion equipment and the Buck circuit in the electrical circuit, the power battery formation equipment is simplified to the buck converter circuit analysis, and the corresponding buck converter circuit model is obtained as follows:

[0008]

[0009] Among them, R load , L, C, and D are the load resistance, inductance, capacitance, and duty cycle parameter values ​​of the buck converter circuit respectively; select the capacitor voltage U c (t) and the inductor current I L (t) is the state of the system, so x(t) = [x1(t) x2(t)] T =[I L (t) U c (t)] T ; Input voltage E(t) and load current I load (t) is the external input variable of the system, and u(t) = [E(t)I load (t)] T .

[0010] After obtaining the corresponding buck converter circuit model, it is discretized to obtain the power battery formation equipment state model containing unknown but bounded noise and faults:

[0011]

[0012] in, represents the state of the i-th model at time k, with the capacitor voltage U in the buck converter circuit c (t) and the inductor current I L (t) as the state of the model; represents the input of the ith model at time k, i.e., the charge and discharge current and / or charge and discharge voltage; represents the output of the i-th model at time k, which is the load battery voltage or load battery capacity;

[0013] A [i] represents the known state transfer matrix, B [i] Represents a known input matrix, C [i] represents the known output matrix;

[0014] represents the unknown but bounded disturbance noise at time k; represents the unknown but bounded measurement noise at time k; h k Indicates the auxiliary input signal.

[0015] Step S2: Design a fully symmetric polyhedral observer to estimate the state at the next time k based on the state at the current time k-1, thereby obtaining a fully symmetric polyhedral expression that encapsulates the true state at time k.

[0016] i∈{0,...,ξ} is the corresponding model index, ξ represents the fault sequence number; when i=0, the model is in a normal state, indicating that the power battery formation equipment is operating normally, otherwise the model is in a fault state, indicating that the power battery formation equipment is faulty;

[0017] Step S3: Obtaining a measurement state set according to the power battery formation equipment model Expression of

[0018] Step S4: Based on the real state at time k In a fully symmetric polytope and measurement state set The auxiliary signal h used to separate the normal model and the fault model is k The design problem is transformed into the following optimization problem:

[0019] min h k

[0020]

[0021] represents the optimal polytopic outer bound containing the system state at time k in the normal model, represents the optimal polytopic external bounding of the system state at time k in the fault model;

[0022] Step S5: Find the normal model by adding test points and fault models The equivalent condition of the intersection being empty forms two different optimization problems. Solving the two optimization problems separately will result in two different auxiliary signals h1 and h2.

[0023] Usually, the state point that makes the polytope representing the normal model and the polytope representing the fault model have only one intersection is called a state separation point, and a series of state points with the same mathematical form and undetermined coefficients as the state separation point are called test points. This application finds the normal model by adding test points. and fault models Equivalence condition for the intersection to be empty.

[0024] Step S6: Select one of the auxiliary signals h1 and h2 as the final auxiliary signal h k The battery-to-electricity conversion equipment state model is introduced to actively isolate faults that occur when the equipment is in operation.

[0025] Optionally, in step S2, the expression for estimating the state at the next time k based on the state at the current time k-1 is:

[0026]

[0027] in, That is, corresponding to the i-th model, the state at the next moment k is estimated based on the state at the current moment k-1; It means that the center of the fully symmetric polyhedron predicted at time k by the i-th model based on the prediction at time k-1, represents the matrix generated by this polyhedron; is a combination of r+n x unit box consisting of unit intervals; r corresponds to the dimension of the fully symmetric polyhedron, n x corresponds to the dimension of the perturbation noise.

[0028] Optionally, step S3 includes:

[0029] According to the power battery formation equipment state model shown in formula (1), the output of the i-th model at time k is derived as follows: Then we get the measurement state set The expression is:

[0030]

[0031] Wherein, σ represents the relative coefficient of the central interval where the measurement noise is located relative to the unit interval. Optionally, in step S4, the real state at time k In a fully symmetric polytope With measuring tape In the intersection, the optimal polytope outside the intersection is bounded The center and generator matrix are:

[0032]

[0033]

[0034] in

[0035]

[0036] I represents the identity matrix and T represents the transpose operation.

[0037] Optionally, step S5 includes:

[0038] For λ1,λ2∈[0,1] that satisfy the following equations (10) and (11):

[0039]

[0040]

[0041] like:

[0042]

[0043] or

[0044]

[0045] Considering equations (12), (13) and λ1, λ2, The prior condition of , the optimization problem described by formula (9) is comprehensively expressed as:

[0046]

[0047]

[0048] or

[0049]

[0050]

[0051] Solve the two different optimization problems in equation (14) and equation (15) respectively, and obtain two different auxiliary signals h1 and h2.

[0052] Optionally, step S6 includes:

[0053] The auxiliary signal h is finally selected according to the following rules k :

[0054] h k =min{|h1|,|h2|} (16)

[0055] The auxiliary signal h k The battery-to-electricity conversion equipment state model is introduced to actively isolate faults that occur when the equipment is in operation.

[0056] Optionally, select the output matrix C [i] When

[01] , the output variable y(t) = U c (t).

[0057] The present application also provides a method for monitoring the operation of a power battery formation equipment, which uses the above method to actively isolate faults that occur during the operation of the power battery formation equipment.

[0058] The beneficial effects of the present invention are:

[0059] Different from the traditional auxiliary signal design method, by obtaining the state model of the power battery formation equipment containing unknown but bounded noise and faults, based on the idea of ​​fully symmetric polyhedral set membership filtering, a spatial update strategy is designed to obtain the optimal polyhedral external delimitation of the state quantity by taking the intersection of the prediction step fully symmetric polyhedral and the measurement band, and a state observer is constructed. At this time, there is no need to assume that the prior distribution of the model noise is known, which enhances the robustness and practicality of the algorithm; by introducing test points to perform fault detection, the tangency of the state feasible set and the fault set is expressed as the critical condition for the occurrence of the fault, and this critical condition is described as two optimization problems to directly obtain the auxiliary signal value, so as to achieve effective fault isolation and avoid the complexity of the problem caused by the traditional method of relying on the observer performance indicators to design auxiliary signals; simulation examples have proved that compared with the traditional method, the method provided by the present application is more sensitive to minor faults, can better avoid delayed or missed detection of faults, and improves the efficiency of fault detection and isolation. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 It is a flow chart of a method for actively isolating power battery formation equipment faults based on polycell filtering and auxiliary signal design disclosed in one embodiment of the present application.

[0062] Figure 2 It is an equivalent circuit model diagram of a power battery formation device (simplified to a buck circuit during analysis).

[0063] Figure 3A It is a simulation diagram of system state estimation and fault isolation results when the power battery fails and no auxiliary signal is added, at time k=100, disclosed in one embodiment of the present application;

[0064] Figure 3B It is a simulation diagram of system state estimation and fault isolation results of the system at time k=200 when no auxiliary signal is added when a power battery fails in one embodiment of the present application;

[0065] Figure 3C It is a simulation diagram of system state estimation and fault isolation results of the system at time k=300 when no auxiliary signal is added when a power battery fails in one embodiment of the present application;

[0066] Figure 3DIt is a simulation diagram of system state estimation and fault isolation results of the system at time k=400 when no auxiliary signal is added when a power battery fails in one embodiment of the present application.

[0067] Figure 4A It is a diagram of system state estimation and fault isolation results when a power battery fails and an auxiliary signal is added by an existing method in one embodiment of the present application, and the system is at time k=100;

[0068] Figure 4B It is a diagram of system state estimation and fault isolation results when a power battery failure occurs and an auxiliary signal is added by an existing method in one embodiment of the present application, and the system is at time k=200;

[0069] Figure 4C It is a diagram of system state estimation and fault isolation results when a power battery fails and an auxiliary signal is added by an existing method in one embodiment of the present application, at time k=300;

[0070] Figure 4D It is a diagram of system state estimation and fault isolation results of the system at time k=400 when an auxiliary signal is added using an existing method when a power battery fails in an embodiment of the present application.

[0071] Figure 5A It is a diagram of system state estimation and fault isolation results when a power battery failure occurs in one embodiment of the present application and an auxiliary signal is added by the method provided in the present application, and the system is at time k=100;

[0072] Figure 5B It is a diagram of system state estimation and fault isolation results when a power battery failure occurs in one embodiment of the present application and an auxiliary signal is added by the method provided in the present application, and the system is at time k=200;

[0073] Figure 5C It is a diagram of system state estimation and fault isolation results when a power battery failure occurs in one embodiment of the present application and an auxiliary signal is added by the method provided in the present application, and the system is at time k=300;

[0074] Figure 5D It is a diagram of system state estimation and fault isolation results of the system at time k=400 when an auxiliary signal is added using the method provided in the present application when a power battery fails in one embodiment of the present application. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the present invention clearer, the following will further describe the implementation mode of the present invention in detail with reference to the accompanying drawings. First, the basic theory involved in this application is introduced as follows:

[0076] (1) This application simplifies the battery formation device into a Buck circuit for analysis. This is based on the fact that the battery formation power conversion device and the Buck circuit have similar functions and behaviors in the electrical circuit, which are specifically manifested in the following five aspects:

[0077] 1. Voltage conversion: During the battery formation process, it is usually necessary to convert a higher DC voltage into a lower voltage required for battery charging; the Buck circuit is a step-down converter that can convert the input higher DC voltage into a lower stable output voltage. This is consistent with the voltage conversion requirements during the battery formation process.

[0078] 2. Energy transfer: During the battery formation process, energy needs to be transferred from the power grid to the battery, which involves energy transfer and conversion; the Buck circuit transfers and converts energy through its switching elements and energy storage elements (inductors and capacitors), so that the input energy can be efficiently transferred to the load (in the case of battery formation, the load is the battery).

[0079] 3. Control interface: In order to achieve accurate battery charging, a control system is required to adjust the charging current and voltage; Buck circuits are usually equipped with a control system that can control the output voltage and current by adjusting the switching frequency and duty cycle, which provides a controllable interface for battery formation.

[0080] 4. Stability: The output voltage and current need to be stable during the battery formation process to ensure the safety and performance of the battery; the Buck circuit is designed to provide a stable output, which can keep the output voltage stable even when the input voltage and load conditions change.

[0081] 5. Efficiency: Efficient energy conversion is very important for battery-based devices because it is related to energy utilization and cost; Buck circuit is known for its high efficiency and can transfer energy from input to output with less loss.

[0082] Therefore, in theoretical analysis, the battery formation device can be simplified to a Buck circuit.

[0083] (2) In the prior art, fault detection based on set membership estimation is generally completed by checking whether the approximate feasible set is an empty set. Its fault detection strategy can be expressed as: if the approximate feasible set is detected as an empty set, it is considered that the system has a fault; otherwise, it is considered that the system has no fault. For normal and fault models (i.e., normal state and fault state), the fault isolation process involves checking whether the intersection of two strips is empty. If the intersection is empty, it means that the system is normal. Otherwise, it cannot be determined that the system is normal, that is, the system has a fault.

[0084] (3) The relevant concepts and symbols of fully symmetric polytope set membership filtering are introduced as follows:

[0085] X = [a, b] represents an interval set {x∈X: a≤x≤b}, B = [-1, 1] represents a unit interval, B r is a unit box consisting of r unit intervals; the Minkowski sum of two sets X and Y is defined as An expression for a fully symmetric polytope is of the form where p∈R n , G∈R n×r , r is the dimension of this fully symmetric polyhedron.

[0086] Based on the above basic theory, the technical solution of this application is described in detail below in combination with specific embodiments:

[0087] Embodiment 1

[0088] This embodiment provides a method for actively isolating power battery formation equipment faults based on multi-cell filtering and auxiliary signal design. Figure 1 , the method comprising:

[0089] Step 1: Simplify the power battery formation equipment into a buck circuit analysis and obtain the following equation:

[0090]

[0091] Please refer to Figure 2 The buck converter circuit model shown in Figure 1 is shown in Figure 2, where R load , L, C, and D are the load resistance, inductance, capacitance, and duty cycle parameters of the buck circuit. c (t) and the inductor current I L (t) is the state of the system, so x(t) = [x1(t) x2(t)] T =[I L (t) U c (t)] T ; Input voltage E(t) and load current I load(t) is the external input variable of the system, and we have u(t) = [E(t) I load (t)] T .

[0092] Step 2: After discretizing the buck circuit model shown in formula (17), the power battery formation equipment state model including unknown but bounded noise and faults shown in formula (1) is obtained:

[0093]

[0094] Among them, i∈{0,...,ξ} is the corresponding model index, ξ represents the fault sequence number. When i=0, the model is in a normal state, otherwise it is in a fault state; the subscript k represents the discrete moment, represents the state of the i-th model at time k, Indicates n x dimensional real number space; represents the input of the i-th model at time k, which is the charge and discharge current (also known as the inductor current I L (t)) or the charge and discharge voltage (i.e. the capacitor voltage U c (t)), in this embodiment, it is a matrix composed of the two; A represents the output of the i-th model at time k, which is usually the load battery voltage or load battery capacity; [i] represents the known state transfer matrix, B [i] Represents a known input matrix, C [i] represents the known output matrix; represents the unknown but bounded disturbance noise at time k, with an initial value W represents the generator matrix of the unknown but bounded perturbation noise; represents the unknown but bounded measurement noise at time k, with an initial value V represents the generator matrix of the unknown but bounded measurement noise; h k Indicates the auxiliary input signal.

[0095] Step 3: Select the output matrix C [i] is [0 1], then the output variable y(t) = U c (t).

[0096] Step 4: According to the current state of k-1 Estimate the state at the next moment

[0097]

[0098]

[0099]

[0100] in, It means that the center of the fully symmetric polyhedron predicted at time k by the i-th model based on the prediction at time k-1, represents the generator matrix of this polyhedron; W is the generator matrix of the unknown but bounded perturbation noise.

[0101] Measurement noise In a central interval V = σB 1 Where (σ represents the relative coefficient of the center interval of the measurement noise relative to the unit interval), according to equation (1), the output measurement can be expressed as So the measurement state set can be described as a band space as follows:

[0102]

[0103] The actual state at time k In a fully symmetric polytope With measuring tape In the intersection, the optimal polytope outside the intersection is bounded The center and generator matrix can be written as:

[0104]

[0105]

[0106] in

[0107]

[0108] I represents the identity matrix and T represents the transpose operation.

[0109] Step 5: In order to separate the normal model and the fault model, the design of the auxiliary signal is transformed into the following optimization problem:

[0110] minh k

[0111]

[0112] represents the optimal polytopic outer bound containing the system state at time k in the normal model, represents the optimal polytopic outer bound containing the system state at time k in the fault model.

[0113] To solve this optimization problem, this application adds test points to find The equivalent condition is that for λ1,λ2∈[0,1] that satisfies the following equations (10) and (11):

[0114]

[0115]

[0116] like:

[0117]

[0118] or

[0119]

[0120] Considering equations (12), (13) and λ1, λ2, The prior condition of , the optimization problem described by formula (9) is comprehensively expressed as:

[0121]

[0122]

[0123] or

[0124]

[0125]

[0126] By solving the two different optimization problems in equation (14) and equation (15) respectively, two different auxiliary signals h1 and h2 can be obtained.

[0127] Step 6: Finally select the auxiliary signal h according to the following rules k :

[0128] h k =min{|h1|,|h2|} (16)

[0129] The auxiliary signal h k Introduce the battery power conversion equipment state model.

[0130] In this embodiment, after executing steps 1 to 6 within a predetermined time range, the normal model and the fault model of the system are separated, and active isolation of the fault occurring in the equipment operation state is completed.

[0131] In order to verify the effectiveness of the method of the present application, this embodiment uses the method of the present application and the existing method to simulate the power battery formation equipment (simplified as a buck circuit during analysis). Table 1 is the model parameters of the buck converter in the simulation experiment. The total number of iterations in the experimental process is k=400, and different numbers of iterations correspond to different moments.

[0132] Table 1 Buck converter model parameters

[0133] Physical quantity Physical meaning Parameter Value E Input voltage 24V C capacitance 470±10μF L inductance 560±10μH <![CDATA[R load ]]> Load resistance 10±0.05Ω D Duty Cycle 0.4

[0134] Figure 3A-3D , Figure 4A-4D , Figure 5A-5D The system state estimation and fault isolation result diagrams of the power battery formation equipment (simplified as buck circuit during analysis) at the same time without adding auxiliary signal method, using the existing method (the existing method can refer to Active fault detection based on set-membership approach for uncertain discrete-time systems, International Journal of Robust and Nonlinear Control.) and using the method provided by this application correspond to different moments, wherein the method of adding auxiliary signal by using the existing method is the method of obtaining auxiliary signal by designing an observer. Different from the auxiliary signal obtained by solving the optimization problem in this application, the auxiliary signal of the existing method is obtained by designing an observer. When designing an observer, on the one hand, it depends on setting a certain threshold to determine whether the system state is abnormal. For minor faults, it is difficult to detect because its characteristics may not be enough to exceed this threshold; on the other hand, the accuracy limitation of the observer design may make it insensitive to the changes of minor faults. If the reconstruction ability of the observer is not enough to capture the minor state changes caused by minor faults, then these faults may be ignored. In addition, the method of designing an observer may also involve complex mathematical operations and derivations, which increases the difficulty of algorithm design. .

[0135] from Figure 3A-3D It can be observed that when the auxiliary signal method is not added, the two polytopes representing the state feasible sets of the normal model and the fault model intersect. From the system state estimation and fault isolation result diagrams at different times, it can be seen that at this time, the buck converter cannot determine whether it is in a normal state or a fault state during operation.

[0136] from Figure 4A-4D It can be observed that after the auxiliary signal obtained by the existing method is introduced, the two polytopes representing the state feasible set of the normal model and the fault model are not completely separated, which shows that in some cases, the traditional auxiliary signal design method may not be able to achieve effective fault isolation. This is caused by the lack of sensitivity of the traditional method to minor faults. At this time, the system may continue to have faults that are easy to ignore during operation, leading to more serious industrial problems.

[0137] from Figure 5A-5DIt can be observed that after adding appropriate auxiliary signals using the method provided by the present application, the polyhedron representing the state feasible sets of the normal model and the fault model can be effectively separated, the state deviation is within an acceptable range, and minor faults can also be quickly diagnosed and effectively separated. This is because the auxiliary signal obtained by the active fault isolation method for power battery formation equipment based on polyhedral filtering and auxiliary signal design proposed in the present application during fault isolation is obtained by solving two optimization problems transformed from equivalent conditions under which the normal model and the fault model can be completely separated, thereby ensuring the effectiveness of the auxiliary signal without overly relying on the model accuracy; in addition, the addition of the auxiliary signal does not change the shape of the original state polyhedron or affect its properties, and does not affect the normal operating state of the equipment.

[0138] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0139] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for actively isolating power battery formation equipment faults, characterized in that: The method is implemented based on polyhedral filtering and auxiliary signal design, and the method comprises: Step S1: Establishing a power battery formation equipment model including unknown but bounded noise and faults; Step S2: Design a fully symmetric polyhedral observer to estimate the state at the next time k based on the state at the current time k-1, thereby obtaining a fully symmetric polyhedral expression that encapsulates the true state at time k. i∈{0,…,ξ} is the corresponding model index, ξ represents the fault sequence number; when i=0, the model is in a normal state, indicating that the power battery formation equipment is operating normally, otherwise the model is in a fault state, indicating that the power battery formation equipment is faulty; Step S3: Obtaining a measurement state set according to the power battery formation equipment model Expression of Step S4: Based on the real state at time k In a fully symmetric polytope and measurement state set The auxiliary signal h used to separate the normal model and the fault model is k The design problem is transformed into the following optimization problem: represents the optimal polytopic outer bound containing the system state at time k in the normal model, (i≠0) represents the optimal polytopic external bounding of the system state at time k in the fault model; Step S5: Find the normal model by adding test points and fault models The equivalent condition of the intersection being empty forms two different optimization problems. Solving the two optimization problems separately will result in two different auxiliary signals h1 and h2. Step S6: Select one of the auxiliary signals h1 and h2 as the final auxiliary signal h k Introducing the battery-to-power conversion equipment state model to complete the active isolation of faults that occur when the equipment is in operation; The power battery formation equipment model established in step S1 includes: Step S1.1, simplifying the power battery formation equipment into a buck converter circuit analysis to obtain a corresponding buck converter circuit model; Step S1.2, discretize the buck converter circuit model to obtain a power battery formation device state model containing unknown but bounded noise and faults: in, represents the state of the i-th model at time k, with the capacitor voltage U in the buck converter circuit c (t) and the inductor current I L (t) as the state of the model; represents the input of the ith model at time k, i.e., the charge and discharge current and / or charge and discharge voltage; represents the output of the i-th model at time k, which is the load battery voltage or load battery capacity; A [i] represents the known state transfer matrix, B [i] Represents a known input matrix, C [i] represents the known output matrix; represents the unknown but bounded disturbance noise at time k; represents the unknown but bounded measurement noise at time k; h k Indicates auxiliary signal.

2. The method according to claim 1, characterized in that In step S2, the expression for estimating the state at the next time k based on the state at the current time k-1 is: in, That is, corresponding to the i-th model, the state at the next moment k is estimated based on the state at the current moment k-1; It means that the center of the fully symmetric polyhedron predicted at time k by the i-th model based on the prediction at time k-1, represents the matrix generated by this polyhedron; is a combination of r+n x unit box consisting of unit intervals; r corresponds to the dimension of the fully symmetric polyhedron, n x corresponds to the dimension of the perturbation noise.

3. The method according to claim 2, characterized in that The step S3 comprises: According to the power battery formation equipment state model shown in formula (1), the output of the i-th model at time k is derived as follows: Then we get the measurement state set The expression is: Among them, σ represents the relative coefficient of the central interval where the measurement noise is located relative to the unit interval.

4. The method according to claim 3, characterized in that In step S4, the real state at time k In a fully symmetric polytope With measuring tape In the intersection, the optimal polytope outside the intersection is bounded The center and generator matrix are: in I represents the identity matrix and T represents the transpose operation.

5. The method according to claim 4, characterized in that The step S5 comprises: For λ1,λ2∈[0,1] that satisfy the following equations (10) and (11): like: or Considering equations (12), (13) and λ1, λ2, The prior condition of , the optimization problem described by formula (9) is comprehensively expressed as: or Solve the two different optimization problems in equation (14) and equation (15) respectively, and obtain two different auxiliary signals h1 and h2.

6. The method according to claim 5, characterized in that The step S6 comprises: The auxiliary signal h is finally selected according to the following rules k : h k =min{|h1|,|h2|} (16) The auxiliary signal h k The battery-to-electricity conversion equipment state model is introduced to actively isolate faults that occur when the equipment is in operation.

7. The method according to claim 6, characterized in that The buck converter circuit model obtained in step S1.1 is: Among them, R load , L, C, and D are the load resistance, inductance, capacitance, and duty cycle parameter values ​​of the buck converter circuit respectively; select the capacitor voltage U c (t) and the inductor current I L (t) is the state of the system, so x(t) = [x1(t)x2(t)] T =[I L (t)U c (t)] T ; Input voltage E(t) and load current I load (t) is the external input variable of the system, and u(t) = [E(t)I load (t)] T .

8. The method according to claim 7, characterized in that Select the output matrix C [i] When [01], the output variable y(t) = U c (t).

9. A method for monitoring the operation of a power battery formation equipment, characterized in that: The method adopts the method described in any one of claims 1 to 8 to actively isolate faults occurring during the operation of power battery formation equipment.

Images