阅读说明：本技术 一种电动汽车绝缘状态在线检测方法 (Online detection method for insulation state of electric automobile ) 是由 李小凡 李慧媛 何佳昊 姚金泽 于 2021-07-21 设计创作，主要内容包括：本发明公开了一种电动汽车绝缘状态在线检测方法,包括：步骤1,建立系统动态模型,并根据其阶跃响应,利用非线性最小二乘法进行模型参数辨识,计算出绝缘电阻值R-(f1)和系统响应周期；步骤2,根据电桥法测量绝缘电阻,并采用神经网络对模型进行参数辨识,计算出绝缘电阻值R-(f2)；步骤3,当R-(f1)与R-(f2)的差值小于预设阈值时,则将较小的绝缘电阻值作为测量值R-(f)。所述检测电路包括动态模型电路及电桥电路。相比传统方法,检测时间缩短,提高绝缘电阻检测的实时性。(The invention discloses an electric automobile insulation state online detection method, which comprises the following steps: step 1, establishing a system dynamic model, identifying model parameters by utilizing a nonlinear least square method according to step response of the model, and calculating an insulation resistance value R f1 And a system response period; step 2, measuring the insulation resistance according to a bridge method, adopting a neural network to carry out parameter identification on the model, and calculating the insulation resistance R f2 (ii) a Step 3, when R is f1 And R f2 When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R f . The detection circuit comprises a dynamic model circuit and a bridge circuit. Compared with the traditional method, the detection time is shortened, and the real-time performance of the insulation resistance detection is improved.)

1. An electric automobile insulation state online detection method is characterized by comprising the following steps:

step 1, establishing a system dynamic model, and utilizing a non-linear model according to a step response of the system dynamic modelIdentifying model parameters by linear least square method, and calculating insulation resistance value R_f1And a system response period;

step 2, measuring the insulation resistance according to a bridge method, adopting a neural network to carry out parameter identification on the model, and calculating the insulation resistance R_f2；

Step 3, when R is_f1And R_f2When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R_f；

In the step 1, the detection circuit further comprises a dynamic model circuit, the dynamic model circuit comprises,

u and R_P、R_NForming a loop I, wherein U is the voltage of the power battery, and R_PIs the resistance, R, between the positive pole of the power battery and the vehicle shell_NThe resistance between the cathode of the power battery and the vehicle shell;

u and R₁、R₂Formed into a loop II in which R₁、R₂Is a current limiting resistor;

R₃forming a loop III with a pulse signal generator for generating a signal having an amplitude of U_SAt the step voltage of (3), the R₃Is a sampling resistor;

wherein, flows through R_P、R_N、R₁、R₂、R₃Respectively is I₁、I₂、I₃、I₄I, then I₁、I₂、I₃、I₄And I should satisfy:

I＝I₁+I₂(formula I)

I＝I₃+I₄(formula II).

2. The online detection method for the insulation state of the electric automobile according to claim 1, characterized in that: the step of obtaining the insulation resistance value detection method from the dynamic model circuit comprises the steps of,

21) calculating to obtain the current flowing through the sampling resistor R₃A current I on;

for the loop I, there is,

for the loop II, there is,

U/s＝R₁·I₃+R₂·I₄(formula IV)

For the loop III, there is,

the sampling resistance R flows through according to the formula I, the formula II, the formula III, the formula IV and the formula V₃And performing inverse Laplace transformation on the current I to obtain:

where t is a time variable, parameter a₁In response to the steady-state component, the parameter a₂In response to gain, parameter a₃Is a response time constant;

22) deriving the sampling resistance R₃Responsive to the steady-state component a₁，

The response steady-state component a₁And the response time constant a₃The parameter expression of (1) is:

where// denotes the equivalent resistance of several resistors connected in parallel, and R ═ R₁＝R₂，

Since the active component of the power cell voltage U can be eliminated in the subsequent operation, then,

23) based on responsive steady-state component a'₁Obtaining the resistance value of the equivalent insulation resistor,

the sampling resistor R₃Voltage U on₃The step response under the pulse injection signal is:

wherein the content of the first and second substances,

b₂＝a₂·R₃

b₃＝a₃

when the sampling resistor R is connected₃When the voltage on the voltage regulator is stable, neglecting the response gain a₂I.e. U₃＝b₁The formula VII is modified to obtain the equivalent insulation resistance R_f：

As can be seen from formula VI, the response value is a nonlinear model at time t, and parameter identification is performed by using a nonlinear least square method.

3. The online detection method for the insulation state of the electric vehicle according to claim 2, characterized in that the detection method further comprises a step of parameter identification based on a nonlinear least square model, comprising,

step 24) measurementMeasure m data points (t)_i，y_i)_{(i＝1，2，3，...，m)}And aligning the model Performing parameter identification;

the mean square error E (x) of the model f (t) satisfies:

x^T＝[b₁，b₂，b₃]

according to the principle of least square method, the condition for obtaining the minimum value of the mean square error E (x) is as follows:

solving by adopting a Newton iteration method based on the iteration rapidity;

according to formula (J)^T·J)·h＝-J^TF, obtaining an algorithm iteration step h by taking a minimum value of x from E (x), wherein J is a Jacobin matrix;

the convergence condition of the model f (t) is as follows:

|E(x+h)-E(x)|＜ε

wherein epsilon is a judgment value of the convergence condition, and epsilon is set according to the model parameters.

4. The online detection method for the insulation state of the electric automobile according to claim 3, characterized in that the algorithm steps of parameter identification are as follows:

41) reading data points (t)_i，y_i) Removing outliers, and performing moving average filtering to ensure the reliability of data;

42) given identification parameter x^TAssigning an initial value, wherein x^T＝[b₁，b₂，b₃]；

43) Calculating a Jacobin matrix J and calculating an iteration step length h;

44) judging whether convergence conditions | E (x + h) -E (x) | < epsilon are met;

if yes, considering E (x) converged, exiting iteration, and outputting the optimal parameter set And step 45) is performed;

if not, the identification parameter x is subjected to iteration step length h^TReassign the value and return to execute step 43);

45) the optimal parameters are measuredSubstituting into formula VIII to obtain equivalent insulation resistance R_f1(ii) a From an optimum response time constantAnd determining the response time of the system to obtain the measurement period of the system.

5. The online insulation state detection method of an electric vehicle according to claim 4, wherein a system response period in the system dynamic model is the same as a sampling period for measuring the insulation resistance according to a bridge method.

6. The online detection method for the insulation state of the electric automobile according to claim 1, characterized in that: the bridge circuit in the step 2 comprises,

a detection resistor including a resistor R₄Resistance R₅And a resistance R₆；

The insulation resistor comprisesAnd insulation resistance

A second end of the selection switch S is provided with A, B two ports;

and a DC power supply V_DC；

The connection mode of the bridge method measuring circuit is as follows:

the resistor R₄First terminal and the insulation resistorSecond terminal, said insulation resistanceIs connected to the first terminal of the resistor R₄The second end of the first switch is connected with a selection switch S;

the resistor R₅First terminal and the insulation resistorIs connected to the first terminal of the resistor R₅The second end of the second switch is connected with the end A of the selection switch S;

the resistor R₆Is connected to the terminal B of the selection switch S, and a resistor R₆Second terminal and insulation resistorAre connected with each other;

the insulation resistorAnd the first terminal of the DC power supply V_DCIs connected with the positive electrode of the insulating resistorThe second end of the second end is connected with the shell;

the insulation resistorThe first end of the resistor is connected with the shell, and the insulation resistorSecond terminal and dc power supply V_DCThe negative electrodes are connected;

the DC power supply V_DCIs connected with the vehicle shell.

Technical Field

The invention relates to the technical field of insulation state detection, in particular to an online detection method for the insulation state of an electric automobile.

Background

In recent years, the automotive industry, particularly hybrid and electric vehicles, has rapidly developed in order to overcome energy and environmental problems. Many parts of electric vehicles, including power batteries, motors, chargers, energy recovery devices, auxiliary battery charging devices, etc., all involve high voltage electrical insulation problems.

The voltage of the storage battery, the fuel cell and the super capacitor of the electric automobile far exceeds the safety limit of human body, and the working conditions of the elements are worse: vibration, acid and alkali gas corrosion, and temperature and humidity change can cause rapid aging and even insulation damage of the insulating materials of the components. If the insulation strength of the equipment is greatly reduced, the voltage of the whole electric automobile system can endanger the personal safety. When the insulation performance of a plurality of points between the high voltage circuit and the vehicle case is degraded, heat energy is accumulated, and in a more serious case, a fire or explosion may be caused, so that insulation state detection on the electric vehicle is necessary.

At present, the traditional detection schemes of the insulation resistance of the power battery of the electric automobile comprise a balanced bridge measurement method, an external injection low-frequency signal measurement scheme and an unbalanced bridge measurement method, the common measurement method is difficult to accurately measure the insulation state, and when a direct-current power supply V of the electric automobile_DCInsulating circuit between anode and vehicle shell and direct current power supply V_DCWhen an insulating circuit between the negative electrode and the vehicle shell breaks down simultaneously, the fault cannot be detected, and the fault is easily influenced by external factors.

Disclosure of Invention

The invention provides an electric automobile insulation state online detection method, which comprises the following steps:

step 1, establishing a system dynamic model, identifying model parameters by utilizing a nonlinear least square method according to step response of the model, and calculating an insulation resistance value R_f1And a system response period;

step 2, measuring the insulation resistance according to a bridge method, adopting a neural network to carry out parameter identification on the model, and calculating the insulation resistance R_f2；

Step 3, when R is_f1And R_f2When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R_f；

In the step 1, the detection circuit further comprises a dynamic model circuit, the dynamic model circuit comprises,

u and R_P、R_NForming a loop I, wherein U is the voltage of the power battery, and R_PIs the resistance, R, between the positive pole of the power battery and the vehicle shell_NThe resistance between the cathode of the power battery and the vehicle shell;

u and R₁、R₂Formed into a loop II in which R₁、R₂Is a current limiting resistor;

R₃forming a loop III with a pulse signal generator for generating a signal having an amplitude of U_SAt the step voltage of (3), the R₃Is a sampling resistor;

wherein, flows through R_P、R_N、R₁、R₂、R₃Respectively is I₁、I₂、I₃、I₄I, then I₁、I₂、I₃、I₄And I should satisfy:

I＝I₁+I₂(formula I)

I＝I₃+I₄(formula II)

Preferably, the step of obtaining the insulation resistance value detection method from the dynamic model circuit includes,

21) calculating to obtain the current flowing through the sampling resistor R₃A current I on;

for the loop I, there is,

for the loop II, there is,

U/s＝R₁·I₃+R₂·I₄(formula IV)

For the loop III, there is,

the sampling resistance R flows through according to the formula I, the formula II, the formula III, the formula IV and the formula V₃And performing inverse Laplace transformation on the current I to obtain:

where t is a time variable, parameter a₁In response to the steady-state component, the parameter a₂In response to gain, parameter a₃Is a response time constant;

22) deriving the sampling resistance R₃Responsive to the steady-state component a₁，

The response steady-state component a₁And the response time constant a₃The parameter expression of (1) is:

where// denotes the equivalent resistance of several resistors connected in parallel, and R ═ R₁＝R₂Since the active component of the power cell voltage U can be eliminated in the subsequent operation, then,

23) based on responsive steady-state component a'₁Obtaining the resistance value of the equivalent insulation resistor,

the sampling resistor R₃Voltage U on₃The step response under the pulse injection signal is:

wherein the content of the first and second substances,

b₂＝a₂·R₃

b₃＝a₃

when the sampling resistor R is connected₃When the voltage on the voltage regulator is stable, neglecting the response gain a₂I.e. U₃＝b₁The formula VII is modified to obtain the equivalent insulation resistance R_f：

As can be seen from formula VI, the response value is a nonlinear model at time t, and parameter identification is performed by using a nonlinear least square method.

Preferably, the detection method further comprises a step of parameter identification based on a nonlinear least squares model, comprising,

step 24) measuring m data points (t)_i，y_i)_{(i＝1，2，3，...，m)}And aligning the model Performing parameter identification;

the mean square error E (x) of the model f (t) satisfies:

according to the principle of least square method, the condition for obtaining the minimum value of the mean square error E (x) is as follows:

solving by adopting a Newton iteration method based on the iteration rapidity;

according to formula (J)^T·J)·h＝-J^TF, obtaining an algorithm iteration step h by taking a minimum value of x from E (x), wherein J is a Jacobin matrix;

the convergence condition of the model f (t) is as follows:

|E(x+h)-E(x)|＜ε

wherein epsilon is a judgment value of the convergence condition, and epsilon is set according to the model parameters.

Preferably, the algorithm steps of parameter identification are as follows:

41) reading data points (t)_i，y_i) Removing outliers, and performing moving average filtering to ensure the reliability of data;

42) given identification parameter x^TAssigning an initial value, wherein x^T＝[b₁，b₂，b₃]；

43) Calculating a Jacobin matrix J and calculating an iteration step length h;

44) judging whether convergence conditions | E (x + h) -E (x) | < epsilon are met;

if yes, considering E (x) converged, exiting iteration, and outputting the optimal parameter set And step 45) is performed;

if not, the identification parameter x is subjected to iteration step length h^TReassign the value and return to execute step 43);

45) the optimal parameters are measuredSubstituting into formula VIII to obtain equivalent insulation resistance R_f1(ii) a From an optimum response time constantAnd determining the response time of the system to obtain the measurement period of the system.

Preferably, the system response period in the system dynamic model is the same as the sampling period for measuring the insulation resistance according to the bridge method.

In the step 2, the insulation resistance is measured by using a bridge method, and a bridge method measuring circuit comprises:

a detection resistor including a resistor R₄Resistance R₅And a resistance R₆；

The insulation resistor comprisesAnd insulation resistance

A second end of the selection switch S is provided with A, B two ports;

and a DC power supply V_DC。

The connection mode of the bridge method measuring circuit is as follows:

the resistor R₄First terminal and the insulation resistorSecond terminal, said insulation resistanceIs connected to the first terminal of the resistor R₄The second end of the first switch is connected with a selection switch S;

the resistor R₅First terminal and the insulation resistorIs connected to the first terminal of the resistor R₅The second end of the second switch is connected with the end A of the selection switch S;

the resistor R₆Is connected to the terminal B of the selection switch S, and a resistor R₆Second terminal and insulation resistorSecond end phase ofConnecting;

the insulation resistorAnd the first terminal of the DC power supply V_DCIs connected with the positive electrode of the insulating resistorThe second end of the second end is connected with the shell;

the insulation resistorThe first end of the resistor is connected with the shell, and the insulation resistorSecond terminal and dc power supply V_DCThe negative electrodes are connected;

the DC power supply V_DCIs connected with the vehicle shell and has 0 potential.

The invention adopts the measuring resistor R₅And R₆To detect an insulation state between a battery and a case of an electric vehicle.

Based on the bridge method measuring circuit, the invention discloses an online detection method for measuring insulation resistance by using a bridge method, which adopts a neural network to identify parameters of a bridge circuit model, and comprises the following steps:

step 51, connecting an A end of the selector switch S into a circuit to build an asymmetric circuit for analyzing the insulation state;

and step 52, connecting a circuit to the terminal B of the selector switch S to build an asymmetric circuit for analyzing the insulation state.

In step 51, based on the DC power supply V_DC、R₄、R₅、Formed equivalent circuit, DC power supply V_DCHas a resistance R between the positive electrode and the vehicle body_aD.C. power supply V_DCHas a resistance R between the negative electrode and the vehicle body_bThen R is_a、R_bSatisfies the following conditions:

DC power supply V_DCVoltage V between the positive pole of (b) and the vehicle casing_aComprises the following steps:

R₅voltage V on_R5Comprises the following steps:

in step 52, based on the DC power supply V_DC、R₄、R₆、Formed equivalent circuit, DC power supply V_DCHas a resistance R between the positive electrode and the vehicle body_cD.C. power supply V_DCHas a resistance R between the negative electrode and the vehicle body_dThen R is_c、R_dSatisfies the following conditions:

DC power supply V_DCVoltage between the negative electrode and the vehicle shell is V_dComprises the following steps:

R₆voltage V on_R6Comprises the following steps:

by reverse calculation, the insulation resistance is expressed as:

further, to simplify circuit design, the R is set to be₅、R₆Set as resistors with the same resistance value, thenSatisfies the following conditions:

wherein R is_e＝R₄+R₅＝R₄+R₆

From this, the insulation resistance is obtained

The insulation resistance R obtained based on the dynamic model circuit_f1And is based onInsulation resistance R obtained by the bridge circuit_f2Comparing: when R is_f1And R_f2When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R_f。

In the bridge circuit disclosed by the invention, an asymmetric circuit is built through a selection switch S, and the voltage of a detection resistor is measured: solving for V by step 51_R5V is solved by step 52_R6And is based on V_R5And V_R6And reversely calculating the resistance value of the insulation resistor, and reflecting the insulation state of the circuit through the calculated resistance value of the insulation resistor.

One of the above technical solutions has the following advantages and beneficial effects:

the bridge circuit detection system disclosed by the invention is connected with different end points through the change-over switch S to construct an unbalanced circuit, and the voltage of two detection resistors is measured to solve the insulation resistor, so that whether the insulation state of the electric automobile is qualified or not is deduced. The problem that when a circuit between a direct current power supply anode and a car shell of an electric car and a circuit between a direct current power supply cathode and the car shell break down simultaneously, the circuit cannot be accurately judged is solved.

On the basis of a low-frequency injection method steady-state measurement model, a system dynamic model is established, model parameter identification is carried out by utilizing a nonlinear least square method according to step response of the model, and insulation resistance R is accurately calculated_f1And the system response period, and further measuring the insulation resistance R by a bridge method_f2On the one hand, according to the response sequence, a nonlinear least square method is adopted to identify circuit model parameters, and the equivalent insulation resistance value and the measurement period are calculated according to the model parameters. The measurement method not only effectively inhibits measurement noise, but also can adjust the measurement period in real time, improves the accuracy and response speed of the insulation resistance measurement system, and on the other hand, the accuracy of the result can be further ensured by the two measurement methods, and meanwhile, the influence caused by external adverse factors can not be eliminated one by one, and the effectiveness and accuracy of measurement are improved.

Drawings

FIG. 1 is a schematic connection diagram of a bridge circuit according to the present embodiment;

FIG. 2 is a circuit diagram illustrating the connection of the terminal A of the selection switch S in step 51 according to the present embodiment;

FIG. 3 is a circuit diagram illustrating the connection of the terminal B of the selection switch S in step 52 according to the present embodiment;

FIG. 4 is the insulation resistance data of the bridge circuit of this embodiment under different measuring resistances;

FIG. 5 shows the insulation resistance data of the bridge circuit of this embodiment under different power voltages;

FIG. 6 is an equivalent circuit diagram of a dynamic model according to the present embodiment;

FIG. 7 is a flowchart illustrating dynamic model identification according to the present embodiment.

Detailed Description

To facilitate an understanding of this patent, it will now be described more fully with reference to the accompanying drawings. Embodiments of the present application are set forth in the accompanying drawings. This application may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. The meaning is generally understood by the skilled person. The terminology used in the description of the patent herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

In the description of the present invention, it is to be understood that the terms "first end", "second end", "upper", "lower", "left", "right", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplification of description, but do not indicate or imply that the elements referred to must have a specific orientation, and thus, are not to be construed as limiting the present invention. In the description of the present invention, it is specifically stated that: if the component is vertically placed in the circuit diagram, the first end of the component means the upper end, and the second end of the component means the lower end; if this component is placed laterally in the circuit diagram, then "the first end of the component" means the left end and then "the second end of the component" means the right end.

In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "connected" and "connected" are to be interpreted broadly, e.g., as being fixed or detachable or integrally connected; they may be electrically connected, directly connected, indirectly connected through an intermediary, or interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

As used herein, the singular forms "a", "an" and "the" may include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises/comprising," "includes" or "including," etc., specify the presence of stated features, integers, steps, operations, components, parts, or combinations thereof, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof. Also, as used in this specification, the term "and/or" includes any and all combinations of the associated listed items.

A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

The invention provides an electric automobile insulation state online detection method, which comprises the following steps:

step 1, establishing a system dynamic model, identifying model parameters by utilizing a nonlinear least square method according to step response of the model, and calculating an insulation resistance value R_f1And a system response period;

step 2, measuring the insulation resistance according to a bridge method, adopting a neural network to carry out parameter identification on the model, and calculating the insulation resistance R_f2；

Step 3, when R is_f1And R_f2When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R_f；

In step 1, the detection circuit further comprises a dynamic model circuit, as shown in fig. 6, the dynamic model circuit comprises,

u and R_P、R_NForming a loop I, wherein U is the voltage of the power battery, and R_PIs the resistance, R, between the positive pole of the power battery and the vehicle shell_NThe resistance between the cathode of the power battery and the vehicle shell;

u and R₁、R₂Formed into a loop II in which R₁、R₂Is a current limiting resistor;

R₃forming a loop III with a pulse signal generator for generating a signal having an amplitude of U_SAt the step voltage of (3), the R₃Is a sampling resistor;

wherein, flows through R_P、R_N、R₁、R₂、R₃Respectively is I₁、I₂、I₃、I₄I, then I₁、I₂、I₃、I₄And I should satisfy:

I＝I₁+I₂(formula I)

I＝I₃+I₄(formula II)

Preferably, the step of obtaining the insulation resistance value detection method from the dynamic model circuit includes,

21) calculating to obtain the current flowing through the sampling resistor R₃A current I on;

for the loop I, there is,

for the loop II, there is,

U/s＝R₁·I₃+R₂·I₄(formula IV)

For the loop III, there is,

the sampling resistance R flows through according to the formula I, the formula II, the formula III, the formula IV and the formula V₃And performing inverse Laplace transformation on the current I to obtain:

where t is a time variable, parameter a₁In response to the steady-state component, the parameter a₂In response to gain, parameter a₃Is a response time constant;

22) deriving the sampling resistance R₃Responsive to the steady-state component a₁，

The response steady-state component a₁And the response time constant a₃The parameter expression of (1) is:

where// denotes the equivalent resistance of several resistors connected in parallel, and R ═ R₁＝R₂，

Since the active component of the power cell voltage U can be eliminated in the subsequent operation, then,

23) based on responsive steady-state component a'₁Obtaining the resistance value of the equivalent insulation resistor,

the sampling resistor R₃Voltage U on₃The step response under the pulse injection signal is:

wherein the content of the first and second substances,

b₂＝a₂·R₃

b₃＝a₃

when the sampling resistor R is connected₃When the voltage on the voltage regulator is stable, neglecting the response gain a₂I.e. U₃＝b₁The formula VII is modified to obtain the equivalent insulation resistance R_f：

As can be seen from formula VI, the response value is a nonlinear model at time t, and parameter identification is performed by using a nonlinear least square method.

Preferably, the detection method further comprises a step of parameter identification based on a nonlinear least squares model, comprising,

step 24) measuring m data points (t)_i，y_i)_{(i＝1，2，3，...，m)}And aligning the model Performing parameter identification;

the mean square error E (x) of the model f (t) satisfies:

according to the principle of least square method, the condition for obtaining the minimum value of the mean square error E (x) is as follows:

solving by adopting a Newton iteration method based on the iteration rapidity;

according to formula (J)^T·J)·h＝-J^TF, obtaining an algorithm iteration step h by taking a minimum value of x from E (x), wherein J is a Jacobin matrix;

the convergence condition of the model f (t) is as follows:

|E(x+h)-E(x)|＜ε

wherein epsilon is a judgment value of the convergence condition, and epsilon is set according to the model parameters.

Preferably, the algorithm steps of parameter identification are as follows, as shown in fig. 7:

41) reading data points (t)_i，y_i) Removing outliers, and performing moving average filtering to ensure the reliability of data;

42) given identification parameter x^TAssigning an initial value, wherein x^T＝[b₁，b₂，b₃]；

43) Calculating a Jacobin matrix J and calculating an iteration step length h;

44) judging whether convergence conditions | E (x + h) -E (x) | < epsilon are met;

if yes, considering E (x) converged, exiting iteration, and outputting the optimal parameter set And step 45) is performed;

if not, the identification parameter x is subjected to iteration step length h^TReassign the value and return to execute step 43);

45) the optimal parameters are measuredSubstituting into formula VIII to obtain equivalent insulation resistance R_f1(ii) a From an optimum response time constantAnd determining the response time of the system to obtain the measurement period of the system.

Preferably, the system response period in the system dynamic model is the same as the sampling period for measuring the insulation resistance according to the bridge method.

In step 2, the insulation resistance is measured by using a bridge method, referring to fig. 1, the bridge method measurement circuit includes:

a detection resistor including a resistor R₄Resistance R₅And a resistance R₆；

The insulation resistor comprisesAnd insulation resistance

A second end of the selection switch S is provided with A, B two ports;

and a DC power supply V_DC。

The connection mode of the bridge method measuring circuit is as follows:

the resistor R₄First terminal and the insulation resistorSecond terminal, said insulation resistanceIs connected to the first terminal of the resistor R₄The second end of the first switch is connected with a selection switch S;

the resistor R₅First terminal and the insulation resistorIs connected to the first terminal of the resistor R₅The second end of the second switch is connected with the end A of the selection switch S;

the resistor R₆Is connected to the terminal B of the selection switch S, and a resistor R₆Second end of (2)And insulation resistanceAre connected with each other;

the insulation resistorAnd the first terminal of the DC power supply V_DCIs connected with the positive electrode of the insulating resistorThe second end of the second end is connected with the shell;

the insulation resistorThe first end of the resistor is connected with the shell, and the insulation resistorSecond terminal and dc power supply V_DCThe negative electrodes are connected;

the DC power supply V_DCIs connected with the vehicle shell and has 0 potential.

The invention adopts the measuring resistor R₅And R₆To detect an insulation state between a battery and a case of an electric vehicle.

Based on the bridge method measuring circuit, the invention discloses an online detection method for measuring insulation resistance by using a bridge method, which adopts a neural network to identify parameters of a bridge circuit model, and comprises the following steps:

step 51, connecting an A end of the selector switch S into a circuit to build an asymmetric circuit for analyzing the insulation state;

and step 52, connecting a circuit to the terminal B of the selector switch S to build an asymmetric circuit for analyzing the insulation state.

In step 51, referring to fig. 2, the DC power supply V is used_DC、R₄、R₅、Formed equivalent circuit, DC power supply V_DCHas a resistance R between the positive electrode and the vehicle body_aD.C. power supply V_DCHas a resistance R between the negative electrode and the vehicle body_bThen R is_a、R_bSatisfies the following conditions:

DC power supply V_DCVoltage V between the positive pole of (b) and the vehicle casing_aComprises the following steps:

R₅voltage V on_R5Comprises the following steps:

in step 52, referring to fig. 3, the dc power V is supplied to the battery_DC、R₄、R₆、Formed equivalent circuit, DC power supply V_DCHas a resistance R between the positive electrode and the vehicle body_cD.C. power supply V_DCHas a resistance R between the negative electrode and the vehicle body_dThen R is_c、R_dSatisfies the following conditions:

DC power supply V_DCVoltage between the negative electrode and the vehicle shell is V_dComprises the following steps:

R₆voltage V on_R6Comprises the following steps:

by reverse calculation, the insulation resistance is expressed as:

further, to simplify circuit design, the R is set to be₅、R₆Set as resistors with the same resistance value, thenSatisfies the following conditions:

wherein R is_e＝R₄+R₅＝R₄+R₆

From this, the insulation resistance is obtained

The insulation resistance R obtained based on the dynamic model circuit_f1And an insulation resistance R obtained based on the bridge circuit_f2Comparing: when R is_f1And R_f2When the difference is smaller than a preset threshold value, the smaller insulation resistance value is taken as a measured value R_f。

In the bridge circuit disclosed by the invention, an asymmetric circuit is built through a selection switch S, and the voltage of a detection resistor is measured: solving for V by step 51_R5V is solved by step 52_R6And is based on V_R5And V_R6And reversely calculating the resistance value of the insulation resistor, and reflecting the insulation state of the circuit through the calculated resistance value of the insulation resistor.

In the present embodiment, the insulation state of each circuit is detected by measuring the voltage of the detection resistor and obtaining the integrated resistance of the actual circuit. The influence of the measuring value on an actual circuit is reduced as much as possible while the insulation state of the circuit and a direct-current power supply system is considered.

Furthermore, fig. 4 shows one possible parameter set: FIG. 4 reflects the resistance R₅And R₆Insulation resistance measured at different resistance valuesAndand the actual insulation resistanceAndrelative error rate between resistance values. In this embodiment, V_DC＝800V；R₄＝100kΩ； Listed in the table are insulation resistancesAndis measured. By varying the resistance R₅And R₆To obtain different insulation resistancesAndand the actual insulation resistanceAndrelative error rate between resistance values.

The relative error rate in this example is calculated as follows:

fig. 5 shows another possible parameter set: FIG. 5 reflects the measured insulation resistance at different supply voltagesAndinsulation resistance from the actualAnderror between, in this embodiment R₄＝100kΩ；R₅＝1000kΩ；R₆＝1000kΩ：Listed in the table are insulation resistancesAndis measured. By varying the value of the supply voltage, different insulation resistances are obtainedAndand the actual insulation resistanceAndrelative error rate between resistance values.

The method provided by the invention can be used for detecting the insulation state of the circuit when the positive circuit and the negative circuit of the electric automobile simultaneously have faults.

The present invention is not limited to the above preferred embodiments, and any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.