Power battery pack top cover electric leakage safety detection method and system

Abstract

The invention provides a power battery pack top cover electric leakage safety detection method and system, and relates to the technical field of battery pack top cover electric leakage detection. A pulse signal source and a detection resistor are applied to a top cover of each battery in a to-be-detected battery pack to form a top cover insulation detection circuit; insulating materials between a top cover of the power battery and positive and negative poles are equivalent to positive and negative pole insulation resistors, an equivalent insulation circuit model is constructed, the positive and negative pole insulation resistors are calculated, and a dust precipitation model is established and trained to obtain a dust precipitation model. Environmental parameters and pole parameters of the battery at the current moment are processed through a sliding window method and then serve as input of a trained model, the dust thickness of the positive electrode and the dust thickness of the negative electrode are output through the model, and the resistance value of dust is calculated according to the real-time environmental parameters and the pole parameters of the battery; according to the detected insulation resistance of the positive pole, the insulation resistance of the negative pole, the resistance value of the dust and the battery voltage, the actual electric leakage amount and the real-time electric leakage amount are calculated.

Description

Technical Field

[0001] The present invention relates to the technical field of battery pack top cover leakage detection technology, and in particular to a power battery pack top cover leakage safety detection method and system. Background Art

[0002] Over the past few years, with the widespread application of power battery packs in electric vehicles, energy storage systems and other fields, battery safety has received increasing attention. In particular, the insulation performance of the battery pack top cover is crucial to preventing leakage failures.

[0003] However, in traditional battery pack safety testing methods, the insulation status of the battery's top cover is often determined through simple voltage or current testing. This method is difficult to accurately reflect the actual insulation condition of the top cover, and generally has problems such as inaccurate detection and poor real-time performance. Especially in complex environments, such as high temperature, high humidity, and dust, it is difficult to effectively evaluate the insulation resistance of the battery. As the battery is used for a longer time, factors such as dust and corrosion may cause the insulation layer of the battery top cover to age or become thinner, thereby affecting the safety of the battery pack. Therefore, there is an urgent need for a detection method that can accurately detect the insulation performance of the power battery top cover and reflect real-time and actual changes in battery leakage, so as to improve battery safety and reduce the risk of battery failure.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The object of the present invention is to provide a method and system for detecting leakage safety of a power battery pack top cover, so as to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A power battery pack top cover leakage safety detection method, the specific steps include:

[0008] Step 1: Apply the positive terminal of a pulse signal source to the top cover of each battery in the battery pack to be tested, connect the negative terminal of the pulse signal source to a detection resistor, and then connect the detection resistor in series with a current-limiting resistor and electrically connect it to the positive and negative terminals of the battery to form a top cover insulation detection circuit;

[0009] Step 2: Equivalently equate the insulation material between the power battery top cover and the positive and negative posts to the insulation resistance of the positive and negative posts. Build an equivalent insulation circuit model with the top cover insulation detection circuit. Calculate the insulation resistance of the positive and negative posts based on Kirchhoff's law by comparing the positive and negative voltages applied by the pulse signal source within the signal cycle with the detection voltage of the detection circuit within the cycle.

[0010] Step 3: Obtain the historical operating environment parameters and electrode parameters of different battery packs and stamp them with timestamps to form a time series dataset. Use the sliding window method to segment the time series dataset and establish a dust deposition model. Use the segmented time series dataset at the current moment as the model input and the dust thickness of the positive and negative electrodes at the next moment as the model output to train the model.

[0011] Step 4: The current battery environmental parameters and electrode parameters are processed using the sliding window method as input to the trained model. The model outputs the dust thickness at the positive and negative electrodes. Based on the real-time battery environmental parameters and electrode parameters, and the effects of temperature and humidity on the conductivity of the dust, the dust resistance is calculated.

[0012] Step 5: Calculate the true resistance of the insulating material between the positive and negative electrodes and the battery cover based on the detected insulation resistance of the positive and negative electrodes and the resistance of the dust. Calculate the actual leakage and real-time leakage based on the true resistance and battery voltage. Compare the leakage with the leakage threshold to determine the leakage situation and take appropriate measures.

[0013] Furthermore, the top cover insulation detection circuit includes a pulse signal source U s , detection resistor R f , current limiting resistor R1, current limiting resistor R2, positive pole, negative pole, the pulse signal source U s It includes a pulse positive output terminal and a pulse negative output terminal. The pulse positive output terminal is electrically connected to the battery top cover, and the pulse negative output terminal is electrically connected to the detection resistor R f One end is electrically connected to the detection resistor R f One end of is electrically connected to one end of the current limiting resistor R1 and the current limiting resistor R2 respectively, the other end of the current limiting resistor R1 is electrically connected to the positive electrode column, and the other end of the current limiting resistor R2 is electrically connected to the negative electrode column.

[0014] Furthermore, the equivalent insulation circuit model includes the positive column insulation resistance R p , Negative column insulation resistance R n And the top cover insulation detection circuit, the positive column insulation resistance R p The insulation resistance R between one end of the battery and the top cover and the negative pole n Electrical connection, positive column insulation resistance R p The other end is electrically connected to the positive pole, and the negative pole insulation resistance R n The other end is electrically connected to the negative electrode.

[0015] Furthermore, the specific steps of calculating the insulation resistance of the positive and negative poles are:

[0016] When the voltage applied by the pulse signal during the period is positive, the voltage across the detection resistor can be calculated according to Kirchhoff's law. The calculation formula is:

[0017] I f+ =I 1+ +I 2+

[0018] I f+ =I 3+ +I 4+

[0019] U=R n *I 4+ -R P *I 3+

[0020] U=R1*I 1+ -R2*I2 +

[0021] U s+ =R p *I 3+ +R1*I 1+ +R f *I f+

[0022] Among them, U s+ is the positive voltage applied by the pulse signal during the period, I 1+ , I 2+ , I 3+ , I 4+ , I f+ They are respectively the current limiting resistor R1, the current limiting resistor R2, and the positive column insulation resistance R when the pulse signal applies a positive voltage within the cycle. p , Negative column insulation resistance R n and the detection resistor R f U is the voltage across the positive and negative poles of the power battery;

[0023] Since R1=R2=R, according to the above formula, the detection resistor R can be calculated f The voltage across the two ends, the voltage calculation formula is:

[0024]

[0025] Among them, U f+ When the positive voltage is applied by the pulse signal during the period, the detection resistor R f The voltage across the terminals;

[0026] When the voltage applied by the pulse signal during the cycle is a negative voltage;

[0027] The detection resistor R can be calculatedf The voltage across the two ends, the voltage calculation formula is:

[0028]

[0029] Among them, U f- When the negative voltage is applied by the pulse signal during the period, the detection resistor R f The voltage across the terminals;

[0030] Positive column insulation resistance R p , Negative column insulation resistance R n The calculation formula is:

[0031]

[0032] Among them, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative pole.

[0033] Furthermore, the pole parameters include pole material, pole area, and pole temperature;

[0034] The working environment parameters include temperature, humidity, and particle concentration.

[0035] Furthermore, the calculation formula of the sliding window method is:

[0036]

[0037] Among them, X is the time series data set after segmentation, xn na is the nth influencing factor in the time series data set, ti is the current moment, and nc is the width of the sliding window.

[0038] Furthermore, the dust deposition model is based on a long short-term memory neural network, which includes an input gate, a forget gate, and an output gate;

[0039] The equation of the forget gate is:

[0040] f t =δ(W f ·[h t-1 , x t ]+b f )

[0041] Among them, f t is the current switch state of the forget gate, δ represents the sigmoid function, W f is the forget gate weight, b f is the forget gate bias parameter, h t-1 is the hidden state of the previous moment, x t Input switch for current data;

[0042] Determine how much of the unit state at the previous moment needs to be retained to the current moment;

[0043] The equation for the input gate is:

[0044] i t =δ(W i ·[h t-1 , x t ]+b i )

[0045] Among them, i t is the current switch state of the input gate, δ represents the sigmoid function, W i is the input gate weight, b i is the input gate bias parameter, h t-1 is the output hidden state of the previous moment, x t is the time series dataset after segmentation at the current moment;

[0046]

[0047] in, Candidate cell state, tanh represents the hyperbolic tangent function, W c is the candidate cell weight, b c is the candidate cell state bias parameter, h t-1 is the hidden state of the previous moment, x t Input switch for current data;

[0048] Determine how much of the network's input data needs to be saved to the unit state at the current moment;

[0049] The update equation is:

[0050]

[0051] Among them, C t Current cell state, C t-1 is the cell state at the previous moment;

[0052] The output gate equation is:

[0053] o t =(W o ·[h t-1 , x t ]+b o )

[0054] h t =o t *tanh(C t )

[0055] Among them, t is the current switch state of the output gate, δ represents the sigmoid function, W o is the output gate weight, b o is the output gate bias parameter, h t-1 is the hidden state of the previous moment, x t For the current data input switch, h t is the dust thickness of the positive and negative poles at the next moment;

[0056] Controls how much of the current unit state needs to be output to the current output value;

[0057] Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are:

[0058]

[0059] Among them, δ(x γ ) is the sigmoid function, x γ For the sigmoid function input, tanh(C t ) is the hyperbolic tangent function, C t Input to the function.

[0060] Furthermore, the specific steps of calculating the resistance value of dust according to the influence of temperature and humidity on the conductivity of dust are as follows:

[0061] The effect of humidity on the electrical conductivity of dust is calculated as follows:

[0062] σ s =σ0*(1+k s *Sd)

[0063] Among them, σ s is the conductivity of dust under the influence of humidity, σ0 is the conductivity of dust at standard temperature and humidity, k s is the humidity influence coefficient, Sd is the humidity of dust;

[0064] The effect of temperature on the electrical conductivity of dust is calculated as follows:

[0065]

[0066] Among them, σ w is the conductivity of dust under the influence of temperature, σ0 is the conductivity of dust under standard temperature and humidity, k w is the humidity influence coefficient, T h is the temperature of the dust;

[0067] The resistance value of dust is calculated as follows:

[0068]

[0069] Among them, R h is the resistance value of dust, H h is the dust thickness, S j It is the area covered by dust on the pole.

[0070] Furthermore, the calculation formula for the actual resistance of the positive and negative poles is:

[0071] R zp =R p -R h

[0072] R zn =R n -R h

[0073] Among them, R zp is the actual insulation resistance of the positive column, R zn is the actual insulation resistance of the negative pole, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative column, R h is the resistance value of dust;

[0074] The calculation formulas for the actual leakage and real-time leakage are:

[0075]

[0076] Among them, I sj is the real-time leakage of the battery cover, I sh is the actual leakage of the battery top cover, and U is the voltage across the positive and negative poles of the power battery;

[0077] When I sj When ≥α, it is judged that the dust on the battery cover has a great impact on the leakage of the cover, and the dust should be cleaned immediately;

[0078] When I sh When ≥β, it is judged that the battery cover leakage is too large and the cover should be repaired immediately;

[0079] Among them, α and β are the real-time leakage of the battery top cover and the safety threshold of the actual leakage of the battery top cover, respectively.

[0080] The present invention further provides a power battery pack top cover leakage safety detection system, which is used to execute the above-mentioned power battery pack top cover leakage safety detection method, including:

[0081] An insulation detection module is configured to apply a pulse signal source's positive terminal to the top cover of each battery in the battery pack to be detected, connect the negative terminal of the pulse signal source to a detection resistor, and electrically connect the detection resistor to the positive and negative electrodes of the battery after being connected in series with a current-limiting resistor, thereby forming a top cover insulation detection circuit;

[0082] A circuit equivalent module is used to equate the insulating material between the power battery top cover and the positive and negative poles to the insulation resistance of the positive and negative poles, and to construct an equivalent insulation circuit model with the top cover insulation detection circuit. The positive and negative voltages applied by the pulse signal source during the signal cycle and the detection voltage of the detection circuit during the cycle are compared to calculate the insulation resistance of the positive and negative poles according to Kirchhoff's law.

[0083] A dust prediction module is used to obtain historical operating environment parameters and pole parameters of different battery packs, stamp them with timestamps, form a time series data set, segment the time series data set using a sliding window method, establish a dust deposition model, use the segmented time series data set at the current moment as the model input, and use the dust thickness of the positive and negative poles at the next moment as the model output to train the model;

[0084] A dust resistance calculation module is used to process the current battery environmental parameters and electrode parameters using a sliding window method as input to a trained model. The model outputs the dust thickness at the positive and negative electrodes. Based on the real-time battery environmental parameters and electrode parameters, and the effects of temperature and humidity on the conductivity of the dust, the dust resistance is calculated.

[0085] The leakage detection module is used to calculate the true resistance of the insulating material between the positive and negative poles and the battery top cover based on the detected positive pole insulation resistance, negative pole insulation resistance and dust resistance. Based on the true resistance and battery voltage, the module calculates the actual leakage amount and the real-time leakage amount, compares them with the leakage threshold, determines the leakage situation, and takes measures.

[0086] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention forms a top cover insulation detection circuit by applying a pulse signal source and a detection resistor to the top cover of each battery in the battery pack to be detected, and equates the insulating material between the power battery top cover and the positive and negative poles to the insulation resistance of the positive and negative poles, constructs an equivalent insulation circuit model, calculates the insulation resistance of the positive and negative poles, establishes a dust precipitation model and trains the model, and uses the current environmental parameters and pole parameters of the battery after processing by a sliding window method as the input of the trained model. The model outputs the dust thickness of the positive and negative poles, and calculates the dust resistance value based on the real-time environmental parameters and pole parameters of the battery. The actual leakage amount and the real-time leakage amount are calculated based on the detected positive pole insulation resistance, negative pole insulation resistance and dust resistance value, and the battery voltage;

[0087] The present invention is based on; constructing an equivalent insulation circuit model, calculating the insulation resistance of the positive and negative poles, and calculating the insulation resistance by injecting the reflected wave voltage of the pulse signal source. It can be independent of the power battery voltage and can also be measured when the power battery is disconnected or the voltage is very low. It can detect the insulation resistance in real time and accurately, and has high reliability and system compatibility. By establishing a dust precipitation model, the dust thickness of the positive and negative poles is output through the model, and the dust resistance is calculated in combination with temperature and humidity. It can accurately predict the dust thickness and consider the impact of the environment on the resistance, which helps to provide more accurate leakage detection. According to the detected positive pole insulation resistance, negative pole insulation resistance and dust resistance value, and battery voltage, the actual leakage amount and real-time leakage amount are calculated, which can truly reflect the insulation performance of the battery top cover and the real-time and actual battery leakage amount changes, and use this as a basis to adjust the charging and discharging strategy and take necessary protective measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0089] Figure 2 Schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0090] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0091] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0092] Example:

[0093] See also Figure 1 , the present invention provides a technical solution:

[0094] A power battery pack top cover leakage safety detection method, the specific steps include:

[0095] Step 1: Apply the positive end of a pulse signal source to the top cover of each battery in the battery pack to be tested, connect the negative end of the pulse signal source to a detection resistor, and then connect the detection resistor in series with a current-limiting resistor and electrically connect it to the positive and negative poles of the battery to form a top cover insulation detection circuit.

[0096] Thermal aging and cracking of the insulation material between the positive and negative electrodes and the battery cover are primarily caused by changes in weather and geographic location, such as heat and humidity cycles, temperature shocks, salt spray corrosion, and altitude. High summer temperatures cause a sharp rise in power battery temperatures, while low winter temperatures and changes in altitude also create significant temperature differences. These temperature differences cause the insulating polymer to degrade in high-heat environments and shrink, becoming hard and brittle in low temperatures, accelerating insulation aging and ultimately leading to leakage between the positive and negative electrodes and the battery cover.

[0097] In this embodiment, the top cover insulation detection circuit includes a pulse signal source U s , detection resistor R f , current limiting resistor R1, current limiting resistor R2, positive pole, negative pole, the pulse signal source U s It includes a pulse positive output terminal and a pulse negative output terminal. The pulse positive output terminal is electrically connected to the battery top cover, and the pulse negative output terminal is electrically connected to the detection resistor R f One end is electrically connected to the detection resistor R f One end of is electrically connected to one end of the current limiting resistor R1 and the current limiting resistor R2 respectively, the other end of the current limiting resistor R1 is electrically connected to the positive electrode column, and the other end of the current limiting resistor R2 is electrically connected to the negative electrode column.

[0098] Step 2: Equivalently treat the insulating material between the power battery top cover and the positive and negative poles as the insulation resistance of the positive and negative poles, and construct an equivalent insulation circuit model with the top cover insulation detection circuit. The positive and negative voltages applied by the pulse signal source during the signal cycle and the detection voltage of the detection circuit during the cycle are compared to calculate the insulation resistance of the positive and negative poles according to Kirchhoff's law.

[0099] In this embodiment, the equivalent insulation circuit model includes the positive column insulation resistance R p , Negative column insulation resistance R n And the top cover insulation detection circuit, the positive column insulation resistance R p The insulation resistance R between one end of the battery and the top cover and the negative pole n Electrical connection, positive column insulation resistance R p The other end is electrically connected to the positive pole, and the negative pole insulation resistance R n The other end is electrically connected to the negative electrode.

[0100] The equivalent insulation circuit model injects a low-frequency, low-voltage detection signal of a certain amplitude and frequency into the power battery through the battery top cover. The insulation resistance of the positive and negative poles is calculated based on the reflected wave voltage signal on the detection resistor. The specific measurement process is as follows: a pulse signal source generates a low-frequency detection signal, which is injected into the power battery through the battery top cover. The signal flows through the positive and negative insulation resistors, returns to the top cover insulation detection circuit, and then flows back to the sampling resistor through a current-limiting resistor. At this point, a reflected wave voltage generated by the pulse signal source is generated on the sampling resistor. The magnitude of the reflected wave voltage is used to calculate the insulation resistance of the positive and negative poles.

[0101] The equivalent insulation circuit model can simultaneously calculate the insulation resistance of the positive and negative poles, effectively solving the problem of low detection accuracy caused by simultaneous insulation faults in the positive and negative poles. This method calculates the insulation resistance by injecting the reflected wave voltage of the pulse signal source. It does not rely on the power battery voltage and can also be measured when the power battery is disconnected or the voltage is very low. Therefore, it has certain advantages in reliability and system compatibility.

[0102] Kirchhoff's laws are two basic laws in circuit analysis, used to describe the distribution of current and voltage in a circuit. They include Kirchhoff's current law and Kirchhoff's voltage law. In Kirchhoff's current law, at any node in a circuit, the sum of the currents entering the node is equal to the sum of the currents leaving the node. In Kirchhoff's voltage law, in any closed circuit, the sum of the voltages along the circuit path (loop) is zero, that is, the voltage increases and decreases are equal.

[0103] In this embodiment, the specific steps of calculating the insulation resistance of the positive and negative poles are:

[0104] When the voltage applied by the pulse signal during the period is positive, the voltage across the detection resistor can be calculated according to Kirchhoff's law. The calculation formula is:

[0105] I f+ =I 1+ +I 2+

[0106] I f+ =I 3+ +I 4+

[0107] U=R n *I 4+ -R P *I 3+

[0108] U=R1*I 1+ -R2*I 2+

[0109] U s+ =Rp *I 3+ +R1*I 1+ +R f *I f+

[0110] Among them, U s+ is the positive voltage applied by the pulse signal during the period, I 1+ , I 2+ , I 3+ , I 4+ , I f+ They are respectively the current limiting resistor R1, the current limiting resistor R2, and the positive column insulation resistance R when the pulse signal applies a positive voltage within the cycle. p , Negative column insulation resistance R n and the detection resistor R f U is the voltage across the positive and negative poles of the power battery;

[0111] Since R1=R2=R, according to the above formula, the detection resistor R can be calculated f The voltage across the two ends, the voltage calculation formula is:

[0112]

[0113] Among them, U f+ When the positive voltage is applied by the pulse signal during the period, the detection resistor R f The voltage across the terminals;

[0114] When the voltage applied by the pulse signal during the cycle is a negative voltage;

[0115] The detection resistor R can be calculated f The voltage across the two ends, the voltage calculation formula is:

[0116]

[0117] Among them, U f- When the negative voltage is applied by the pulse signal during the period, the detection resistor R f The voltage across the terminals;

[0118] Positive column insulation resistance R p , Negative column insulation resistance R n The calculation formula is:

[0119]

[0120] Among them, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative pole.

[0121] Step 3: Obtain the historical working environment parameters and pole parameters of different battery packs and stamp them with timestamps to form a time series dataset. Use the sliding window method to segment the time series dataset and establish a dust deposition model. Use the segmented time series dataset at the current moment as the input of the model and the dust thickness of the positive and negative poles at the next moment as the output of the model to train the model.

[0122] In this embodiment, the pole parameters include pole material, pole area, and pole temperature;

[0123] The working environment parameters include temperature, humidity, and particle concentration.

[0124] The material of the battery terminal determines the interaction between the terminal surface and dust particles. Battery terminals made of different materials have different characteristics for dust adsorption, adhesion, and accumulation. The terminal area directly affects the total surface area for dust adhesion, and thus the amount of dust accumulation. A larger surface area means more dust adhesion, while a smaller surface area reduces the total amount of dust adhesion. The terminal temperature is a very important physical factor that directly affects the interaction between the terminal surface and the surrounding moisture and dust particles.

[0125] Temperature affects the mobility and settling rate of dust particles. Higher temperatures can increase the number of dust particles in the air or affect their adhesion. Humidity changes affect the adsorption and deposition characteristics of dust particles. Excessive humidity can make dust particles more likely to clump and adhere. Particle concentration is a factor that directly affects the rate of dust deposition: higher dust concentrations result in a greater amount of dust deposited per unit time.

[0126] In this embodiment, the calculation formula of the sliding window method is:

[0127]

[0128] Among them, X is the time series data set after segmentation, xn na is the nth influencing factor in the time series data set, ti is the current moment, and nc is the width of the sliding window;

[0129] The sliding window method divides time series data into multiple windows of fixed length, with the data within each window serving as an input sample. For example, using the most recent 30 days of data to predict dust deposition thickness on the 31st day, each window contains environmental parameters and polar parameters from multiple time periods, which serve as model inputs.

[0130] The sliding window method effectively preserves historical information and models the temporal dependencies of data using a long-short-term memory neural network. The dust thickness on the battery pole is affected not only by current environmental and pole parameters, but also by those from previous periods. The sliding window method extracts this dependency, enabling more accurate predictions based on historical data. Especially in dynamic environments, the model can continuously adjust its predictions of future dust thickness to adapt to changing environmental trends.

[0131] Long Short-Term Memory (LSTM) neural networks are a special type of recurrent neural network that can effectively learn and remember temporal dependencies over long time spans. The thickness of dust accumulation is affected not only by current environmental parameters but also by environmental changes over time. For example, changes in temperature and humidity can affect the rate and amount of dust accumulation. LSTM neural networks can capture these long-term and short-term dependencies in a time series, enabling more accurate predictions. The environment surrounding battery terminals, such as temperature, humidity, and dust concentration, often changes dynamically. LSTM neural networks can use historical data to identify patterns in how these environmental factors change over time, thereby predicting dust accumulation trends.

[0132] Predicting dust accumulation requires more than simple linear relationships; complex nonlinear relationships exist between environmental parameters like temperature, humidity, and dust concentration, as well as between polar parameters. Long-short-term memory neural networks, with their deep nonlinear structure and learning capabilities, can handle these complex relationships, offering advantages over traditional linear regression or simple neural networks.

[0133] In this embodiment, the dust deposition model is based on a long short-term memory neural network, which includes an input gate, a forget gate, and an output gate;

[0134] The equation of the forget gate is:

[0135] f t =δ(W f ·[h t-1 , x t ]+b f )

[0136] Among them, f t is the current switch state of the forget gate, δ represents the sigmoid function, W f is the forget gate weight, b f is the forget gate bias parameter, h t-1 is the hidden state of the previous moment, x t Input switch for current data;

[0137] Determine how much of the unit state at the previous moment needs to be retained to the current moment;

[0138] The equation for the input gate is:

[0139] i t =δ(W i ·[h t-1 , x t ]+b i )

[0140] Among them, i t is the current switch state of the input gate, δ represents the sigmoid function, W i is the input gate weight, b i is the input gate bias parameter, h t-1 is the output hidden state of the previous moment, x t is the time series dataset after segmentation at the current moment;

[0141]

[0142] in, Candidate cell state, tanh represents the hyperbolic tangent function, W c is the candidate cell weight, b c is the candidate cell state bias parameter, h t-1 is the hidden state of the previous moment, x t Input switch for current data;

[0143] Determine how much of the network's input data needs to be saved to the unit state at the current moment;

[0144] The update equation is:

[0145]

[0146] Among them, C t Current cell state, C t-1 is the cell state at the previous moment;

[0147] The output gate equation is:

[0148] o t =(W o ·[h t-1 , x t ]+b o )

[0149] h t =o t *tanh(C t )

[0150] Among them, t is the current switch state of the output gate, δ represents the sigmoid function, W o is the output gate weight, bo is the output gate bias parameter, h t-1 is the hidden state of the previous moment, x t For the current data input switch, h t is the dust thickness of the positive and negative poles at the next moment;

[0151] Controls how much of the current unit state needs to be output to the current output value;

[0152] Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are:

[0153]

[0154] Among them, δ(x γ ) is the sigmoid function, x γ For the sigmoid function input, tanh(C t ) is the hyperbolic tangent function, C t Input to the function.

[0155] Step 4: The current battery environmental parameters and electrode parameters are processed using the sliding window method and used as the input of the trained model. The model outputs the dust thickness at the positive and negative electrodes. Based on the real-time environmental parameters and electrode parameters of the battery, the resistance of the dust is calculated by considering the effects of temperature and humidity on the conductivity of the dust.

[0156] Dust accumulation and changes in its conductivity can be a precursor to battery pack leakage. By predicting dust thickness and then calculating resistance based on temperature and humidity data, we can accurately detect the impact of dust on leakage. This method, by predicting dust thickness and calculating resistance based on temperature and humidity conditions, allows the detection method to adapt to various environmental conditions, improving the system's versatility and reliability.

[0157] The conductivity of dust is affected by humidity, especially dust containing salt or metal particles. In a humid environment, moisture may form conductive pathways between dust particles, significantly increasing the conductivity.

[0158] In this embodiment, the specific steps of calculating the resistance value of dust according to the effects of temperature and humidity on the conductivity of dust are as follows:

[0159] The effect of humidity on the electrical conductivity of dust is calculated as follows:

[0160] σ s =σ0*(1+k s *Sd)

[0161] Among them, σ s is the conductivity of dust under the influence of humidity, σ0 is the conductivity of dust at standard temperature and humidity, ks is the humidity influence coefficient, and Sd is the humidity of the dust.

[0162] Temperature changes can also affect the conductivity of dust. Generally, an increase in temperature causes a material to become more conductive.

[0163] The effect of temperature on the electrical conductivity of dust is calculated as follows:

[0164]

[0165] Among them, σ w is the conductivity of dust under the influence of temperature, σ0 is the conductivity of dust under standard temperature and humidity, k w is the humidity influence coefficient, T h is the temperature of the dust;

[0166] The resistance value of dust is calculated as follows:

[0167]

[0168] Among them, R h is the resistance value of dust, H h is the dust thickness, S j It is the area covered by dust on the pole.

[0169] Step 5: Calculate the true resistance of the insulating material between the positive and negative electrodes and the battery cover based on the detected insulation resistance of the positive and negative electrodes and the resistance of the dust. Calculate the actual leakage and real-time leakage based on the true resistance and battery voltage. Compare the leakage with the leakage threshold to determine the leakage situation and take appropriate measures.

[0170] By calculating the actual resistance value and combining it with the battery voltage and the resistance value of dust, the battery leakage can be monitored in real time. The real-time leakage calculation can accurately reflect the leakage situation under different working conditions such as temperature, humidity, dust accumulation and other factors.

[0171] The actual resistance value calculated by measuring the insulation resistance of the positive and negative terminals and combining it with the battery voltage reflects the actual leakage of the power battery and is not affected by operating conditions. It is caused by the aging of the insulation material between the positive and negative terminals and the top cover.

[0172] Based on the monitoring of real-time leakage and actual leakage, the battery status can be analyzed more intelligently, and the charging and discharging strategies can be adjusted according to the battery leakage situation, and necessary protective measures can be taken.

[0173] In this embodiment, the calculation formula for the actual resistance of the positive and negative poles is:

[0174] R zp =R p-R h

[0175] R zn =R n -R h

[0176] Among them, R zp is the actual insulation resistance of the positive column, R zn is the actual insulation resistance of the negative pole, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative column, R h is the resistance value of dust;

[0177] The calculation formulas for the actual leakage and real-time leakage are:

[0178]

[0179] Among them, I sj is the real-time leakage of the battery cover, I sh is the actual leakage of the battery top cover, and U is the voltage across the positive and negative poles of the power battery;

[0180] When I sj When ≥α, it is judged that the dust on the battery cover has a great impact on the leakage of the cover, and the dust should be cleaned immediately;

[0181] When I sh When ≥β, it is judged that the battery cover leakage is too large and the cover should be repaired immediately;

[0182] Among them, α and β are the real-time leakage of the battery top cover and the safety threshold of the actual leakage of the battery top cover, respectively.

[0183] See also Figure 2 The present invention further provides a power battery pack top cover leakage safety detection system, which is used to execute the above-mentioned power battery pack top cover leakage safety detection method, including:

[0184] An insulation detection module is configured to apply a pulse signal source's positive terminal to the top cover of each battery in the battery pack to be detected, connect the negative terminal of the pulse signal source to a detection resistor, and electrically connect the detection resistor to the positive and negative electrodes of the battery after being connected in series with a current-limiting resistor, thereby forming a top cover insulation detection circuit;

[0185] A circuit equivalent module is used to equate the insulating material between the power battery top cover and the positive and negative poles to the insulation resistance of the positive and negative poles, and to construct an equivalent insulation circuit model with the top cover insulation detection circuit. The positive and negative voltages applied by the pulse signal source during the signal cycle and the detection voltage of the detection circuit during the cycle are compared to calculate the insulation resistance of the positive and negative poles according to Kirchhoff's law.

[0186] A dust prediction module is used to obtain historical operating environment parameters and pole parameters of different battery packs, stamp them with timestamps, form a time series data set, segment the time series data set using a sliding window method, establish a dust deposition model, use the segmented time series data set at the current moment as the model input, and use the dust thickness of the positive and negative poles at the next moment as the model output to train the model;

[0187] A dust resistance calculation module is used to process the current battery environmental parameters and electrode parameters using a sliding window method as input to a trained model. The model outputs the dust thickness at the positive and negative electrodes. Based on the real-time battery environmental parameters and electrode parameters, and the effects of temperature and humidity on the conductivity of the dust, the dust resistance is calculated.

[0188] The leakage detection module is used to calculate the true resistance of the insulating material between the positive and negative poles and the battery top cover based on the detected positive pole insulation resistance, negative pole insulation resistance and dust resistance. Based on the true resistance and battery voltage, the module calculates the actual leakage amount and the real-time leakage amount, compares them with the leakage threshold, determines the leakage situation, and takes measures.

[0189] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0190] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0191] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0192] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A power battery pack top cover leakage safety detection method, characterized in that: The specific steps include: Step 1: Apply the positive end of the pulse signal source to the top cover of each battery in the battery pack to be tested, connect the negative end of the pulse signal source to the detection resistor, and the detection resistor is electrically connected to the positive and negative poles of the battery after being connected in series with the current limiting resistor to form a top cover insulation detection circuit; Step 2: The insulating material between the power battery top cover and the positive and negative poles is equivalent to the insulation resistance of the positive and negative poles, and an equivalent insulation circuit model is constructed with the top cover insulation detection circuit. The positive and negative voltages applied by the pulse signal source in the signal cycle and the detection voltage of the detection circuit in the cycle are compared, and the insulation resistance of the positive and negative poles is calculated according to Kirchhoff's law; Step 3: Obtain the historical working environment parameters and pole parameters of different battery packs and stamp them with timestamps to form a time series data set. Use the sliding window method to segment the time series data set, establish a dust precipitation model, use the segmented time series data set at the current moment as the input of the model, and use the dust thickness of the positive and negative poles at the next moment as the output of the model to train the model; Step 4: The current environmental parameters and pole parameters of the battery are processed by the sliding window method as the input of the trained model. The model outputs the dust thickness at the positive and negative electrodes. According to the real-time environmental parameters and pole parameters of the battery, the resistance value of the dust is calculated by the influence of temperature and humidity on the conductivity of the dust. Step 5: Calculate the true resistance of the insulating material between the positive and negative poles and the battery top cover based on the detected insulation resistance of the positive pole, the negative pole and the resistance of the dust. Calculate the actual leakage amount and the real-time leakage amount based on the true resistance and the battery voltage. Compare them with the leakage threshold to determine the leakage situation and take measures.

2. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The top cover insulation detection circuit includes a pulse signal source U s , detection resistor R f , current limiting resistor R1, current limiting resistor R2, positive pole, negative pole, the pulse signal source U s It includes a pulse positive output terminal and a pulse negative output terminal, the pulse positive output terminal is electrically connected to the battery top cover, and the pulse negative output terminal is electrically connected to the detection resistor R f One end of the detection resistor R f One end of is electrically connected to one end of the current limiting resistor R1 and the current limiting resistor R2 respectively, the other end of the current limiting resistor R1 is electrically connected to the positive electrode column, and the other end of the current limiting resistor R2 is electrically connected to the negative electrode column.

3. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The equivalent insulation circuit model includes the positive pole insulation resistance R p , Negative pole insulation resistance R n And the top cover insulation detection circuit, the positive pole insulation resistance R p One end of the battery is connected to the top cover and the negative pole by the insulation resistance R n Electrical connection, positive pole insulation resistance R p The other end is electrically connected to the positive pole, and the negative pole insulation resistance R n The other end is electrically connected to the negative electrode.

4. A power battery pack top cover leakage safety detection method according to claim 2, characterized in that: The specific steps of calculating the insulation resistance of the positive and negative poles are: When the voltage applied by the pulse signal during the period is a positive voltage, the voltage across the detection resistor can be calculated according to Kirchhoff's law. The calculation formula is: I f+ =I 1+ +I 2+ I f+ =I 3+ +I 4+ U=R n *I 4+ -R P *I 3+ U=R1*I 1+ -R2*I 2+ U s+ =R p *I 3+ +R1*I 1+ +R f *I f+ Among them, U s+ is the positive voltage applied by the pulse signal during the period, I 1+ ,I 2+ ,I 3+ ,I 4+ ,I f+ They are respectively the current limiting resistor R1, the current limiting resistor R2, and the positive column insulation resistance R when the pulse signal applies a positive voltage within the cycle. p , Negative pole insulation resistance R n and the detection resistor R f U is the voltage across the positive and negative poles of the power battery; Since R1=R2=R, according to the above formula, the detection resistor R can be calculated f The voltage across the two ends, the voltage calculation formula is: Among them, U f+ When the positive voltage is applied by the pulse signal during the period, the detection resistor R f The voltage across the terminals; When the voltage applied by the pulse signal during the cycle is a negative voltage; The detection resistor R can be calculated f The voltage across the two ends, the voltage calculation formula is: Among them, U f- When the pulse signal applies a negative voltage during the cycle, the detection resistor R f The voltage across the terminals; Positive column insulation resistance R p , Negative pole insulation resistance R n The calculation formula is: Among them, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative pole.

5. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The pole parameters include pole material, pole area, and pole temperature; The working environment parameters include temperature, humidity, and particle concentration.

6. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The calculation formula of the sliding window method is: Among them, X is the time series data set after segmentation, xn na is the nath influencing factor in the time series data set, ti is the current moment, and nc is the width of the sliding window.

7. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The dust precipitation model is based on a long short-term memory neural network, which includes an input gate, a forget gate and an output gate; The equation of the forget gate is: f t =δ(W f ·[h t-1 ,x t ]+b f ) Among them, f t is the current switch state of the forget gate, δ represents the sigmoid function, W f is the forget gate weight, b f is the forget gate bias parameter, h t-1 is the hidden state at the previous moment, x t Input switch for current data; Determine how much of the unit state at the previous moment needs to be retained to the current moment; The equation for the input gate is: i t =δ(W i ·[h t-1 ,x t ]+b i ) Among them, i t is the current switch state of the input gate, δ represents the sigmoid function, W i is the input gate weight, b i is the input gate bias parameter, h t-1 is the output hidden state of the previous moment, x t is the time series data set after segmentation at the current moment; in, Candidate cell state, tanh represents the hyperbolic tangent function, W c is the candidate cell weight, b c is the candidate cell state bias parameter, h t-1 is the hidden state at the previous moment, x t Input switch for current data; Determine how much of the network's input data needs to be saved to the unit state at the current moment; The update equation is: Among them, C t Current cell state, C t-1 is the cell state at the previous moment; The output gate equation is: o t =(W o ·[h t-1 ,x t ]+b o ) h t =o t *tanh(C t ) Among them, t is the current switch state of the output gate, δ represents the sigmoid function, W o is the output gate weight, b o is the output gate bias parameter, h t-1 is the hidden state at the previous moment, x t is the current data input switch, h t is the dust thickness of the positive and negative poles at the next moment; Controls how much of the current unit state needs to be output to the current output value; Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are: Among them, δ(x γ ) is the sigmoid function, x γ is the input of sigmoid function, tanh(C t ) is the hyperbolic tangent function, C t Input to the function.

8. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The specific steps of calculating the resistance value of dust according to the influence of temperature and humidity on the conductivity of dust are as follows: The effect of humidity on the conductivity of dust is calculated as follows: s s =σ0*(1+k s *Sd) Among them, σ s is the conductivity of dust under the influence of humidity, σ0 is the conductivity of dust at standard temperature and humidity, k s is the humidity influence coefficient, Sd is the humidity of dust; The effect of temperature on the conductivity of dust is calculated as follows: Among them, σ w is the conductivity of dust under the influence of temperature, σ0 is the conductivity of dust under standard temperature and humidity, k w is the humidity influence coefficient, T h is the temperature of the dust; The resistance value of dust is calculated as: Among them, R h is the resistance value of dust, H h is the dust thickness, S j It is the area covered by dust on the pole.

9. A power battery pack top cover leakage safety detection method according to claim 1, characterized in that: The calculation formula for the actual resistance of the positive and negative poles is: R zp =R p -R h R zn =R n -R h Among them, R zp is the actual resistance of the positive column insulation resistance, R zn is the actual insulation resistance of the negative pole, R p is the insulation resistance of the positive column, R n is the insulation resistance of the negative pole, R h is the resistance value of dust; The calculation formulas for the actual leakage and real-time leakage are: Among them, I sj is the real-time leakage of the battery cover, I sh is the actual leakage of the battery top cover, and U is the voltage at both ends of the positive and negative poles of the power battery; When I sj When ≥α, it is judged that the dust on the battery cover has too great an impact on the leakage of the cover, and the dust should be cleaned immediately; When θ sh When ≥β, it is judged that the battery cover leakage is too large and the cover should be repaired immediately; Among them, α and β are the real-time leakage of the battery top cover and the safety threshold of the actual leakage of the battery top cover, respectively.

10. A power battery pack top cover leakage safety detection system, characterized in that: The power battery pack top cover leakage safety detection system is used to execute a power battery pack top cover leakage safety detection method according to any one of claims 1 to 9, comprising: An insulation detection module, wherein the insulation detection module is used to apply the positive end of a pulse signal source to the top cover of each battery in the battery pack to be detected, the negative end of the pulse signal source is connected to a detection resistor, and the detection resistor is respectively connected in series with a current limiting resistor and electrically connected to the positive and negative poles of the battery to form a top cover insulation detection circuit; A circuit equivalent module, which is used to equate the insulating material between the power battery top cover and the positive and negative poles to the insulation resistance of the positive and negative poles, and to construct an equivalent insulation circuit model with the top cover insulation detection circuit, and to calculate the insulation resistance of the positive and negative poles according to Kirchhoff's law by comparing the positive and negative voltages applied by the pulse signal source within the signal cycle with the detection voltage of the detection circuit during the cycle. A dust prediction module is used to obtain the historical working environment parameters of different battery packs and the pole parameters and print the timestamps to form a time series data set, perform time series segmentation on the time series data set by the sliding window method, establish a dust precipitation model, use the time series data set after segmentation at the current moment as the input of the model, and use the dust thickness of the positive and negative poles at the next moment as the output of the model to train the model; A dust resistance calculation module, which is used to process the current environmental parameters and pole parameters of the battery through a sliding window method as inputs of a trained model, output the dust thickness of the positive and negative electrodes through the model, and calculate the resistance value of the dust according to the real-time environmental parameters and pole parameters of the battery and the influence of temperature and humidity on the conductivity of the dust respectively; The leakage detection module is used to calculate the real resistance of the insulating material between the positive and negative poles and the battery top cover according to the detected positive pole insulation resistance, negative pole insulation resistance and dust resistance, calculate the actual leakage amount and the real-time leakage amount according to the real resistance and battery voltage, compare with the leakage threshold to determine the leakage situation, and take measures.

Images