A data processing method and apparatus

Abstract

The embodiment of the present application provides a kind of data processing method and device, the method comprises: obtaining the power battery data of electric bus;Process the power battery data, obtain the multiple dimensional data of the power battery in the charging state and in the non-charging state;Based on 3σ criterion, process the multiple dimensional data, determine abnormal value;The target feature data is obtained by carrying out PCA dimension reduction processing to the abnormal value;SVM weighted processing is carried out to the target feature data, to determine whether the target feature data is target type data, the target type data indicates that the dimensional data corresponding to the target feature data needs to be rejected.The data processing method and device provided by the present application can exclude unnecessary power battery data, so that the power battery data after simplification can be more quickly, efficiently and accurately determined whether to carry out battery safety early warning.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of computer, in particular to a data processing method and device. BACKGROUND

[0002] In the past two years, the safety problem of new energy vehicles in China has attracted widespread attention. Related research shows that the new energy vehicle industry in China generally lacks understanding of safety, and the safety of product design is not enough. The safety interaction mechanism in the whole chain has not been formed, which has led to the frequent occurrence of electric vehicle safety accidents in recent period. Many pure electric bus fire accidents have a negative impact on the development of the industry. The safety warning research of pure electric bus power battery has important significance.

[0003] The existing pure electric bus power battery safety warning method usually gives a threshold value according to the battery manufacturer for warning, but when calculating the state value of the battery to determine whether to issue a warning, the parameters involved in the calculation are not verified, so that the calculation result has a large error. SUMMARY

[0004] The present application provides a data processing method and device which can exclude unnecessary power battery data, so that the power battery data after simplification can be used to determine whether to perform battery safety warning more quickly, efficiently and accurately.

[0005] In order to solve the above technical problems, the embodiment of the present application provides a data processing method, comprising:

[0006] Obtain power battery data of an electric bus;

[0007] Process the power battery data to obtain multiple dimension data of the power battery in a charging state and in a non-charging state;

[0008] Determine an abnormal value based on 3 sigma criterion processing the multiple dimension data;

[0009] Perform PCA dimension reduction processing on the abnormal value to obtain target feature data;

[0010] Perform SVM weighted processing on the target feature data to determine whether the target feature data is target type data, wherein the target type data represents that the dimension data corresponding to the target feature data needs to be eliminated.

[0011] As an optional embodiment, it further comprises:

[0012] Determine target dimension data corresponding to the target feature data;

[0013] Eliminate the target dimension data, and calculate and determine the state of the power battery based on the remaining dimension data.

[0014] determining whether to perform a safety warning based on the state of the power battery.

[0015] As an optional embodiment, the processing the power battery data to obtain the plurality of dimension data of the power battery in the non-charging state comprises:

[0016] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0017] slicing the first voltage data to form a plurality of cell data;

[0018] calculating, for each of the cell data, a maximum value, a minimum value, a cell upper limit, a cell lower limit, a cell range, a coefficient of variation, a cell lower quartile, a cell middle quartile, a cell upper quartile, an upper level endpoint, and a lower level endpoint.

[0019] As an optional embodiment, the processing the power battery data to obtain the plurality of dimension data of the power battery in the non-charging state comprises:

[0020] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0021] slicing the first voltage data to form a plurality of cell data, each of the cell data corresponding to a complete charging cycle;

[0022] calculating a plurality of differences of the cell data that meet a time sequence;

[0023] determining, based on the plurality of differences, a maximum value, a minimum value, a cell upper limit, a cell lower limit, a cell range, a coefficient of variation, a cell lower quartile, a cell middle quartile, a cell upper quartile, an upper level endpoint, and a lower level endpoint.

[0024] As an optional embodiment, the processing the plurality of dimension data based on the 3σ criterion to determine an abnormal value comprises:

[0025] determining a standard deviation and a mean value of the plurality of dimension data;

[0026] determining a detection threshold based on the standard deviation and the mean value;

[0027] determining the abnormal value based on at least the detection threshold and the plurality of dimension data.

[0028] As an optional embodiment, the processing the abnormal value by PCA dimension reduction to obtain target feature data comprises:

[0029] When each dimension data has the abnormal value, the plurality of abnormal values are processed by PCA dimension reduction to obtain the target feature data meeting the feature requirement, and the target feature data is feature data of part of the first abnormal values in the plurality of abnormal values.

[0030] As an optional embodiment, the SVM weighting processing on the target feature data is performed to determine whether the target feature data is target type data, including:

[0031] The SVM weighting processing is performed on the target feature data to determine the category of each target feature data.

[0032] It is determined whether the target feature data is target type data based on the category.

[0033] Another embodiment of the application provides a data processing device, including:

[0034] An obtaining module is configured to obtain power battery data of an electric bus.

[0035] A first processing module is configured to process the power battery data to obtain a plurality of dimension data of the power battery in a charging state and in a non-charging state.

[0036] A second processing module is configured to process the plurality of dimension data according to a 3σ criterion to determine abnormal values.

[0037] A third processing module is configured to perform PCA dimension reduction processing on the abnormal values to obtain target feature data.

[0038] A fourth processing module is configured to perform SVM weighting processing on the target feature data to determine whether the target feature data is target type data, and the target type data represents that dimension data corresponding to the target feature data needs to be removed.

[0039] As an optional embodiment, the application further includes:

[0040] A determining module is configured to determine target dimension data corresponding to the target feature data.

[0041] A removing module is configured to remove the target dimension data, and determine the state of the power battery based on the remaining dimension data.

[0042] A warning module is configured to determine whether to perform a safety warning according to the state of the power battery.

[0043] As an optional embodiment, the processing of the power battery data to obtain the plurality of dimension data of the power battery in the non-charging state includes:

[0044] determine first voltage data of the SOC in the power battery data meeting a target threshold range of value range;

[0045] slice the first voltage data to form a plurality of single data;

[0046] calculate maximum value, minimum value, single upper limit, single lower limit, single range, coefficient of variation, single lower quartile, single middle quartile, single upper quartile, upper level endpoint, and lower level endpoint for each single data.

[0047] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art upon examination of the following or can be learned by practice of the application. The objects and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0048] The technical solutions of the present application will be further described in detail below with the help of the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings incorporated in and forming a part of the specification, illustrate several aspects of the present application, and together with the description serve to explain the principles of the application. In the drawings:

[0050] Figure 1 The flow chart of the data processing method in the embodiment of the present application.

[0051] Figure 2 The actual application flow chart of the data processing method in the embodiment of the present application.

[0052] Figure 3 The actual application flow chart of the data processing method in another embodiment of the present application.

[0053] Figure 4 The flow chart of the data processing device in the embodiment of the present application. DETAILED DESCRIPTION

[0054] The specific embodiments of the present application will be described in detail below with the help of the accompanying drawings, but not as a limitation of the present application.

[0055] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered to be limiting, but merely an example of embodiments. Other modifications within the scope and spirit of the disclosure will occur to those skilled in the art.

[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and, together with the general description of the disclosure given above, and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0057] These and other characteristics of the present application will become apparent from the following description of the preferred forms given, by way of non-limiting example, with reference to the attached drawings.

[0058] It is also to be understood that even though a number of embodiments of the application have been described herein, the application covers all

[0059] The above and other aspects, features, and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which:

[0060] Specific embodiments of the present disclosure will be described hereinafter with reference to the accompanying drawings; however, it will be understood that the disclosed embodiments are merely examples of the present disclosure, which can be implemented in numerous ways. Well-known and / or repetitive functions and structures are not described in detail to avoid obscuring the present disclosure unnecessarily. Therefore, specific structural and functional details disclosed herein are not intended to limit the present disclosure, but merely to serve as a representative basis for teaching one skilled in the art to employ the present disclosure in virtually any appropriate detailed structure.

[0061] The specification can use phrases such as "in one embodiment", "in another embodiment", "in yet another embodiment", or "in other embodiments", which can refer to one or more of the same or different embodiments under the present disclosure.

[0062] Hereinafter, embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0063] As shown in Figure 1 and Figure 2 The embodiment of the present application provides a data processing method, comprising:

[0064] Obtaining power battery data of an electric bus;

[0065] Processing the power battery data to obtain multiple dimension data of the power battery in a charging state and in a non-charging state;

[0066] Processing the multiple dimension data based on a 3σ criterion to determine an abnormal value;

[0067] Performing PCA dimension reduction processing on the abnormal value to obtain target feature data;

[0068] The target feature data is subjected to SVM weighting processing to determine whether the target feature data is target type data, the target type data representing that the dimension data corresponding to the target feature data needs to be eliminated.

[0069] Further, the method in the embodiment further includes:

[0070] determining target dimension data corresponding to the target feature data;

[0071] eliminating the target dimension data, and calculating and determining the state of the power battery based on the remaining dimension data.

[0072] determining whether to perform safety warning based on the state of the power battery.

[0073] The method in the embodiment can be applied to a safety warning model for performing safety warning on a power battery, so as to optimize the model and enable the model to perform safety warning on the power battery based on more concise and effective data. The method in the embodiment performs data processing, takes abnormal values, reduces dimensions, and performs weighting processing, and the data processing adopts normal distribution of a mathematical model, and the processing forms multiple dimensions of data. When 3Sigma is taken, 3Sigma values of multiple dimensions of data are adopted, and the abnormal values taken by 3Sigma are subjected to PCA dimension reduction and SVM weighting, so as to simplify the data of the power battery and remove invalid data. This not only reduces the data processing load when calculating and determining whether to output safety warning, but also ensures the scientificity and accuracy of the safety warning method. In addition, the method provided in the embodiment can discover and eliminate safety hazards of the power battery as early as possible, accelerate the improvement of safety standards of new energy vehicles and power batteries, and improve the quality and safety level of the whole life cycle of new energy vehicles.

[0074] Specifically, the embodiment processes the power battery data to obtain multiple dimensions of data of the power battery in a non-charged state, including:

[0075] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0076] performing slice processing on the first voltage data to form multiple single data;

[0077] calculating a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint for each single data.

[0078] The embodiment processes the power battery data to obtain multiple dimensions of data of the power battery in a charged state, including:

[0079] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0080] slicing the first voltage data based on one complete charging cycle to form a plurality of single data, each of the single data corresponding to one complete charging cycle;

[0081] calculating a difference value of a plurality of the single data meeting a time sequence;

[0082] determining a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint based on a plurality of the difference values.

[0083] For example, the data processing in the embodiment is mainly to facilitate data analysis and assist in determining abnormal values such as missing values and null values in the data. The abnormal value of the single voltage (to be described in detail below) can be judged by the following threshold value: the single voltage of the lithium iron phosphate battery ∈(0, 4.5), the single voltage of the lithium manganese battery ∈(0, 5), and the single voltage exceeding the above threshold value is considered to be an abnormal value.

[0084] When the power battery is in a non-charging state, first voltage data in which a value range of SOC is between 40% and 70% is extracted from the power battery data, and then the first voltage data is sliced to obtain a plurality of single voltage data. For each single voltage data, a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint are calculated, totaling 11 dimensions of data.

[0085] When the power battery is in a charging state, data in which a value range of SOC is between 40% and 70% is extracted from the power battery data, and then slicing is performed. Specifically, slicing can be performed according to the time dimension, with each charging cycle as a slicing standard, to obtain a plurality of single voltage values. Then, each single voltage value is processed by Δ (difference value processing), including subtracting the previous single voltage value from the next single voltage value to obtain a plurality of Δ values, and so on. The maximum value, the minimum value, the single upper limit, the single lower limit, the single range, the coefficient of variation, the single lower quartile, the single middle quartile, the single upper quartile, the upper level endpoint, and the lower level endpoint are calculated and processed by concatenating the plurality of slicing results, i.e., the plurality of Δ values, totaling 11 dimensions of data.

[0086] where max is the maximum, min is the minimum, Q1 is the upper quartile, Q2 is the median, Q3 is the lower quartile, Upper is the upper cell limit, Lower is the lower cell limit, Range is the cell range, Variable is the coefficient of variation, exUp is the upper extreme, exLow is the lower extreme, IQR is the interquartile range, prctile is the percentile function, σ is the standard deviation percentile function, μ is the mean. The specific formulas are as follows:

[0087] The upper quartile formula is as follows: Q1 = prctile (cell voltage array, for example, 25)

[0088] The median quartile formula is as follows: Q2 = prctile (cell voltage array, for example, 50)

[0089] The lower quartile formula is as follows: Q3 = prctile (cell voltage array, for example, 75)

[0090] The interquartile range formula is as follows: IQR = Q3 - Q1

[0091] The upper cell limit formula is as follows: Upper = Q3 + 1.5 * IQR;

[0092] The lower cell limit formula is as follows: Lower = Q3 - 1.5 * IQR;

[0093] The upper extreme formula is as follows: exUp = Q3 + 3 * IQR;

[0094] The lower extreme formula is as follows: exLow = Q3 - 3 * IQR;

[0095] The cell range formula is as follows: Range = max - min;

[0096] The coefficient of variation formula is as follows: Variable = σ / μ;

[0097] The △ processing formula is as follows: Δy n = y n+1 - y n

[0098] Further, when the plurality of dimensional data is processed based on the 3σ criterion to determine the abnormal value, comprising:

[0099] Determine the standard deviation and mean of the plurality of dimensional data;

[0100] Determine the detection threshold based on the standard deviation and mean;

[0101] Determine the abnormal value based on at least the detection threshold and the plurality of dimensional data.

[0102] For example, the standard deviation and mean of the plurality of dimensional data are calculated, σ represents the standard deviation, μ represents the mean, and x = μ is the symmetry axis of the image; the detection threshold is determined based on the standard deviation and the mean.

[0103] The 3σ principle includes one or more of the following detection thresholds:

[0104] P(μ-σ≤x≤μ+σ)≈0.6826

[0105] P(μ-2σ≤x≤μ+2σ)≈0.9544

[0106] P(μ-3σ≤x≤μ+3σ)≈0.9974

[0107] The third detection threshold is selected for subsequent calculation in this embodiment, and it is considered in this embodiment that the value of the monomer voltage Y is almost concentrated in the interval (μ-3σ, μ+3σ), and the Y value exceeding this range is an abnormal value.

[0108] Next, the number of each monomer voltage value (hereinafter referred to as monomer) exceeding the upper extreme point and the number exceeding the lower extreme point can be calculated;

[0109] The number of each monomer exceeding the upper limit and the number exceeding the lower limit is calculated;

[0110] The proportion p1 of the number of each monomer exceeding the upper extreme point to the number exceeding the upper limit and the proportion p2 of the number of each monomer exceeding the lower extreme point to the number exceeding the lower limit are calculated, and the proportion is assigned a value of 0 when the proportion is NaN; The average values mp1 and mp2 of all p1 and p2 are calculated.

[0111] The average values are processed, and the deviation degree deviate = (p-mp) / mp of p1 and p2 of each monomer relative to the average value mp is calculated, and the deviation degree is assigned a value of 0 when the deviation degree is NaN. Based on the above calculation steps and the selected detection threshold, the abnormal values in each dimensional data can be determined.

[0112] Further, the abnormal values are subjected to PCA dimensionality reduction processing to obtain target feature data, comprising:

[0113] When each dimensional data has the abnormal value, the plurality of abnormal values are subjected to PCA dimensionality reduction processing to obtain the target feature data satisfying the feature requirement, and the target feature data is the feature data of part of the first abnormal values in the plurality of abnormal values.

[0114] For example, in the embodiment, there are m pieces of n-dimensional data, the m pieces of n-dimensional data are grouped into an n-row m-column matrix X according to columns, each row (representing an attribute field) of X is zero-meaned, that is, the mean value of the row is subtracted. A covariance matrix is calculated, eigenvalues and corresponding eigenvectors r of the covariance matrix are calculated. The eigenvectors are arranged in a matrix from top to bottom according to the corresponding eigenvalue size, and the first k rows are taken to form a matrix P, that is, the data after dimension reduction to k dimensions, and the following formula is used:

[0115]

[0116] error represents an error value after compression and dimension reduction, where m is the number of all features, and then a threshold x is preset, for example, 0.01, the threshold range is indefinite, and can be changed according to actual conditions. A k is selected, so that error < x, that is, only error < x, the value of k is qualified, and the dimension reduction to k dimensions is qualified. In the embodiment, 11-dimensional data with the most characteristics can be finally obtained from the 22-dimensional data processed by the PCA algorithm, that is, k is 11.

[0117] Further, the target feature data is subjected to SVM weighting processing to determine whether the target feature data is target type data, comprising:

[0118] The target feature data is subjected to SVM weighting processing to determine the category of each target feature data.

[0119] Based on the category, it is determined whether the target feature data is target type data.

[0120] For example, in the weighted support vector machine, a penalty parameter C is determined, the penalty parameter C is selected for each sample, and the following formula can be combined:

[0121]

[0122]

[0123] ξ i ≥0, i = 1, 2,..., l

[0124] Wherein, si is the weighting coefficient of the i-th training sample to C. The Lagrange multiplier method is used to solve the quadratic programming problem with linear inequalities, that is:

[0125]

[0126] s.t. 0 ≤ α i

[0127] 0 ≤ β i

[0128] The dual is obtained as follows:

[0129]

[0130] s.t.0≤α i ≤Cs i

[0131]

[0132] The discriminant function formula is obtained by calculating the above quadratic programming problem:

[0133]

[0134] Wherein b is:

[0135]

[0136] Through the above PCA algorithm, the 11 most characteristic dimensions of data obtained can be finally classified to determine whether they have a significant impact on the safety warning of the power battery.

[0137] Optionally, as shown in Figure 3 the method in the embodiment can further include:

[0138] After the PCA algorithm dimension reduction, the reduced data can be classified by the SVM algorithm and the random forest, and then the classified data can be weighted by feature fusion to further optimize the safety warning model.

[0139] As shown in Figure 4 the application also provides a data processing device, which comprises:

[0140] An obtaining module is configured to obtain power battery data of an electric bus;

[0141] A first processing module is configured to process the power battery data to obtain multiple dimension data of the power battery in a charging state and in a non-charging state;

[0142] A second processing module is configured to process the multiple dimension data according to a 3σ rule to determine an abnormal value;

[0143] A third processing module is configured to perform PCA dimension reduction processing on the abnormal value to obtain target feature data;

[0144] A fourth processing module is configured to perform SVM weighting processing on the target feature data to determine whether the target feature data is target type data, wherein the target type data represents that the dimension data corresponding to the target feature data needs to be removed.

[0145] As an optional embodiment, further comprising:

[0146] a determining module configured to determine target dimension data corresponding to the target feature data;

[0147] a culling module configured to cull the target dimension data, and calculate the state of the power battery based on the remaining dimension data;

[0148] a pre-warning module configured to determine whether to perform a safety pre-warning according to the state of the power battery.

[0149] As an optional embodiment, processing the power battery data to obtain a plurality of dimension data of the power battery in a non-charging state comprises:

[0150] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0151] performing slice processing on the first voltage data to form a plurality of single data;

[0152] calculating a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint for each single data.

[0153] As an optional embodiment, processing the power battery data to obtain a plurality of dimension data of the power battery in a charging state comprises:

[0154] determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range;

[0155] performing slice processing on the first voltage data based on a complete charging cycle to form a plurality of single data, each single data corresponding to a complete charging cycle;

[0156] calculating a difference value of a plurality of single data meeting a time sequence;

[0157] calculating a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint based on a plurality of difference values.

[0158] As an optional embodiment, processing the plurality of dimension data based on the 3σ criterion to determine an abnormal value comprises:

[0159] determining a standard deviation and a mean value of the plurality of dimension data;

[0160] determining a detection threshold based on the standard deviation and the mean value;

[0161] determine the abnormal value based on the detection threshold and the plurality of dimension data.

[0162] As an optional embodiment, the PCA dimension reduction processing is performed on the abnormal value to obtain target feature data, comprising:

[0163] When each dimension data has the abnormal value, the PCA dimension reduction processing is performed on a plurality of abnormal values to obtain the target feature data meeting the feature requirement, and the target feature data is feature data of part of the first abnormal value in the plurality of abnormal values.

[0164] As an optional embodiment, the SVM weighting processing is performed on the target feature data to determine whether the target feature data is target type data, comprising:

[0165] The SVM weighting processing is performed on the target feature data to determine the category of each target feature data.

[0166] It is determined whether the target feature data is target type data based on the category.

[0167] Another embodiment of the present application further provides an electronic device, comprising:

[0168] one or more processors;

[0169] a memory configured to store one or more programs;

[0170] When the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0171] An embodiment of the present application further provides a storage medium having a computer program stored thereon, and the program is executed by a processor to implement the method described above. It should be understood that each scheme in the embodiment has the corresponding technical effects in the above method embodiments, which will not be repeated here.

[0172] An embodiment of the present application further provides a computer program product, which is tangibly stored on a computer readable medium and includes computer readable instructions that, when executed, cause at least one processor to perform a method such as in the above embodiments. It should be understood that each scheme in the embodiment has the corresponding technical effects in the above method embodiments, which will not be repeated here.

[0173] It should be noted that the computer storage media of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable medium may, for example, but not limited to, be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of computer readable storage media can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a carrier wave in a propagated data signal, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium that can send, propagate or transmit the program configured for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, cable, RF, etc. or any suitable combination of the above.

[0174] It should be understood that although the present application is described in terms of various embodiments, not every embodiment contains only one independent technical solution, and the description of the specification is only for the sake of clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that those skilled in the art can understand.

[0175] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements to the present application within the spirit and protection scope of the present application, and such modifications or equivalent replacements should also be considered to fall within the protection scope of the present application.

Claims

1. A data processing method, characterized by, The method comprises the following steps: obtaining power battery data of an electric bus; processing the power battery data to obtain multiple dimension data of the power battery in a charging state and in a non-charging state; determining abnormal values based on 3σ criterion processing of the multiple dimension data; performing PCA dimension reduction processing on the abnormal values to obtain target feature data; performing SVM weighting processing on the target feature data to determine whether the target feature data is target type data, which represents that the dimension data corresponding to the target feature data needs to be excluded; the processing of the power battery data to obtain multiple dimension data of the power battery in a non-charging state comprises: determining first voltage data in which the value range of SOC in the power battery data meets a target threshold range; performing slicing processing on the first voltage data to form multiple single data; calculating the maximum value, the minimum value, the single upper limit, the single lower limit, the single range, the coefficient of variation, the single lower quartile, the single middle quartile, the single upper quartile, the upper level endpoint, and the lower level endpoint for each single data.

2. The method of claim 1, wherein, Further comprising: determining target dimension data corresponding to the target feature data; excluding the target dimension data, and determining the state of the power battery based on the remaining dimension data; determining whether to perform a safety warning based on the state of the power battery.

3. The method of claim 1, wherein, The processing of the power battery data to obtain multiple dimension data of the power battery in a charging state comprises: determining first voltage data in which the value range of SOC in the power battery data meets a target threshold range; performing slicing processing on the first voltage data based on a complete charging cycle to form multiple single data, each single data corresponding to a complete charging cycle; calculating the difference value of multiple single data that meet the time sequence; determining the maximum value, the minimum value, the single upper limit, the single lower limit, the single range, the coefficient of variation, the single lower quartile, the single middle quartile, the single upper quartile, the upper level endpoint, and the lower level endpoint based on multiple difference values.

4. The method of claim 1, wherein, The determination of abnormal values based on 3σ criterion processing of the multiple dimension data comprises: determining the standard deviation and the mean value of the multiple dimension data; determining a detection threshold based on the standard deviation and the mean value; determining the abnormal values based on at least the detection threshold and the multiple dimension data.

5. The method of claim 1, wherein, The PCA dimension reduction processing of the abnormal values to obtain target feature data comprises: when each dimension data has the abnormal value, performing PCA dimension reduction processing on multiple abnormal values to obtain target feature data that meets the feature requirement, the target feature data being feature data of part of the first abnormal values in the multiple abnormal values.

6. The method of claim 1, wherein, The SVM weighting processing of the target feature data to determine whether the target feature data is target type data comprises: performing SVM weighting processing on the target feature data to determine the category of each target feature data; determining whether the target feature data is target type data based on the category.

7. A data processing apparatus, characterized by, The method comprises the following steps: an obtaining module is configured to obtain power battery data of an electric bus; The first processing module is configured to process the power battery data to obtain multiple dimension data of the power battery in a charging state and in a non-charging state; The second processing module is configured to process the multiple dimension data according to a 3σ criterion to determine an abnormal value; The third processing module is configured to perform PCA dimension reduction processing on the abnormal value to obtain target feature data; The fourth processing module is configured to perform SVM weighting processing on the target feature data to determine whether the target feature data is target type data, the target type data indicating that dimension data corresponding to the target feature data needs to be removed; The processing of the power battery data to obtain the multiple dimension data of the power battery in the non-charging state comprises: determining first voltage data in which a value range of SOC in the power battery data meets a target threshold range; performing slice processing on the first voltage data to form multiple single data; calculating a maximum value, a minimum value, a single upper limit, a single lower limit, a single range, a coefficient of variation, a single lower quartile, a single middle quartile, a single upper quartile, an upper level endpoint, and a lower level endpoint for each single data.

8. The data processing apparatus according to claim 7, characterized in that, Further comprising: a determination module configured to determine target dimension data corresponding to the target feature data; a removal module configured to remove the target dimension data and determine a state of the power battery based on remaining dimension data; an early warning module configured to determine whether to perform a safety early warning according to the state of the power battery.

Images