A new energy vehicle charging pile intelligent operation management method and system
By performing stability calculation and differential processing on the temperature and charging data of the charging pile, an ARIMA prediction model is built, which solves the problem of data instability under the influence of environmental factors, and improves the accuracy and user experience of charging pile data prediction.
Patent Information
- Application Number
- CN202410164041.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-02-05
AI Technical Summary
When the prior art uses differential autoregressive sliding average model (ARIMA) to predict charging pile data, the influence of environmental factors leads to instability of data, resulting in low accuracy of prediction results and affecting user interaction experience.
By obtaining the temperature dimension data sequence and the charging dimension data sequence, calculating its stability and performing differential processing, obtaining the stable temperature-charging data sequence, calculating the temperature-charging optimization correlation coefficient, and building an ARIMA prediction model to predict the latest charging of the charging pile data.
It improves the accuracy of charging pile data prediction, reduces the impact of environmental data changes on prediction results, and improves user interaction experience and the health management effect of charging piles.
Smart Images

Figure CN118082580B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault prediction data processing, and in particular to a smart operation and management method and system for a new energy vehicle charging pile. Background Art
[0002] Fault prediction technology is one of the key technologies to ensure the stable operation of charging piles. By collecting and analyzing the operating data of charging piles to predict potential faults, this prediction not only helps to reduce the unexpected downtime of charging piles, but also reduces maintenance costs. Combined with fault prediction, health management is crucial to the long-term stability of charging piles. By continuously monitoring and evaluating the health of charging piles, more refined maintenance strategies can be implemented; by combining the charging data of charging piles with environmental data for data prediction, users can be provided with a more intelligent and efficient charging experience, which also helps to improve the reliability and efficiency of charging pile equipment and reduce operating costs and environmental impacts.
[0003] When using the Autoregressive Imaging Average (ARIMA) algorithm to predict the latest charging data of a charging pile, since the data needs to be stable when using the ARIMA algorithm for prediction processing, and the prediction results are often affected by environmental factors, if only a single-dimensional charging data is constructed and analyzed using an ARIMA prediction model, the prediction results will be affected by environmental factors, resulting in errors, resulting in low accuracy in predicting the latest charging data of the charging pile, affecting the interactive experience when interacting with users based on the prediction results. Summary of the invention
[0004] In order to solve the above problems, the present invention provides a new energy vehicle charging pile intelligent operation and management method and system.
[0005] An embodiment of the present invention provides a method for intelligent operation and management of a new energy vehicle charging pile, the method comprising the following steps:
[0006] Acquire a temperature dimension data sequence and a charging dimension data sequence of a charging pile; the temperature dimension data sequence includes temperature data segments within a plurality of sampling time ranges, and the charging dimension data sequence includes charging data segments within a plurality of sampling time ranges; the temperature data segment includes a plurality of temperature data, and the charging data segment includes a plurality of charging data; within the same sampling time range, one temperature data segment corresponds to one charging data segment;
[0007] According to the distribution of temperature data in the temperature data segment of each sampling time range in the temperature dimension data sequence, the stability of the temperature dimension data sequence is obtained; according to the distribution of charging data in the charging data segment of each sampling time range in the charging dimension data sequence, the stability of the charging dimension data sequence is obtained; according to the stability, the temperature dimension data sequence and the charging dimension data sequence are differentially processed to obtain a stabilized temperature-charging data sequence, wherein the stabilized temperature-charging data sequence includes temperature data segments and charging data segments of several sampling time ranges;
[0008] According to the temperature data segment and charging data segment of each sampling time range in the stabilized temperature-charging data sequence, the temperature-charging optimization correlation coefficient of each sampling time range is obtained; according to the temperature-charging optimization correlation coefficient, the ARIMA prediction model is obtained, and the latest charging data of the charging pile is predicted according to the ARIMA prediction model.
[0009] Preferably, the stability of the temperature dimension data sequence is obtained according to the distribution of the temperature data in the temperature data segment of each sampling time range in the temperature dimension data sequence, and the specific method includes:
[0010] According to the difference between each temperature data and the adjacent temperature data in the temperature data segment of each sampling time range, the difference factor of each temperature data in the temperature data segment of each sampling time range is obtained; according to the discrete situation of all temperature data in the temperature data segment of each sampling time range, the stable weight of the temperature data segment of each sampling time range is obtained;
[0011] According to the difference factor of each temperature data in the temperature data segment of each sampling time range and the stability weight of the temperature data segment of each sampling time range, the calculation method for obtaining the stability of the temperature dimension data sequence is:
[0012]
[0013] In the formula, Staw represents the stability of the temperature dimension data sequence; U represents the total number of temperature data segments in all sampling time ranges in the temperature dimension data sequence; I represents the total number of all temperature data in the temperature data segment of each sampling time range; CD u,i Represents the difference factor of the u-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence; var u It represents the stable weight of the temperature data segment in the u-th sampling time range in the temperature dimension data sequence; exp() represents an exponential function with a natural constant as the base; || represents taking the absolute value.
[0014] Preferably, the specific method of obtaining the difference factor of each temperature data in the temperature data segment of each sampling time range according to the difference between each temperature data in the temperature data segment of each sampling time range and the adjacent temperature data includes:
[0015] A hyperparameter ε is preset, and the difference between the i-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence and the i-1-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence is recorded as the first difference; the difference between the i-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence and the i+1-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence is recorded as the second difference, and the sum of the second difference and the hyperparameter ε is recorded as the first sum value; the ratio of the first difference to the first sum value is recorded as the first ratio, and the difference between 1 and the first ratio is used as the difference factor of the i-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence.
[0016] Preferably, the method of obtaining the stable weight of the temperature data segment of each sampling time range according to the discrete conditions of all temperature data in the temperature data segment of each sampling time range includes the following specific methods:
[0017] The variance of all temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence is used as the stable weight of the temperature data segment of the u-th sampling time range.
[0018] Preferably, the step of performing differential processing on the temperature dimension data sequence and the charging dimension data sequence according to stability to obtain a stabilized temperature-charging data sequence includes the following specific methods:
[0019] A stability parameter W is preset. If the stability of the temperature dimension data sequence is less than the stability parameter W, the temperature dimension data sequence is differentially processed until the stability of the temperature dimension data sequence is greater than or equal to the stability parameter W, and a stabilized temperature dimension data sequence is obtained. If the stability of the charging dimension data sequence is less than the stability parameter W, the charging dimension data sequence is differentially processed until the stability of the charging dimension data sequence is greater than or equal to the stability parameter W, and a stabilized charging dimension data sequence is obtained. The stabilized charging dimension data sequence and the stabilized temperature dimension data sequence are combined to form a stabilized temperature-charging data sequence.
[0020] Preferably, the method of obtaining the temperature-charging optimization correlation coefficient of each sampling time range according to the temperature data segment and the charging data segment of each sampling time range in the stabilized temperature-charging data sequence includes the following specific methods:
[0021] For the kth sampling time range, the autocorrelation coefficient between the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range, the autocorrelation coefficient between the charging data segment of the kth sampling time range and the charging data segment of the first sampling time range, and the partial autocorrelation coefficient between the temperature data segment of the kth sampling time range and the charging data segment of the kth sampling time range are obtained by using the ARIMA algorithm;
[0022] According to the autocorrelation coefficient of the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range, the optimization weight of the kth sampling time range is obtained; according to the partial autocorrelation coefficient of the temperature data segment of the kth sampling time range and the charging data segment of the kth sampling time range and the optimization weight of the kth sampling time range, the optimization factor of the kth sampling time range is obtained; according to the optimization factor of the kth sampling time range, the optimization degree of the kth sampling time range is obtained;
[0023] The product of the autocorrelation coefficient of the charging data segment of the kth sampling time range and the charging data segment of the first sampling time range and the optimization degree of the kth sampling time range is used as the temperature-charging optimization correlation coefficient of the kth sampling time range.
[0024] Preferably, the step of obtaining the optimization weight of the kth sampling time range according to the autocorrelation coefficient of the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range includes the following specific methods:
[0025] The absolute value of the autocorrelation coefficient between the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range is used as the optimization weight of the kth sampling time range.
[0026] Preferably, the specific formula for obtaining the optimization factor of the kth sampling time range according to the partial autocorrelation coefficient of the temperature data segment of the kth sampling time range and the charging data segment of the kth sampling time range and the optimization weight of the kth sampling time range is:
[0027]
[0028] In the formula, YH k represents the optimization factor of the kth sampling time range; I' represents the total number of all sampling time ranges in the stable temperature-charging data sequence; Q k represents the optimization weight of the kth sampling time range; Zfg j Represents the partial autocorrelation coefficient between the temperature data segment in the j-th sampling time range and the charging data segment in the j-th sampling time range; || represents the absolute value.
[0029] Preferably, the step of obtaining the optimization degree of the kth sampling time range according to the optimization factor of the kth sampling time range includes the following specific methods:
[0030] After all optimization factors are linearly normalized, each optimization factor is recorded as the degree of preference.
[0031] The present invention also proposes an intelligent operation and management system for new energy vehicle charging piles, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of the intelligent operation and management method for new energy vehicle charging piles.
[0032] The beneficial effects of the technical solution of the present invention are: the present invention performs differential processing on the temperature dimension data sequence and the charging dimension data sequence according to stability to obtain a stabilized temperature-charging data sequence, so that the data distribution is suitable for the ARIMA algorithm, ensuring that the data is stable data; according to the temperature data segment and the charging data segment of each sampling time range in the stabilized temperature-charging data sequence, the temperature-charging optimization correlation coefficient of each sampling time range is obtained; according to the temperature-charging optimization correlation coefficient, the ARIMA prediction model is obtained, and the latest charging data of the charging pile is predicted according to the ARIMA prediction model, thereby constructing an ARIMA prediction model based on the temperature data and the charging data, and obtaining the prediction results accompanied by temperature data, which solves the problem of inaccurate prediction results caused by changes in environmental data, thereby improving the accuracy of predicting the latest charging data of the charging pile. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0034] Figure 1 A flowchart of the steps of a new energy vehicle charging pile intelligent operation and management method of the present invention;
[0035] Figure 2 The process of obtaining the ARIMA forecast model. DETAILED DESCRIPTION
[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the intelligent operation and management method and system of a new energy vehicle charging pile proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0038] The specific scheme of the intelligent operation and management method and system of new energy vehicle charging piles provided by the present invention is described in detail below with reference to the accompanying drawings.
[0039] See also Figure 1 , which shows a flowchart of a method for intelligent operation and management of new energy vehicle charging piles provided by an embodiment of the present invention, the method comprising the following steps:
[0040] Step S001: Obtain the temperature dimension data sequence and charging dimension data sequence of the charging pile.
[0041] It should be noted that when using the ARIMA algorithm to predict the charging data of the user's charging pile, the charging current or voltage is unstable due to the fluctuations in the ambient temperature of the charging pile. As a result, the algorithm cannot obtain the most accurate ARIMA prediction data, resulting in a deviation between the prediction result and the true value, and cannot effectively achieve the prediction effect. This in turn leads to low accuracy of the fault prediction results and inability to effectively manage the health of the charging pile.
[0042] Specifically, it is necessary to first collect the temperature dimension data sequence and charging dimension data sequence of the charging pile. The specific process is as follows:
[0043] The temperature data and charging data of the charging pile are collected every 10 minutes, and the collection is carried out for 4 hours. 4 hours is used as a sampling time range, and the data sequences formed by all the temperature data and charging data within a sampling time range are recorded as temperature data segments and charging data segments respectively; a total of 24 hours are collected to obtain the temperature data segments and charging data segments of several sampling time ranges of the charging pile; the temperature data segments of all sampling time ranges of the charging pile are combined into a data sequence as the temperature dimension data sequence of the charging pile; the charging data segments of all sampling time ranges of the charging pile are combined into a data sequence as the charging dimension data sequence of the charging pile.
[0044] Among them, within the same sampling time range, one temperature data segment corresponds to one charging data segment; in this embodiment, current data or voltage data can be used as charging data, and this embodiment does not make specific limitations; current data can be collected through the current sensor of the charging pile, voltage data can be collected through the voltage sensor of the charging pile, and temperature data can be collected through the temperature sensor of the charging pile.
[0045] So far, the temperature dimension data sequence and charging dimension data sequence of the charging pile are obtained through the above method.
[0046] Step S002: According to the stability of the temperature dimension data sequence and the charging dimension data sequence, a stabilized temperature-charging data sequence is obtained.
[0047] It should be noted that the environment also has a certain impact on the charging data segment. For example, during the day, the ambient temperature is higher, which is more conducive to improving the efficiency of the charging pile. On the contrary, at night, the ambient temperature is lower and the efficiency of the charging pile is lower than during the day, so a larger charging data segment is required. In order to achieve a higher user interaction effect, it is necessary to predict the charging data of the charging pile and feed the prediction results back to the user to improve the interactive experience. When using the ARIMA algorithm for prediction processing, in order to make the collected data distribution suitable for the model, it is necessary to ensure that each data segment is stable data.
[0048] It should be further explained that the higher the data stability, the more conducive it is to the predictive analysis of the ARIMA algorithm, and the less necessary it is to perform differentiation; corresponding to the scenario, the higher the data stability, that is, the closer the charging data of different users are, the more conducive it is to the prediction of subsequent charging data, and the less necessary it is to perform differentiation, because differentiation is for data that does not meet the stability requirements; the closer the data values are in different time periods, the higher their stability; the closer the difference between the data values, the higher their stability; when the difference is high, it needs to be differentiated.
[0049] Specifically, for the temperature dimension data sequence, the stability of the temperature dimension data sequence is obtained according to the distribution of temperature data in the temperature data segment of each sampling time range in the temperature dimension data sequence.
[0050] As an example, the variance of all temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence is used as the stability weight of the temperature data segment of the u-th sampling time range. The calculation method for obtaining the stability of the temperature dimension data sequence is:
[0051]
[0052] In the formula, Staw represents the stability of the temperature dimension data sequence; U represents the total number of temperature data segments in all sampling time ranges in the temperature dimension data sequence; I represents the total number of all temperature data in the temperature data segment of each sampling time range; g u,i represents the i-th temperature data in the temperature data segment of the u-th sampling time range in the temperature dimension data sequence; g u,i-1 Indicates the i-1th temperature data in the temperature data segment of the uth sampling time range in the temperature dimension data sequence; g u,i+1 Represents the i+1th temperature data in the temperature data segment of the uth sampling time range in the temperature dimension data sequence; var u It represents the stable weight of the temperature data segment in the u-th sampling time range in the temperature dimension data sequence; ε represents a preset hyperparameter. In this implementation, ε=1 is preset to prevent the denominator from being 0; exp() represents an exponential function with a natural constant as the base. The implementation example adopts the exp(-x) model to present the inverse proportional relationship. x is the input of the model. The implementer can select the inverse proportional function according to the actual situation; || represents taking the absolute value.
[0053] It should be noted that var u Represents the stable weight of the temperature data segment of the u-th sampling time range in the temperature dimension data sequence, which is used to quantify the difference between all temperature data in the temperature data segment of the u-th sampling time range; It represents the difference factor of the ith temperature data in the temperature data segment of the uth sampling time range. The smaller the value, the more stable the temperature data segment. In the process of calculating the stability of the temperature dimension data sequence, when i=1, the zero-filling operation is used to make g u,0 When i=I, the zero-filling operation makes g u,i+1 is 0.
[0054] Further, referring to the method for obtaining the stability of the temperature dimension data sequence, the temperature dimension data sequence is replaced with the charging dimension data sequence to obtain the stability of the charging dimension data sequence.
[0055] A stability parameter W is preset, wherein this embodiment is described by taking W=0.8 as an example, and this embodiment is not specifically limited, wherein W is determined according to specific implementation conditions.
[0056] Specifically, if the stability of the temperature dimension data sequence is less than the stability parameter W, the temperature dimension data sequence is differentially processed until the stability of the temperature dimension data sequence is greater than or equal to the stability parameter W, and a stabilized temperature dimension data sequence is obtained; if the stability of the charging dimension data sequence is less than the stability parameter W, the charging dimension data sequence is differentially processed until the stability of the charging dimension data sequence is greater than or equal to the stability parameter W, and a stabilized charging dimension data sequence is obtained; the stabilized charging dimension data sequence and the stabilized temperature dimension data sequence are combined to form a stabilized temperature-charging data sequence; the stabilized temperature-charging data sequence includes temperature data segments and charging data segments within several sampling time ranges.
[0057] The difference processing is a well-known content of the ARIMA algorithm, and will not be described in detail in this embodiment.
[0058] At this point, a stabilized temperature-charging data sequence is obtained through the above method.
[0059] Step S003: According to the temperature-charging optimization correlation coefficient of each sampling time range in the stabilized temperature-charging data sequence, an ARIMA prediction model is obtained to predict the latest charging data of the charging pile.
[0060] It should be noted that temperature data has the most direct and largest impact on charging data. The lower the temperature, the more charging data is required. Therefore, the accuracy of the prediction results is also limited by the temperature. Furthermore, since the prediction results are indirectly determined by the autocorrelation coefficient and the partial correlation coefficient, and the autocorrelation coefficient is only related to the data itself, that is, it is only related to the charging data, the obtained prediction results will not include the influence of temperature data. In order to eliminate this influence, it is chosen to add temperature data to the autocorrelation coefficient graph when generating the autocorrelation coefficient graph, thereby changing the order of the ARIMA model. When the model order changes, the prediction results will also change accordingly, thereby optimizing the prediction results.
[0061] Specifically, for the kth sampling time range, the autocorrelation coefficient of the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range is obtained by the ARIMA algorithm, the autocorrelation coefficient of the charging data segment of the kth sampling time range and the charging data segment of the first sampling time range is obtained by the ARIMA algorithm, and the partial autocorrelation coefficient of the temperature data segment of the kth sampling time range and the charging data segment of the kth sampling time range is obtained by the ARIMA algorithm.
[0062] As an example, the calculation method for obtaining the optimization factor of the kth sampling time range is:
[0063]
[0064] In the formula, YH k represents the optimization factor of the kth sampling time range; I' represents the total number of all sampling time ranges in the stable temperature-charging data sequence; Zg 1,k The autocorrelation coefficient between the temperature data segment of the kth sampling time range and the temperature data segment of the first sampling time range; Zfg j Represents the partial autocorrelation coefficient between the temperature data segment in the j-th sampling time range and the charging data segment in the j-th sampling time range; || represents the absolute value.
[0065] The optimization factors of all sampling time ranges are obtained, and each optimization factor after linear normalization of all optimization factors is recorded as the degree of optimization.
[0066] Among them, |Zg 1,k | represents the optimization weight of the kth sampling time range. The larger the value, the greater the overall weight, and thus the greater the temperature-charging optimization correlation coefficient of the kth sampling time range. The larger its value is when establishing the correlation coefficient graph, the more orders are calculated, and the better the fit of the prediction results.
[0067] As an example, the calculation method for obtaining the temperature-charging optimization correlation coefficient of the kth sampling time range is:
[0068] Sim k =Zf 1,k ×YH k '
[0069] In the formula, Sim k represents the temperature-charging optimization correlation coefficient for the kth sampling time range; Zf 1,k represents the autocorrelation coefficient between the charging data segment of the kth sampling time range and the charging data segment of the first sampling time range; YH k 'Indicates the optimization degree of the kth sampling time range.
[0070] The autocorrelation graph and partial autocorrelation graph are constructed according to the temperature-charge optimization correlation coefficients of all sampling time ranges, and the autocorrelation graph and partial autocorrelation graph are input into the ARIMA algorithm to obtain the optimal order.
[0071] Specifically, the optimal order is input into the ARIMA algorithm to obtain the ARIMA prediction model. The latest charging data of the charging pile is obtained through the ARIMA prediction model. The user sets the charging time according to the latest charging data, thereby achieving the effect of intelligent operation and management of the charging pile.
[0072] Among them, obtaining the autocorrelation coefficient, the partial autocorrelation coefficient and the optimal order, constructing the autocorrelation graph and the partial autocorrelation graph, and obtaining the ARIMA prediction model are prior arts, and this embodiment will not be described in detail here; the specific process of obtaining the ARIMA prediction model is as follows Figure 2 .
[0073] Through the above steps, a smart operation and management method for new energy vehicle charging piles is completed.
[0074] The present invention also proposes a smart operation and management system for new energy vehicle charging piles, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a smart operation and management method for new energy vehicle charging piles in steps S001 to S003 are implemented.
[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A smart operation and management method for new energy vehicle charging piles, characterized in that: The method comprises the following steps: Acquire a temperature dimension data sequence and a charging dimension data sequence of a charging pile; the temperature dimension data sequence includes temperature data segments within a plurality of sampling time ranges, and the charging dimension data sequence includes charging data segments within a plurality of sampling time ranges; the temperature data segment includes a plurality of temperature data, and the charging data segment includes a plurality of charging data; within the same sampling time range, one temperature data segment corresponds to one charging data segment; According to the distribution of temperature data in the temperature data segment of each sampling time range in the temperature dimension data sequence, the stability of the temperature dimension data sequence is obtained; according to the distribution of charging data in the charging data segment of each sampling time range in the charging dimension data sequence, the stability of the charging dimension data sequence is obtained; according to the stability, the temperature dimension data sequence and the charging dimension data sequence are differentially processed to obtain a stabilized temperature-charging data sequence, wherein the stabilized temperature-charging data sequence includes temperature data segments and charging data segments of several sampling time ranges; According to the temperature data segment and the charging data segment of each sampling time range in the stabilized temperature-charging data sequence, a temperature-charging optimization correlation coefficient of each sampling time range is obtained; The specific method of obtaining the temperature-charging optimization correlation coefficient of each sampling time range according to the temperature data segment and the charging data segment of each sampling time range in the stabilized temperature-charging data sequence includes: For The sampling time range is obtained by ARIMA algorithm. The autocorrelation coefficient of the temperature data segment of the first sampling time range and the temperature data segment of the first sampling time range, The autocorrelation coefficient of the charging data segment of the first sampling time range and the charging data segment of the first sampling time range and the The temperature data segment of the sampling time range is The partial autocorrelation coefficient of the charging data segment within the sampling time range; According to The autocorrelation coefficient of the temperature data segment of the first sampling time range and the temperature data segment of the first sampling time range is obtained. The optimization weight of the sampling time range; The temperature data segment of the sampling time range is The partial autocorrelation coefficient of the charging data segment in the sampling time range and the The optimization weight of the sampling time range is obtained The optimization factor for the sampling time range; The optimization factor for the sampling time range is obtained The degree of optimization of the sampling time range; The first The autocorrelation coefficient of the charging data segment of the first sampling time range and the charging data segment of the first sampling time range is The product of the optimization degree of the sampling time range is taken as the Temperature-charging optimization correlation coefficient for each sampling time range; According to the The autocorrelation coefficient of the temperature data segment of the first sampling time range and the temperature data segment of the first sampling time range is obtained. The optimization weights of the sampling time range include the following specific methods: The first The absolute value of the autocorrelation coefficient between the temperature data segment of the first sampling time range and the temperature data segment of the first sampling time range is taken as the The optimization weights of the sampling time ranges; According to the The temperature data segment of the sampling time range is The partial autocorrelation coefficient of the charging data segment in the sampling time range and the The optimization weight of the sampling time range is obtained The specific formula for the optimization factor of a sampling time range is: ; In the formula, Indicates Optimization factor for a sampling time range; Represents the total number of all sampling time ranges in the stable temperature-charging data series; Indicates The optimization weights of the sampling time ranges; Indicates The temperature data segment of the sampling time range is The partial autocorrelation coefficient of the charging data segment within the sampling time range; Indicates taking the absolute value; The ARIMA prediction model is obtained according to the temperature-charging optimization correlation coefficient, and the latest charging data of the charging pile is predicted according to the ARIMA prediction model.
2. According to claim 1, a new energy vehicle charging pile intelligent operation and management method is characterized in that: The specific method of obtaining the stability of the temperature dimension data sequence according to the distribution of the temperature data in the temperature data segment of each sampling time range in the temperature dimension data sequence is as follows: According to the difference between each temperature data and the adjacent temperature data in the temperature data segment of each sampling time range, the difference factor of each temperature data in the temperature data segment of each sampling time range is obtained; according to the discrete situation of all temperature data in the temperature data segment of each sampling time range, the stable weight of the temperature data segment of each sampling time range is obtained; According to the difference factor of each temperature data in the temperature data segment of each sampling time range and the stability weight of the temperature data segment of each sampling time range, the calculation method for obtaining the stability of the temperature dimension data sequence is: ; In the formula, Indicates the stability of the temperature dimension data series; Represents the total number of temperature data segments in all sampling time ranges in the temperature dimension data series; Indicates the total number of all temperature data in the temperature data segment of each sampling time range; Indicates the first The temperature data segment within the sampling time range The difference factor of each temperature data; Indicates the first Stable weights of temperature data segments within a sampling time range; represents an exponential function with a natural constant as base; Indicates taking the absolute value.
3. According to claim 2, a new energy vehicle charging pile intelligent operation and management method is characterized in that: The specific method of obtaining the difference factor of each temperature data in the temperature data segment of each sampling time range according to the difference between each temperature data and the adjacent temperature data in the temperature data segment of each sampling time range includes: Preset a hyperparameter , the first The temperature data segment within the sampling time range The temperature data and the temperature dimension data sequence The temperature data segment within the sampling time range The difference of the temperature data is recorded as the first difference; the The temperature data segment within the sampling time range The temperature data and the temperature dimension data sequence The temperature data segment within the sampling time range The difference of temperature data is recorded as the second difference, and the second difference is added to the hyperparameter The sum of is recorded as the first sum value; The ratio of the first difference to the first sum is recorded as the first ratio, and the difference between 1 and the first ratio is recorded as the first value in the temperature dimension data sequence. The temperature data segment within the sampling time range The difference factor of the temperature data.
4. According to claim 2, a new energy vehicle charging pile intelligent operation and management method is characterized in that: The specific method of obtaining the stable weight of the temperature data segment of each sampling time range according to the discrete conditions of all temperature data in the temperature data segment of each sampling time range is as follows: The temperature dimension data sequence The variance of all temperature data in the temperature data segment within the sampling time range is taken as the The stability weight of the temperature data segment in the sampling time range.
5. According to the method for intelligent operation and management of new energy vehicle charging piles according to claim 1, it is characterized in that: The specific method of performing differential processing on the temperature dimension data sequence and the charging dimension data sequence according to stability to obtain a stabilized temperature-charging data sequence includes: Preset a stable parameter , if the stability of the temperature dimension data sequence is less than the stability parameter , perform differential processing on the temperature dimension data sequence until the stability of the temperature dimension data sequence is greater than or equal to the stability parameter , and obtain the stable temperature dimension data sequence; if the stability of the charging dimension data sequence is less than the stable parameter , perform differential processing on the charging dimension data sequence until the stability of the charging dimension data sequence is greater than or equal to the stability parameter , obtain a stabilized charging dimension data sequence; the stabilized charging dimension data sequence and the stabilized temperature dimension data sequence are combined to form a stabilized temperature-charging data sequence.
6. According to the method for intelligent operation and management of new energy vehicle charging piles according to claim 1, it is characterized in that: According to the The optimization factor for the sampling time range is obtained The optimization degree of each sampling time range includes the following specific methods: After all optimization factors are linearly normalized, each optimization factor is recorded as the degree of preference.
7. A smart operation and management system for new energy vehicle charging piles, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the smart operation and management method of a new energy vehicle charging pile as described in any one of claims 1-6 are implemented.
