An extended-range electric energy storage and discharge station operation status monitoring and early warning system

Abstract

The present invention relates to the technical field of equipment operation monitoring and early warning, and specifically to an extended-range electric energy storage and discharge station operation status monitoring and early warning system, the system includes a processor and a memory, the processor executes a computer program stored in the memory to implement the following steps: obtaining turning data, non-turning data and stable index values ​​of non-turning data in a historical operation data sequence; according to the turning data and the stable index values ​​of non-turning data, obtaining weight factors corresponding to each historical operation data, and according to the ARIMA algorithm, the historical operation data sequence and the weight factors corresponding to each historical operation data, obtaining the prediction data corresponding to the current operation time; finally, according to the prediction data and the current operation data, monitoring and early warning the operation status of the extended-range electric energy storage and discharge station. And the present invention can accurately and reliably monitor and early warning the operation status of the extended-range electric energy storage and discharge station.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment operation monitoring and early warning technology, and in particular to an extended-range electric energy storage and discharge station operation status monitoring and early warning system. Background Art

[0002] At present, extended-range electric energy storage and discharge stations play an important role in the power system. The main reason is that extended-range electric energy storage and discharge stations have the performance of balancing supply and demand, regulating electricity, improving energy utilization efficiency, and improving power quality; however, when the extended-range electric energy storage and discharge station has an operating abnormality, the performance of the extended-range electric energy storage and discharge station will be affected. For example, when the extended-range electric energy storage and discharge station has an operating abnormality, it may not be able to balance supply and demand. Therefore, during the operation of the extended-range electric energy storage and discharge station, it is very important to monitor and warn the operating status of the extended-range electric energy storage and discharge station.

[0003] At present, the operation data of the extended-range energy storage and discharge station during operation is usually collected, such as the battery voltage, battery current, battery temperature, etc. of the extended-range energy storage and discharge station, which are all the operation data of the extended-range energy storage and discharge station during operation. Then, the preset fixed threshold is used to monitor the real-time collected operation data, that is, when the real-time collected operation data exceeds the preset fixed threshold, an early warning is generally issued. However, this method of monitoring and early warning based on the preset fixed threshold may fail to issue an early warning in time or issue an erroneous early warning. For example, the charging and discharging characteristics of the extended-range energy storage and discharge station or the changes in the load in the power grid, the changes in the weather environment, etc., will cause the battery voltage to have certain fluctuation characteristics, and the existence of such characteristics may make the preset battery voltage threshold unable to keenly capture the abnormal situation of the battery voltage or the situation of early warning error. For example, the collected battery voltage data may not exceed the preset battery voltage threshold, but the extended-range energy storage and discharge station is in abnormal operation at this time. It is also possible that the extended-range energy storage and discharge station is in normal operation at this time, but the collected battery voltage exceeds the preset battery voltage threshold; and in order to avoid using the preset fixed threshold for early warning, the preset fixed threshold is used to monitor the operation data collected in real time. The problem of failure to issue early warnings or erroneous early warnings caused by monitoring and early warning has led to the emergence of a method of monitoring and early warning by means of data prediction, that is, prediction is made based on the collected operating data and the ARIMA algorithm, and then the operating status of the extended-range electric energy storage and discharge station is monitored and warned based on the difference between the predicted data and the actual collected operating data. However, when the ARIMA algorithm is currently used for prediction, the weight of the data used to obtain the prediction result is generally determined by the residual between the fitting result and the corresponding data. For example, if a sequence is mainly used for the prediction of subsequent data, then the weight of each data in the sequence is determined by the residual between the fitting result of each data in the sequence and the corresponding data during the prediction process. However, this method of determining data weights will lead to unreliable or inaccurate prediction results in the subsequent prediction results. When the prediction results are unreliable or inaccurate, the subsequent warning will be inaccurate or unreliable when monitoring and warning the operating status of the extended-range electric energy storage and discharge station based on the prediction results. Therefore, how to accurately and reliably monitor and warn the operating status of the extended-range electric energy storage and discharge station has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above problems, the present invention provides an extended-range electric energy storage and discharge station operation status monitoring and early warning system, and the technical solutions adopted are as follows:

[0005] An embodiment of the present invention provides an extended-range electric energy storage and discharge station operation status monitoring and early warning system, comprising a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the following steps:

[0006] Obtain current operation data and historical operation data sequence of the extended-range electric energy storage and discharge station at the current operation time;

[0007] According to the preceding data and the succeeding data of each historical operation data in the historical operation data sequence, a change index value of each historical operation data is obtained, and according to the change index value, turning data and non-turning data in the historical operation data sequence are obtained;

[0008] According to adjacent historical running data of the non-turning data in the historical running data sequence, a slope difference index value of the non-turning data is obtained, and according to the difference between the non-turning data in the historical running data sequence and the adjacent historical running data and the slope difference index value, a stability index value of the non-turning data is obtained;

[0009] According to the stable index values ​​of the turning data and the non-turning data, the weight factors corresponding to the various historical operation data are obtained, and according to the ARIMA algorithm, the historical operation data sequence and the weight factors corresponding to the various historical operation data, the forecast data corresponding to the current operation moment is obtained;

[0010] Based on the predicted data and the current operating data, the operating status of the extended-range electric energy storage and discharge station is monitored and warned.

[0011] Beneficial effect: The present invention first obtains the current operating data and the historical operating data sequence of the extended-range electric energy storage and discharge station at the current operating moment; then obtains the change index value of each historical operating data according to the preceding data and the succeeding data of each historical operating data in the historical operating data sequence, obtains the turning data and non-turning data in the historical operating data sequence according to the change index value, obtains the slope difference index value of the non-turning data according to the adjacent historical operating data of the non-turning data in the historical operating data sequence, and obtains the stable index value of the non-turning data according to the difference between the non-turning data and the adjacent historical operating data in the historical operating data sequence and the slope difference index value; then obtains the weight factor corresponding to each historical operating data according to the stable index value of the turning data and the non-turning data, and obtains the predicted data corresponding to the current operating moment according to the ARIMA algorithm, the historical operating data sequence and the weight factor corresponding to each historical operating data; finally, the operating status of the extended-range electric energy storage and discharge station is monitored and warned according to the predicted data and the current operating data. Moreover, the weight factors corresponding to various historical operating data obtained by the present invention based on the stable index values ​​of turning point data and non-turning point data can make the obtained prediction results more reliable and accurate. Based on the reliable and accurate prediction results, the operating status of the extended-range electric energy storage and discharge station can also be accurately and reliably monitored and warned. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

[0013] Figure 1 The present invention is a flow chart of an extended-range electric energy storage and discharge station operation status monitoring and early warning method. DETAILED DESCRIPTION

[0014] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the embodiments of the present invention.

[0015] 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.

[0016] This embodiment provides an extended-range electric energy storage and discharge station operation status monitoring and early warning system, including a processor and a memory, wherein the processor executes a computer program stored in the memory to implement an extended-range electric energy storage and discharge station operation status monitoring and early warning method, such as Figure 1 As shown, the extended-range electric energy storage and discharge station operation status monitoring and early warning method includes the following steps:

[0017] Step S001, obtaining current operation data and historical operation data sequence of the extended-range electric energy storage and discharge station at the current operation time.

[0018] The main purpose of this embodiment is to improve the reliability and accuracy of early warning when monitoring and warning the operating status of an extended-range electric energy storage and discharge station; in addition, for ease of understanding, this embodiment will describe the monitoring and early warning process of the operating status of any extended-range electric energy storage and discharge station, that is, all extended-range electric energy storage and discharge stations that appear later in this embodiment are the same extended-range electric energy storage and discharge station.

[0019] Since the data related to the battery of the extended-range electric energy storage and discharge station can reflect the operating status of the extended-range electric energy storage and discharge station, the present embodiment will subsequently use the data related to the battery of the extended-range electric energy storage and discharge station as the operating data of the extended-range electric energy storage and discharge station, and then use the collection equipment to collect A types of operating data of the extended-range electric energy storage and discharge station at each working moment, including A types of operating data at the current working moment and A types of operating data at each historical operating moment; in addition, in specific applications, the implementer needs to set the value of A and the specific type of operating data collected according to actual conditions. For example, in the present embodiment, the value of A can be set to 3, and the types of operating data collected at each working moment can be the battery voltage, battery current, and battery temperature of the extended-range electric energy storage and discharge station.

[0020] To facilitate subsequent analysis, this embodiment will take any type of operating data as an example for analysis, that is, all operating data that appear in this embodiment are the same type of operating data. For example, this embodiment can take battery voltage as an example for analysis, that is, all operating data that appear in this embodiment are battery voltage. Based on the above-mentioned acquisition process, this embodiment can acquire the operating data of the extended-range electric energy storage and discharge station at the current operating moment. This embodiment records the acquired operating data at the current operating moment as the current operating data at the current operating moment, and the current operating data is the battery voltage at the current operating moment.

[0021] Since the ARIMA algorithm will be used for prediction in the future in this embodiment, the present embodiment first needs to obtain a preset number of autoregressive terms P, and the preset number of autoregressive terms P in this embodiment is an empirical value, such as the preset number of autoregressive terms P can be set to 10; then a time series sequence constructed by the operating data at P consecutive historical operating moments adjacent to the current operating moment is obtained, and recorded as the historical operating data sequence of the extended-range electric energy storage and discharge station at the current operating moment, the operating data in the historical operating data sequence is the battery voltage, and the historical operating data sequence is mainly used to obtain the predicted data at the current working moment, that is, the subsequent predicted data at the current working moment is mainly obtained through the historical operating data sequence.

[0022] In addition, in a specific application, the implementer needs to set the time interval between adjacent working moments according to the monitoring accuracy. If a higher monitoring accuracy is required, the time interval between adjacent working moments can be set to a smaller value, such as 1 second.

[0023] Therefore, this embodiment can obtain the current operating data and historical operating data sequence of the extended-range electric energy storage and discharge station at the current operating time through the above process; in addition, the historical operating data sequence in this embodiment is a preprocessed sequence, and the preprocessing includes denoising, missing value completion and other operations.

[0024] Step S002, obtaining a change index value of each historical operation data according to the preceding data and succeeding data of each historical operation data in the historical operation data sequence, and obtaining the turning data and non-turning data in the historical operation data sequence according to the change index value; obtaining a slope difference index value of the non-turning data according to the adjacent historical operation data of the non-turning data in the historical operation data sequence, and obtaining a smoothness index value of the non-turning data according to the difference between the non-turning data in the historical operation data sequence and the adjacent historical operation data and the slope difference index value.

[0025] At present, extended-range electric energy storage and discharge stations will store renewable energy to meet the diversified needs of modern society for electricity, and when the power of the grid is sufficient, the extended-range electric energy storage and discharge station will convert the electrical energy into chemical energy and store it in the battery; when the power supply of the grid is insufficient, the extended-range electric energy storage and discharge station will convert the stored chemical energy into electrical energy through the discharge equipment to supply the grid, that is, the working principle of the extended-range electric energy storage and discharge station is to charge when the power of the grid is sufficient, and to discharge when the power of the grid is insufficient; and when the extended-range electric energy storage and discharge station is charging and discharging, the changes in the data related to the battery in the extended-range electric energy storage and discharge station are different, and the conversion between discharge and charging of the extended-range electric energy storage and discharge station will also cause turning point data in the collected operation data sequence. The battery voltage in the power station will increase, and when the extended-range energy storage and discharge station discharges, the battery voltage in the extended-range energy storage and discharge station will decrease. Therefore, the conversion of the extended-range energy storage and discharge station from discharge to charging or charging to discharge will cause turning point data in the collected battery voltage sequence, that is, the conversion between discharge and charging will cause turning point data in the collected operation data sequence. If from a certain moment, the extended-range energy storage and discharge station begins to convert from discharge to charging, then turning point data may appear in the future moments adjacent to this moment, and the turning point data is also called watershed number data, that is, in the operation data sequence, the data on the left and right of the turning point data have different attributes. For example, in a certain sequence, the data on the left side of the turning point data may be monotonically increasing, while the data on the right side of the turning point data may be monotonically decreasing.

[0026] In addition, the collected operating data sequence may not only contain turning data, but also some data with unstable characteristics, that is, unstable data, and these data with unstable characteristics may be caused by changes in the load of the power grid, weather factors, changes in the operation mode of the power grid, etc.

[0027] However, if the historical data used in the forecast contains turning data and data with non-stationary characteristics, then if the traditional data weight determination method is still used to determine the data weight, it may have a more serious impact on the reliability and accuracy of the forecast results. That is, the existence of turning data and data with non-stationary characteristics may make the forecast results unreliable and inaccurate. The main reason is that when the existing technology uses the ARIMA model for forecasting, in order to make the forecast result more accurate, it generally first determines the data weight for subsequent forecasting based on the residual between the fitting result of the collected data and the collected data. For example, if a certain sequence is mainly used for the forecast of subsequent data, then in the forecast process, the weight of each data in the sequence is obtained by calculating the weight of each data in the sequence. The traditional method of determining data weights is based on the residual between the fitting result of the data and the corresponding data. However, this traditional method of determining data weights may cause the turning point data to be smoothed. Since the turning point data can reflect the changing characteristics of the operating data during charging and discharging of the extended-range energy storage and discharge station, when the turning point data is smoothed, the tightness of the combination between the predicted data and the actual data will be lower. In other words, when the turning point data is smoothed, there may be a large deviation between the predicted data and the actual data, which means that when the turning point data is smoothed, the reliability and accuracy of the subsequent prediction results are also relatively low. In addition, this traditional method of determining data weights may also amplify the interference of data with non-stationary characteristics on the prediction results, thereby making the accuracy and reliability of the obtained prediction results lower.

[0028] Therefore, based on the above description, it can be known that the existence of turning point data and data with unstable characteristics is an important reason for the unreliability and inaccuracy of the prediction results. When the prediction results are unreliable or inaccurate, it will also cause the subsequent warning to be inaccurate or unreliable when monitoring and warning the operating status of the extended-range electric energy storage and discharge station based on the prediction results. That is, the reliability or accuracy of the prediction results is crucial to the reliability and accuracy of the subsequent warnings. In order to ensure the reliability and accuracy of the prediction results, this embodiment needs to avoid the turning point data from being smoothed as much as possible and the data with unstable characteristics from being amplified as much as possible in the subsequent prediction process. The amplification refers to the interference of the amplification on the prediction results. In order to avoid the turning point data from being smoothed as much as possible and the data with unstable characteristics from being amplified as much as possible, it is necessary to make the weight of the turning point data higher and the weight of the data with unstable characteristics lower during the prediction, that is, the more obvious the unstable characteristics, the lower the weight of the data, and the more stable the characteristics, the higher the weight of the data. The more obvious the data weight, the higher the weight; thus, it can be known that the subsequent embodiment needs to pay attention to the turning data and the data with non-stationary characteristics in the sequence used for data prediction, and because the historical operation data sequence at the current operation time obtained by the above steps is mainly used for the subsequent acquisition of the prediction data at the current operation time, the present embodiment needs to first obtain the turning data and non-turning data in the historical operation data sequence, and then determine the weight factor corresponding to each historical operation data in the historical operation data sequence by analyzing the stability index value of the non-turning data, and the stability index value can reflect the obvious degree of the non-stationary characteristics of the corresponding data; however, before obtaining the turning data and non-turning data, it is also necessary to first obtain the change index value of each historical operation data according to the preceding data and the succeeding data of each historical operation data in the historical operation data sequence, and the size of the change index value can reflect the possibility that the corresponding historical operation data is the turning data or the non-turning data, then the specific process of obtaining the change index value of each historical operation data is as follows:

[0029] First, a two-dimensional coordinate system is constructed, and the coordinate axes of the two-dimensional coordinate system constructed at this time are related to the type of operating data. Since this embodiment takes battery voltage as an example for analysis, the horizontal axis of the two-dimensional coordinate system constructed at this time is time, and the vertical axis is battery voltage; then, each historical operating data in the historical operating data sequence at the current operating moment and the collection time of each historical operating data are mapped to the constructed two-dimensional coordinate system, and the data points corresponding to each historical operating data in the historical operating data sequence are obtained, and the vertical axis value of the data point corresponding to each historical operating data is the corresponding historical operating data, and the horizontal axis value is the collection time of the corresponding historical operating data.

[0030] After obtaining the data points corresponding to the historical operation data, for ease of understanding, this embodiment takes the process of obtaining the change index value of the a-th historical operation data in the historical operation data sequence as an example to describe the process based on the data points corresponding to the historical operation data that have been obtained, and a is not equal to 1 and P, P is the preset number of autoregressive terms, and is also the total number of operation data in the historical operation data sequence, then the process of obtaining the change index value of the a-th historical operation data is:

[0031] First, a straight line formed by the data points corresponding to the 1st historical operation data in the historical operation data sequence and the data points corresponding to the ath historical operation data in the historical operation data sequence is obtained, and recorded as the first straight line. A straight line formed by the data points corresponding to the Pth historical operation data in the historical operation data sequence and the data points corresponding to the ath historical operation data is obtained, and recorded as the second straight line. Then, the absolute value of the difference between the slope of the first straight line and the slope of the second straight line is obtained, and recorded as the first absolute value of the difference. After that, the first absolute value of the difference is normalized, and the value obtained by the normalization is recorded as the first eigenvalue. Next, the Pearson correlation coefficient between the first straight line and the second straight line is obtained, and the Pearson correlation coefficient between the first straight line and the second straight line is normalized, and the value obtained by the normalization is recorded as the target correlation coefficient between the first straight line and the second straight line. Then, the inverse of the target correlation coefficient between the first straight line and the second straight line is obtained, and recorded as the second eigenvalue. Finally, the mean of the first eigenvalue and the second eigenvalue is obtained, and recorded as the change index value of the ath historical operation data.

[0032] In this embodiment, the specific expression for obtaining the change index value of the a-th historical operation data is:

[0033]

[0034] is the change index value of the ath historical operation data, norm() is the normalization function, K1 is the slope of the first straight line, K2 is the slope of the second straight line, R is the Pearson correlation coefficient between the first straight line and the second straight line; and when The bigger and The more hours, The larger the value of The larger the value of is, the more likely it is that the a-th historical running data is a turning point data. The smaller and The bigger the The smaller the value of The smaller the value is, the less likely it is that the a-th historical running data is a turning point data.

[0035] Therefore, this embodiment can obtain the change index value of each historical operation data through the above process. Since the size of the change index value can reflect the probability that the historical operation data is turning point data, after obtaining the change index value, this embodiment obtains the turning point data and non-turn point data in the historical operation data sequence according to the change index value of each historical operation data in the historical operation data sequence, and the specific acquisition process is:

[0036] First, the change index values ​​of the historical operation data in the historical operation data sequence are sorted in order from small to large, and the sorted sequence is recorded as the change index value sequence, that is, the data in the change index value sequence is arranged in order from small to large; then the average of the two change index values ​​corresponding to the second quartile and the third quartile in the change index value sequence is obtained, and recorded as the target judgment index value; and the change index value corresponding to the second quartile in the change index value sequence refers to the change index value at the middle position of the change index value sequence, and the change index value corresponding to the third quartile in the change index value sequence refers to the change index value at the 75% position of the change index value sequence.

[0037] Afterwards, in the historical operation data sequence, all historical operation data whose change index value is not less than the target judgment index value are obtained and recorded as turning point data; then all data except the turning point data in the historical operation data sequence are recorded as non-turning point data, and the 1st historical operation data and the Pth historical operation data in the historical operation data sequence are also non-turning point data.

[0038] Therefore, this embodiment obtains the turning data and non-turning data in the historical operation data sequence through the above process; and after obtaining the turning data and non-turning data in the historical operation data sequence, in order to ensure the reliability and accuracy of subsequent predictions, it is also necessary to analyze the non-stationary characteristics of the non-turning data, that is, the stationary index value of the non-turning data. However, before obtaining the stationary index value of the non-turning data, it is also necessary to obtain the slope difference index value of each non-turning data based on the adjacent historical operation data of the non-turning data in the historical operation data sequence. Subsequently, on the basis of the known slope difference index value, the stationary index value of the non-turning data can be determined. Then the specific acquisition process of the slope difference index value of the non-turning data is as follows:

[0039] First, the set constructed by the first two historical operation data and the last two historical operation data in the historical operation data sequence is recorded as the feature set; and because the calculation value of the stable index value of each data in the feature set is small, that is, the reference value of the stable index value of each data in the feature set obtained by the calculation method is small, so in this embodiment, the stable index value of the data in the feature set is not calculated, that is, the assignment method is used later to determine the stable index value of each data in the feature set; and because the purpose of obtaining the slope difference index value is to obtain the stable index value, the slope difference index value of the data in the feature set is not calculated in this embodiment; in addition, for ease of understanding, this embodiment will take the acquisition process of the slope difference index value of any non-turning data f in the historical operation data sequence as an example to describe, and the turning data f does not belong to the feature set, then the acquisition process of the slope difference index value of the non-turning data f is:

[0040] First, in the historical running data sequence, two historical running data located on the left side of the non-turning data f and closest to the non-turning data f are obtained, and they are recorded as the first left neighbor data and the second left neighbor data of the non-turning data f, respectively. Then, the slope of the straight line formed by the data points corresponding to the first left neighbor data and the data points corresponding to the second left neighbor data are obtained, and they are recorded as the left adjacent slope of the non-turning data f; then, in the historical running data sequence, two historical running data located on the right side of the non-turning data f and closest to the non-turning data f are obtained, and they are recorded as the first right neighbor data and the second right neighbor data of the non-turning data f, respectively. Then, the slope of the straight line formed by the data points corresponding to the first right neighbor data and the data points corresponding to the second right neighbor data are obtained, and they are recorded as the right adjacent slope of the non-turning data f; then, the absolute value of the difference between the left adjacent slope and the right adjacent slope is obtained, which is recorded as the slope difference index value of the non-turning data f, and the slope difference index value of the non-turning data is an important parameter for subsequently obtaining the stability index value of the corresponding non-turning data. For example, if the non-turning data f is the vth historical running data in the historical running data sequence, then the first left neighbor data and the second left neighbor data of the non-turning data f are the v-1th historical running data and the v-2th historical running data in the historical running data sequence, respectively, and the first right neighbor data and the second right neighbor data of the non-turning data f are the v+1th historical running data and the v+2th historical running data in the historical running data sequence, respectively.

[0041] Therefore, this embodiment can obtain the slope difference index value of all non-turning data except the feature set through the above process. After obtaining the slope difference index value, this embodiment then obtains the stability index value of the non-turning data according to the difference between the non-turning data in the historical operation data sequence and the adjacent historical operation data and the slope difference index value; and for ease of understanding, this embodiment will take the acquisition process of the stability index value of any non-turning data h in the historical operation data sequence as an example to describe, that is, the acquisition process of the stability index value of the non-turning data h is:

[0042] First, determine whether the non-turning data h belongs to the feature set. If so, use the preset second constant as the stationary index value of the non-turning data h; otherwise, obtain the second characterization value of the non-turning data h, and record the normalized value of the second characterization value of the non-turning data h as the stationary index value of the non-turning data h.

[0043] In this embodiment, when the non-turning data h does not belong to the feature set, the process of obtaining the second characterization value of the non-turning data h is: first, the adjacent difference characterization value of the non-turning data h is obtained, and the slope difference index value of the non-turning data h is multiplied by the adjacent difference characterization value of the non-turning data h, and then the value obtained after adding the preset first constant is recorded as the first characterization value of the non-turning data h, and then the inverse of the first characterization value of the non-turning data h is obtained, and recorded as the second characterization value of the non-turning data h.

[0044] In this embodiment, the method for obtaining the adjacent difference characterization values ​​of the non-turning data h is:

[0045] In the historical running data sequence, a historical running data located on the left side of the non-turning data h and closest to the non-turning data h is obtained, and recorded as the first most adjacent data of the non-turning data h, and a historical running data located on the right side of the non-turning data h and closest to the non-turning data h is obtained, and recorded as the second most adjacent data of the non-turning data h; for example, if the non-turning data h is the vth historical running data in the historical running data sequence, then the first most adjacent data of the non-turning data h is the v-1th historical running data in the historical running data sequence, and the second most adjacent data of the non-turning data h is the v+1th historical running data in the historical running data sequence. Then, the absolute value of the difference between the first most adjacent data and the non-turning data h is calculated, and recorded as the first difference value, the absolute value of the difference between the second most adjacent data and the non-turning data h is calculated, and recorded as the second difference value, and the sum of the first difference value and the second difference value is recorded as the adjacent difference characterization value of the non-turning data h.

[0046] In specific applications, the implementer needs to set the preset first constant and the preset second constant according to actual conditions. For example, in this embodiment, the preset first constant and the preset second constant can be set to 1 and 0.5 respectively, and the value range of the preset second constant is required to be neither less than 0 nor greater than 1.

[0047] In addition, when the non-turning data h does not belong to the feature set, the specific expression for calculating the stationary index value of the non-turning data h is:

[0048]

[0049] is the stable index value of the non-turning data h, is the left adjacent slope of the non-turning data h, is the right adjacent slope of the non-turning data h, W0 is the non-turning data h, W1 is the first most adjacent data of the non-turning data h, and W2 is the second most adjacent data of the non-turning data h. is the slope difference index value of the non-turning data h, is the adjacent difference representation value of the non-turning data h.

[0050] And when The bigger and The larger the The smaller the value of The smaller the value of is, the more obvious the instability characteristics of the non-turning data h are, that is, the more likely the non-turning data h is to be unstable. The smaller and The smaller the time, the The larger the value of The larger the value of , the more obvious the stability characteristics of the non-turning data h are, that is, the less likely the non-turning data h is to be data with unstable characteristics.

[0051] Therefore, this embodiment can obtain the stable index value of each non-turning data through the above process.

[0052] Step S003, according to the stability index values ​​of the turning data and the non-turning data, obtain the weight factors corresponding to the various historical operation data, and according to the ARIMA algorithm, the historical operation data sequence and the weight factors corresponding to the various historical operation data, obtain the prediction data corresponding to the current operation moment.

[0053] After obtaining the stable index value, this embodiment obtains the weight factor corresponding to each historical operation data in the historical operation data sequence according to the stable index values ​​of the turning data and the non-turning data, and the size of the weight factor corresponding to the historical operation data can reflect the importance of the corresponding historical operation data to the prediction result in the prediction process. The larger the weight factor corresponding to the historical operation data, the more important the corresponding historical operation data is to the prediction result in the prediction process. On the contrary, the smaller the weight factor corresponding to the historical operation data, the less important the corresponding historical operation data is to the prediction result in the prediction process. Therefore, the method for obtaining the weight factor corresponding to each historical operation data includes:

[0054] First, a set constructed by the stationary index values ​​of all non-turning data in the historical operation data sequence is obtained, and recorded as the stationary index value set; then the maximum stationary index value in the stationary index value set is selected, and recorded as the maximum assigned index value; then, according to the maximum assigned index value and the stationary index value of the non-turning data, the influence degree of the prediction result of each historical operation data in the historical operation data sequence is obtained; then, the cumulative sum of the influence degrees of the prediction results of all historical operation data in the historical operation data sequence is obtained, and recorded as the comprehensive influence degree, and then the ratio of the influence degree of the prediction result of each historical operation data to the comprehensive influence degree is obtained, and recorded as the weight factor corresponding to the corresponding historical operation data.

[0055] And the process of obtaining the influence degree of each historical operation data on the prediction result is: for the g-th historical operation data in the historical operation data sequence, if the g-th historical operation data is judged to be turning data, the maximum assigned index value is used as the influence degree of the prediction result of the g-th historical operation data; if the g-th historical operation data is judged to be non-turning data, the stability index value of the g-th historical operation data is used as the influence degree of the prediction result of the g-th historical operation data; and the greater the influence degree of the prediction result of the historical operation data, the more important the corresponding historical operation data is to the prediction result in the prediction process; and the method of obtaining the influence degree of the prediction result in this embodiment can make the weight factor corresponding to the turning data higher, and make the weight factor corresponding to the data with unstable characteristics lower.

[0056] Therefore, this embodiment can obtain the weight factors corresponding to each historical operation data in the historical operation data sequence through the above process. After obtaining the weight factors, the ARIMA algorithm is used to predict the data. Since the historical data used for prediction at this time is the historical operation data sequence at the current operation moment, the prediction result at this time is the predicted data corresponding to the current operation moment under the preset number of autoregressive terms P, that is, when the historical operation data sequence at the current operation moment and the weight factors corresponding to each historical operation data in the historical operation data sequence are known, the predicted data obtained by using the ARIMA algorithm is recorded as the predicted data corresponding to the current operation moment under the preset number of autoregressive terms P.

[0057] Therefore, this embodiment can obtain the prediction data corresponding to the current operating time according to the ARIMA algorithm, the historical operating data sequence at the current operating time, and the weight factors corresponding to each historical operating data in the historical operating data sequence; and because the historical operating data sequence at the current operating time and the weight factors corresponding to each historical operating data in the historical operating data sequence are known, the process of using the ARIMA algorithm to obtain prediction data is a well-known technology, and therefore it will not be described again in this embodiment.

[0058] Step S004: monitor and warn the operating status of the extended-range electric energy storage and discharge station based on the predicted data and the current operating data.

[0059] After obtaining the prediction data corresponding to the current operating time and the current operating data at the current operating time, this embodiment completes the monitoring and early warning of the operating status of the extended-range electric energy storage and discharge station based on the obtained prediction data corresponding to the current operating time and the current operating data at the current operating time, specifically:

[0060] First, a set of P historical running moments that are adjacent and continuous to the current running moment is obtained, and recorded as the historical running moment set. Then, the prediction data corresponding to each historical running moment in the historical running moment set is obtained. Since the method for obtaining the prediction data corresponding to the historical running moment is the same as the method for obtaining the prediction data corresponding to the current running moment, it will not be described in detail. It should be noted that the obtained prediction data are all prediction data under the preset number of autoregressive terms P.

[0061] Next, the absolute value of the difference between the current operating data at the current operating moment and the predicted data corresponding to the current operating moment is calculated, and recorded as the current actual predicted difference value; then the absolute value of the difference between the operating data at each historical operating moment in the historical operating moment set and the predicted data corresponding to the corresponding historical operating moment is calculated, and recorded as the historical actual predicted difference value of the corresponding historical operating moment, that is, the historical actual predicted difference value of any historical operating moment refers to the absolute value of the difference between the operating data at the historical operating moment and the predicted data corresponding to the historical operating moment; then the mean of the historical actual predicted difference values ​​of all historical operating moments in the historical operating moment set is obtained, and recorded as the mean of the historical actual predicted differences, the normalized value of the absolute value of the difference between the current actual predicted difference value and the mean of the historical actual predicted differences is obtained, and recorded as the warning indicator value at the current operating moment, and the operating data at the operating moments obtained in this embodiment are all data actually collected by the collection device, and in this embodiment, the normalization process is performed using the normalization function norm(). In addition, the larger the warning index value obtained at the current operating time, the more likely it is that the operating status of the extended-range electric energy storage and discharge station will be abnormal. Conversely, the smaller the warning index value obtained at the current operating time, the smaller the possibility of abnormality in the operating status of the extended-range electric energy storage and discharge station.

[0062] Therefore, it is determined whether the value of the warning indicator at the current operating time is greater than the preset warning threshold. If so, it is determined that an early warning needs to be triggered immediately and relevant personnel are notified to respond, such as to conduct inspections and repairs. Otherwise, it is determined that no early warning is required, that is, work can be continued. In specific applications, the implementer needs to set the preset warning threshold according to the actual situation. For example, in this embodiment, the preset warning threshold can be set to 0.6.

[0063] In addition, it should be noted that the present embodiment uses the battery voltage as an example for analysis, from which it can be seen that the warning index values ​​obtained are different for different types of operating data analyzed, that is, if the type of operating data involved in the present embodiment is A, then the number of warning index values ​​finally obtained at the current operating moment is also A, and among the A warning index values ​​obtained at the current operating moment, as long as there is a warning index value greater than the preset warning threshold, the warning is triggered immediately.

[0064] So far, this embodiment has completed the monitoring and early warning of the operating status of the extended-range electric energy storage and discharge station.

[0065] To summarize, this embodiment first obtains the current operating data and historical operating data sequence of the extended-range electric energy storage and discharge station at the current operating moment; then, according to the preceding data and succeeding data of each historical operating data in the historical operating data sequence, obtains the change index value of each historical operating data; according to the change index value, obtains the turning data and non-turning data in the historical operating data sequence; according to the adjacent historical operating data of the non-turning data in the historical operating data sequence, obtains the slope difference index value of the non-turning data; and according to the difference between the non-turning data and the adjacent historical operating data in the historical operating data sequence and the slope difference index value, obtains the stable index value of the non-turning data; then, according to the stable index value of the turning data and the non-turning data, obtains the weight factor corresponding to each historical operating data, and according to the ARIMA algorithm, the historical operating data sequence and the weight factor corresponding to each historical operating data, obtains the predicted data corresponding to the current operating moment; finally, according to the predicted data and the current operating data, the operating status of the extended-range electric energy storage and discharge station is monitored and warned. Moreover, the weight factors corresponding to the various historical operating data obtained according to the stable index values ​​of the turning data and the non-turning data in this embodiment can make the obtained prediction results more reliable and accurate. Based on the more reliable and accurate prediction results, the operating status of the extended-range electric energy storage and discharge station can also be accurately and reliably monitored and warned.

[0066] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An extended-range electric energy storage and discharge station operation status monitoring and early warning system, comprising a processor and a memory, characterized in that: The processor executes the computer program stored in the memory to implement the following steps: Obtain current operation data and historical operation data sequence of the extended-range electric energy storage and discharge station at the current operation time; According to the preceding data and the succeeding data of each historical operation data in the historical operation data sequence, a change index value of each historical operation data is obtained, and according to the change index value, turning data and non-turning data in the historical operation data sequence are obtained; According to adjacent historical running data of the non-turning data in the historical running data sequence, a slope difference index value of the non-turning data is obtained, and according to the difference between the non-turning data in the historical running data sequence and the adjacent historical running data and the slope difference index value, a stability index value of the non-turning data is obtained; According to the stable index values ​​of the turning data and the non-turning data, the weight factors corresponding to the various historical operation data are obtained, and according to the ARIMA algorithm, the historical operation data sequence and the weight factors corresponding to the various historical operation data, the forecast data corresponding to the current operation moment is obtained; Monitoring and early warning the operating status of the extended-range electric energy storage and discharge station according to the predicted data and the current operating data; The method for obtaining the change index value of each historical operation data includes: Constructing a two-dimensional coordinate system, and mapping each historical operation data in the historical operation data sequence and the collection time of each historical operation data to the two-dimensional coordinate system, to obtain a data point corresponding to each historical operation data, wherein the ordinate value of the data point is the historical operation data, and the abscissa value is the collection time of the corresponding historical operation data; For the ath historical operation data in the historical operation data sequence, a is not equal to 1 and P: the straight line formed by the data points corresponding to the 1st historical operation data in the historical operation data sequence and the data points corresponding to the ath historical operation data is recorded as the first straight line; the straight line formed by the data points corresponding to the Pth historical operation data in the historical operation data sequence and the data points corresponding to the ath historical operation data is recorded as the second straight line; the normalized value of the absolute value of the difference between the slope of the first straight line and the slope of the second straight line is recorded as the first eigenvalue; the inverse of the target correlation coefficient between the first straight line and the second straight line is recorded as the second eigenvalue, and the target correlation coefficient refers to the normalized value of the Pearson correlation coefficient between the first straight line and the second straight line; the mean of the first eigenvalue and the second eigenvalue is recorded as the change index value of the ath historical operation data; The method for obtaining the slope difference index value of the non-turning data includes: Recording a set constructed by the first two historical operation data and the last two historical operation data in the historical operation data sequence as a feature set; For any non-turning data f in the historical operation data sequence, the non-turning data f does not belong to the feature set: the slope of the straight line formed by the data points corresponding to the two historical operation data closest to the non-turning data f on the left side of the non-turning data f is recorded as the left adjacent slope; the slope of the straight line formed by the data points corresponding to the two historical operation data closest to the non-turning data f on the right side of the non-turning data f is recorded as the right adjacent slope; the absolute value of the difference between the left adjacent slope and the right adjacent slope is recorded as the slope difference index value of the non-turning data f; The method for obtaining the stable index value of the non-turning data includes: For any non-turning data h in the historical operation data sequence: determine whether the non-turning data h belongs to the feature set, if so, use the preset second constant as the stability index value of the non-turning data h; otherwise, obtain the second characterization value of the non-turning data h, and record the normalized value of the second characterization value as the stability index value of the non-turning data h, the second characterization value is the reciprocal of the first characterization value of the non-turning data h, the first characterization value is the result of multiplying the slope difference index value of the non-turning data h and the adjacent difference characterization values ​​of the non-turning data h and adding the preset first constant; The method for obtaining adjacent difference characterization values ​​of the non-turning data h includes: Record a historical running data located on the left side of the non-turning data h and closest to the non-turning data h as the first most adjacent data; record a historical running data located on the right side of the non-turning data h and closest to the non-turning data h as the second most adjacent data; record the absolute value of the difference between the first most adjacent data and the non-turning data h as the first difference value, record the absolute value of the difference between the second most adjacent data and the non-turning data h as the second difference value, and record the sum of the first difference value and the second difference value as the adjacent difference characterization value of the non-turning data f; The method for obtaining the weight factor corresponding to each historical operation data includes: The set constructed by the stationary index values ​​of all non-turning data in the historical operation data sequence is recorded as the stationary index value set; the maximum stationary index value in the stationary index value set is selected and recorded as the maximum assigned index value; the influence degree of the prediction result of each historical operation data in the historical operation data sequence is obtained according to the maximum assigned index value and the stationary index value of the non-turning data; the cumulative sum of the influence degrees of the prediction results of all historical operation data in the historical operation data sequence is recorded as the comprehensive influence degree, and the ratio of the influence degree of the prediction result of each historical operation data to the comprehensive influence degree is recorded as the weight factor of the corresponding historical operation data; The method for obtaining the influence degree of the prediction result of each historical operation data includes: For the g-th historical operation data in the historical operation data sequence, if the g-th historical operation data is turning point data, the maximum assigned index value is used as the influence degree of the prediction result of the g-th historical operation data; if the g-th historical operation data is non-turning point data, the stable index value of the g-th historical operation data is used as the influence degree of the prediction result of the g-th historical operation data.

2. The extended-range electric energy storage and discharge station operation status monitoring and early warning system according to claim 1, characterized in that: The method for obtaining the historical operation data sequence includes: Obtain a preset number of autoregressive terms P and the operating data of the extended-range electric energy storage and discharge station at historical operating times, and record a time series constructed from the operating data at P consecutive historical operating times adjacent to the current operating time as the historical operating data sequence at the current operating time.

3. The extended-range electric energy storage and discharge station operation status monitoring and early warning system according to claim 1, characterized in that: The method for obtaining turning point data and non-turn point data in the historical operation data sequence includes: Sorting the change index values ​​of the historical operation data in the historical operation data sequence in order from small to large, and recording the sorted sequence as the change index value sequence; recording the average of the two change index values ​​corresponding to the second quartile and the third quartile in the change index value sequence as the target judgment index value; In the historical operation data sequence, all historical operation data whose change index value is not less than the target judgment index value are recorded as turning point data; and all data in the historical operation data sequence except the turning point data are recorded as non-turning point data.

4. The extended-range electric energy storage and discharge station operation status monitoring and early warning system according to claim 1, characterized in that: The method for monitoring and early warning the operating state of the extended-range electric energy storage and discharge station according to the prediction data and the current operating data comprises: Recording a set of P consecutive historical running moments adjacent to the current running moment as a historical running moment set, obtaining prediction data corresponding to each historical running moment in the historical running moment set, wherein the method for obtaining the prediction data corresponding to the historical running moment is the same as the method for obtaining the prediction data corresponding to the current running moment; The absolute value of the difference between the current operating data at the current operating moment and the predicted data corresponding to the current operating moment is recorded as the current actual predicted difference value; the absolute value of the difference between the operating data at each historical operating moment in the historical operating moment set and the predicted data corresponding to the corresponding historical operating moment is recorded as the historical actual predicted difference value of the corresponding historical operating moment, and the average of the historical actual predicted difference values ​​of all historical operating moments in the historical operating moment set is recorded as the historical actual predicted difference average; the normalized value of the absolute value of the difference between the current actual predicted difference value and the historical actual predicted difference average is recorded as the warning indicator value at the current operating moment; It is determined whether the warning indicator value is greater than a preset warning threshold value. If so, it is determined that a warning is immediately issued; otherwise, it is determined that a warning is not required.

Images