Probabilistic prediction method and system for safe temperature range of cascaded energy storage system

Abstract

A probabilistic prediction method and system for the safe temperature range of a cascaded energy storage system, establishing a temperature prediction model based on a graph neural network and a fully connected layer, and establishing a temperature probability distribution prediction model based on a graph neural network and the last Bayesian layer; establishing a temperature prediction error model; using historical values ​​of temperature and temperature-related monitoring quantities for model training; using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window, and the scaling factor to establish a temperature range model; when the confidence mean of the temperature range is equal to the set threshold, determining the optimal value of the scaling factor; obtaining the monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascaded energy storage system in the operating state, using the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor to obtain the safe temperature range of the cascaded energy storage system, providing a basis for monitoring the operating status of the energy storage system.

Description

Technical Field

[0001] The present invention belongs to the technical field of health status assessment of cascade utilization energy storage systems, and specifically relates to a probabilistic prediction method and system for a safe temperature range of a cascade utilization energy storage system. Background Art

[0002] In new power systems, the proportion of renewable energy, represented by wind power and photovoltaics, continues to increase. Due to the intermittent and unstable nature of natural resources like wind and solar energy, renewable energy generation is highly random and volatile, making it increasingly difficult to maintain a balanced power system. This has led to a surge in demand for peak and frequency regulation on the power grid, making it difficult to effectively address traditional peak and frequency regulation methods, such as thermal power units. Consequently, the introduction of battery energy storage power stations into the power system is urgently needed to improve system stability, reliability, and efficiency.

[0003] However, the high cost of batteries has become one of the main obstacles limiting the large-scale promotion and application of battery energy storage power stations in new power systems. Building second-use battery energy storage power stations based on retired batteries and participating in peak-shaving and frequency regulation auxiliary services has brought new options for reducing power station construction costs. However, for second-use battery energy storage power stations, ensuring safe operation is the primary issue. Retired batteries have served in various scenarios, with varying degrees of degradation and uneven performance and quality, and the probability of failure is higher than that of new energy storage batteries. When responding to peak-shaving and frequency regulation auxiliary services, energy storage batteries may need to be frequently charged and discharged, further increasing the probability of thermal runaway. Therefore, there is an urgent need to establish effective temperature monitoring methods for second-use energy storage systems.

[0004] The electrochemical reactions within energy storage batteries are complex, and numerous factors can influence their state. Monitoring the temperature changes of energy storage systems using methods driven by physical and chemical models is challenging. Existing data-driven methods often predict the maximum and minimum temperatures during normal operation of the energy storage system to provide a safe temperature range for the system. This can weaken the correlation between the upper and lower bounds of the temperature range, affecting prediction accuracy. Summary of the Invention

[0005] To address the deficiencies in the prior art, the present invention provides a probabilistic prediction method and system for the safe temperature range of a cascaded energy storage system. By constructing a probability distribution of the temperature at the time to be predicted, the upper and lower bounds of the temperature are obtained, and then the safe temperature range of the energy storage system is constructed, providing a basis for monitoring the operating status of the energy storage system.

[0006] The present invention adopts the following technical solutions.

[0007] The present invention proposes a probabilistic prediction method for the safe temperature range of a cascade utilization energy storage system, comprising:

[0008] Obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring quantities;

[0009] A temperature prediction model was established based on a graph neural network and a fully connected layer, and a temperature probability distribution prediction model was established based on a graph neural network and the last Bayesian layer. A temperature prediction error model was established using a shallow machine learning method.

[0010] The temperature prediction model and temperature prediction error model are trained using historical temperature values ​​and historical values ​​of temperature-related monitoring variables. The parameters of the graph neural network in the trained temperature prediction model are set as the initial parameters of the graph neural network in the temperature probability distribution prediction model. The temperature probability distribution prediction model with the initial parameters set is trained. Based on the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model, the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window.

[0011] A temperature interval model is established using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean within the prediction window, and a scaling factor. When the confidence mean of the temperature interval is equal to the set threshold, the optimal value of the scaling factor is determined.

[0012] The monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in operation are obtained, and the safe temperature range of the cascade utilization energy storage system is obtained using the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor.

[0013] Preferably, obtaining historical values ​​of the temperature of the cascade utilization energy storage system in normal operating conditions and historical values ​​of temperature-related monitoring variables includes:

[0014] Collect historical values ​​of different temperature-related monitoring quantities, including state of charge, current, voltage, and temperature;

[0015] Calculate the correlation coefficients between the various associated monitoring quantities, and determine the monitoring quantities that have a correlation relationship based on the correlation coefficients;

[0016] Divide the historical values ​​of temperature and temperature-related monitoring variables into sequences of different time periods, and detect suspected abnormal data in each time period sequence;

[0017] If other monitoring quantities that are associated with the monitoring quantity to which the suspected abnormal data belongs also have suspected abnormal data at the same time node, the suspected abnormal data is determined to be abnormal state data; otherwise, with the suspected abnormal data as the center of the sliding window, it is determined whether the absolute value of the difference between the normalized value of the abnormal data and the mean of the normalized data in the sliding window is greater than the set abnormal threshold. If it is greater, the suspected abnormal data is determined to be abnormal state data; after eliminating the abnormal state data and performing interpolation and repair, together with the historical values ​​of the temperature, a multi-segment historical normal data time series is formed.

[0018] Preferably, each period of historical normal data time series satisfies the following relationship:

[0019]

[0020] Where X is the historical normal data time series, 、 、 、……、 The historical value time series of the temperature is formed, 、 、 、……、 For the A time series of historical values ​​of temperature-related monitoring quantities, , is the total number of monitored quantities, is the length of each time series in the historical normal data time series X;

[0021] In each period of historical normal data time series, the time nodes in the historical normal data time series are Establish the moment for the sliding window, is the sliding window length, is the prediction window length, from time To time Constructing a sliding window, from time To time Constitute the prediction window.

[0022] Preferably, the fully connected layer is the last layer of the temperature prediction model;

[0023] The last Bayesian layer is the last layer of the temperature probability distribution prediction model;

[0024] Moreover, the graph neural network in the temperature prediction model has the same structure as that in the temperature probability distribution prediction model.

[0025] Preferably, based on the historical values ​​of the temperature in the sliding window and the historical values ​​of the temperature-related monitoring quantity, a temperature prediction model is used to output the temperature of the cascade utilization energy storage system in the prediction window;

[0026] The input data of the temperature prediction model satisfies the following relationship:

[0027]

[0028] Where, is the input data of the temperature prediction model, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities;

[0029] The output data of the temperature prediction model satisfies the following relationship:

[0030]

[0031] Where, is the output data of the temperature prediction model, 、 、 、……、 is the temperature within the prediction window.

[0032] Preferably, based on the historical values ​​of the temperature in the sliding window and the historical values ​​of the temperature-related monitoring quantity, a temperature probability distribution prediction model is used to output the temperature probability distribution of the cascade utilization energy storage system in the prediction window;

[0033] The input data of the temperature probability distribution prediction model satisfies the following relationship:

[0034]

[0035] Where, is the input data of the temperature probability distribution prediction model, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities.

[0036] Preferably, the temperature probability distribution within the prediction window is set to follow a Gaussian distribution , the output data of the temperature probability distribution prediction model satisfies the following relationship:

[0037]

[0038] Where, is the output data of the temperature probability distribution prediction model, 、 、 、……、 is the mean temperature within the prediction window, 、 、 、……、 is the standard deviation of temperature within the prediction window.

[0039] Preferably, according to the temperature in the prediction window, the temperature prediction error model is used to output the absolute value of the prediction error of the temperature of the cascade utilization energy storage system in the prediction window;

[0040] When the temperature prediction error model learns the temperature prediction error, the input data of the temperature prediction error model satisfies the following relationship:

[0041]

[0042] Where, is the input data of the temperature prediction error model, 、 、 、……、 is the temperature within the prediction window;

[0043] The output data of the temperature prediction error model satisfies the following relationship:

[0044]

[0045] Where, is the output data of the temperature prediction error model, 、 、 、……、 is the absolute value of the temperature prediction error within the prediction window.

[0046] Preferably, the temperature prediction model and the temperature prediction error model are trained using the historical values ​​of temperature and the historical values ​​of the temperature-related monitoring variables, including:

[0047] The historical values ​​of the temperature in the sliding window and the historical values ​​of the temperature-related monitoring quantity are used as the first training data set, and the historical values ​​of the temperature in the prediction window are used as the first label data;

[0048] The temperature prediction model is trained using the first training data set and the first label data; during the training process, the temperature within the prediction window output by the temperature prediction model is a deterministic estimate of the temperature history value within the prediction window, and has a corresponding relationship;

[0049] The temperature within the prediction window output by the temperature prediction model constitutes a second training data set, and the absolute value of the difference between the temperature within the prediction window and the corresponding first label data is used as the second label data;

[0050] The temperature prediction error model is trained using the second training data set and the second label, and the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature within the prediction window.

[0051] Preferably, the temperature probability distribution prediction model after the initial parameter setting is trained using the first training data set and the first label data, and the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean in the prediction window is output by the trained temperature probability distribution prediction model according to the temperature probability distribution in the prediction window.

[0052] Preferably, the temperature range model satisfies the following relationship:

[0053]

[0054] Where, and are the upper and lower bound correction values ​​of the temperature probability distribution within the prediction window, respectively. is the mean temperature within the prediction window, is the variance of temperature within the prediction window, is the absolute value of the prediction error of the temperature probability distribution within the prediction window, is the scaling factor;

[0055] The temperature upper limit correction value and the temperature lower limit correction value constitute a temperature interval, and the confidence mean of the temperature interval satisfies the following relationship:

[0056]

[0057] Where, is the confidence mean of the temperature interval, is the total number of groups of temperature upper limit correction value and temperature lower limit correction value, is the jth label value, is the jth indicator function, which is It takes 1 when it is in the temperature range, otherwise it takes 0;

[0058] When the confidence mean is equal to the set confidence level, the scaling factor is determined The confidence level is set to 0.95.

[0059] Preferably, based on the current moment Establish a lookback window of length m, starting from time To time Constitute the lookback window, from time To time The forecast period, is the length of the forecast period;

[0060] Obtaining the monitored values ​​of the temperature and temperature-related monitoring variables within the lookback window and inputting them into the temperature probability distribution prediction model to output the temperature probability distribution for the prediction period; determining the temperature mean for the prediction period based on the temperature probability distribution for the prediction period; inputting the temperature mean for the prediction period into the temperature prediction error model to obtain the absolute value of the prediction error of the temperature mean for the prediction period;

[0061] The input data of the temperature prediction error model satisfies the following relationship:

[0062]

[0063] Where, is the input data of the temperature prediction error model, 、 、 、……、 is the mean temperature during the forecast period;

[0064] Using the temperature mean of the prediction period, the absolute value of the prediction error of the temperature mean of the prediction period and the optimal value of the scaling factor, according to the temperature interval model, the temperature upper limit correction value and the temperature lower limit correction value of the prediction period are determined to form a safe temperature interval for the prediction period.

[0065] The present invention also proposes a probabilistic prediction system for the safe temperature range of a cascade utilization energy storage system, comprising:

[0066] An acquisition module is used to obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring quantities;

[0067] The model building module is used to build a temperature prediction model based on the graph neural network and the fully connected layer, and a temperature probability distribution prediction model based on the graph neural network and the last Bayesian layer; and a temperature prediction error model is built using shallow machine learning methods;

[0068] A model training module is used to train the temperature prediction model and the temperature prediction error model using historical temperature values ​​and historical values ​​of temperature-related monitoring variables, and to set the parameters of the graph neural network in the trained temperature prediction model as the initial parameters of the graph neural network in the temperature probability distribution prediction model; the temperature probability distribution prediction model after the initial parameters are set is trained, and the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean in the prediction window based on the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model;

[0069] The scaling factor optimization module is used to establish a temperature interval model using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean within the prediction window, and the scaling factor; when the confidence mean of the temperature interval is equal to the set threshold, the optimal value of the scaling factor is determined;

[0070] The prediction module is used to obtain the monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in the operating state, and use the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor to obtain the safe temperature range of the cascade utilization energy storage system.

[0071] The present invention also provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute steps of the method.

[0072] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when the program is executed by a processor.

[0073] The beneficial effects of the present invention are as follows: the present invention uses a probability prediction method to obtain the probability distribution of the operating temperature of the energy storage system at the prediction moment, and obtains the upper and lower bounds of the temperature during safe operation of the energy storage system based on the probability distribution, which can make full use of the correlation between the upper and lower bounds. The last layer of Bayes is selected as the implementation method of probability prediction, and the uncertainty quantification mechanism is introduced only in the last layer of the neural network model. It makes few changes to the previous model, has strong compatibility and high computational efficiency in practical applications. A mapping between the temperature prediction value and the absolute value of the temperature prediction error is established based on a shallow machine learning method. Based on the error amount and the established interval scaling factor, the safe temperature interval of the cascade utilization energy storage system under different operating modes is optimized and adjusted to adapt to the different working modes of the cascade utilization energy storage system and the different error modes of the prediction model, providing more accurate reference information for decision makers. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 This is a flow chart of a probabilistic prediction method for a safe temperature range of a cascade utilization energy storage system proposed by the present invention;

[0075] Figure 2 Schematic diagram of the temperature prediction model in an embodiment of the present invention;

[0076] Figure 3 2 is a schematic diagram of the structure of the temperature probability distribution prediction model in an embodiment of the present invention;

[0077] Figure 4 Schematic diagram of the safe temperature range of the energy storage system obtained in an embodiment of the present invention. DETAILED DESCRIPTION

[0078] To make the purpose, technical solutions and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the detailed implementation of the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0079] The present invention provides a probabilistic prediction method for the safe temperature range of a cascade utilization energy storage system. The cascade utilization energy storage system is constructed using retired batteries. The monitoring quantity of the cascade utilization energy storage system includes temperature. Figure 1 As shown, the method includes:

[0080] Step 1: Obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring variables.

[0081] Specifically, step 1 includes:

[0082] Step 1.1: Collect historical values ​​of different monitoring quantities of the cascade utilization energy storage system. In this embodiment, the historical monitoring data used for example testing comes from a long-serving energy storage system, which consists of 6 groups of battery submodules with different degrees of aging. The monitoring quantities of the cascade utilization energy storage system include but are not limited to: state of charge, current, voltage, and temperature;

[0083] Wherein, the temperature includes the temperature measurement value and / or the temperature maximum value, and the temperature maximum value includes the maximum temperature value and the temperature minimum value; when the temperature only includes the temperature maximum value, the average value of the maximum temperature value and the temperature minimum value is taken as the temperature measurement value;

[0084] In the embodiment, the monitoring quantities of the acquired monitoring data of the cascade utilization energy storage system are: state of charge, current, voltage, maximum temperature value and minimum temperature value.

[0085] Step 1.2: Use the correlation analysis method to determine the correlation coefficients between the various correlated monitoring quantities, and use the normalized values ​​of all correlation coefficients to establish a correlation distribution matrix. Rank all elements in the correlation distribution matrix from largest to smallest, and use the top 70% of the elements as candidate values. If the candidate value is not less than 0.7, the two monitoring quantities are determined to be correlated.

[0086] In the embodiment, the grey correlation analysis method is used to determine the correlation between historical monitoring data, which is a non-limiting and preferred choice. The correlation matrix of the present invention is not only used to distinguish which variables are strongly correlated with abnormal variables during data cleaning, but also to distinguish which variables are strongly correlated with temperature.

[0087] Due to the particularity of each battery cluster in the cascade utilization energy storage system, a specific method is required to first determine the temperature-related monitoring quantity and then use the relevant data. However, these historical values ​​contain "dirty" data and need to be cleaned.

[0088] Step 1.3: Divide the historical values ​​of temperature and temperature-related monitoring variables into sequences of different time periods; use unsupervised machine learning methods to detect abnormal data in the sequences of different time periods to obtain abnormal sequences with abnormal data;

[0089] In the embodiment, using a local anomaly factor detection algorithm to detect abnormal data in a sequence is a non-restrictive and preferred choice;

[0090] In step 1.4, if the monitoring quantity associated with the abnormal data also has abnormal data at the same time node, the abnormal data is determined to be abnormal state data; otherwise, the data in the sliding window is normalized with the abnormal data as the center, and the absolute value of the difference between the normalized value of the abnormal data and the mean of the normalized data in the sliding window is determined to be greater than the set abnormal threshold. If so, the abnormal data is considered as data to be cleaned; the abnormal threshold value is 0.5;

[0091] Step 1.5: Use interpolation to correct all the data to be cleaned.

[0092] In the embodiment, using a linear interpolation algorithm to correct all the data to be cleaned is a non-limiting and preferred choice.

[0093] In step 1.6, after correction, the abnormal state data is eliminated, and the historical values ​​of the temperature and the historical values ​​of the temperature-related monitoring variables of the cascaded energy storage system under normal operation are screened out to form multiple historical normal data time series. Each historical normal data time series satisfies the following relationship:

[0094]

[0095] Where X is the historical normal data time series, 、 、 、……、 The historical value time series of the temperature is formed, 、 、 、……、 For the A time series of historical values ​​of temperature-related monitoring quantities, , is the total number of monitored quantities, is the length of each time series in the historical normal data time series X;

[0096] In each period of historical normal data time series, the time nodes in the historical normal data time series are Establish the moment for the sliding window, is the sliding window length, is the prediction window length, from time To time Constructing a sliding window, from time To time The prediction window is constructed; the historical normal data time series is divided into two parts by using a sliding window and a prediction window. The data of the historical normal data time series in the sliding window is used as input data to obtain the prediction data that corresponds one to one with the time series in the prediction window, and the data of the historical normal data time series in the prediction window is used as a label to verify and determine the error of the prediction data, thereby improving the accuracy and reliability of the model's data estimation.

[0097] Step 2: Establish a temperature prediction model based on graph neural network and fully connected layer.

[0098] Specifically, if Figure 2 As shown in Figure 1, the temperature prediction model includes a graph neural network and a fully connected layer, where the fully connected layer is the last layer of the temperature prediction model. Specifically, based on the historical values ​​of the temperature within the sliding window and the historical values ​​of the temperature-related monitoring variables, the temperature prediction model outputs the temperature of the cascade utilization energy storage system within the prediction window.

[0099] Input data for the temperature prediction model Including: historical values ​​of temperature within the sliding window and historical values ​​of temperature-related monitoring quantities; wherein temperature-related monitoring quantities include but are not limited to: state of charge, current, and voltage;

[0100]

[0101] Where, is the input data of the temperature prediction model, Establish moments for the sliding window, is the sliding window length, from time To time Constructing a sliding window, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities;

[0102] The output data of the temperature prediction model includes the temperature outside the sliding window. The output data of the temperature prediction model satisfies the following relationship:

[0103]

[0104] Where, is the output data of the temperature prediction model, Establish moments for the sliding window, is the prediction window length, from time To time Constitute the prediction window, 、 、 、……、 is the temperature within the prediction window;

[0105] In the embodiment, the graph neural network used is a Fourier graph neural network, and the temperature prediction model is composed of a Fourier graph neural network and a fully connected layer. The temperature-related monitoring quantities in the embodiment include: state of charge, current, voltage, maximum temperature value, and minimum temperature value. The length m of the sliding window is 6 in the embodiment, and the length n of the prediction window is 1 in the embodiment.

[0106] It should be noted that the temperature prediction model can also be composed of other types of graph neural networks connected by fully connected layers.

[0107] Step 3: Establish a temperature probability distribution prediction model based on the graph neural network and the last layer of Bayesian.

[0108] Specifically, if Figure 3 As shown in the figure, the Bayesian last layer is the last layer of the temperature probability distribution prediction model; and the graph neural network in the temperature prediction model has the same structure as the graph neural network in the temperature probability distribution prediction model.

[0109] According to the historical values ​​of temperature in the sliding window and the historical values ​​of temperature-related monitoring quantities, the temperature probability distribution prediction model is used to output the temperature probability distribution of the cascade utilization energy storage system in the prediction window; based on the temperature probability distribution in the prediction window, the upper limit, lower limit and mean temperature in the prediction time period are determined.

[0110] The prediction of the safe temperature range in the prior art is usually a deterministic prediction mode, which assumes that the maximum temperature value and the minimum temperature value of the safe temperature range are two monitoring quantities. The time series prediction method is used to directly predict the values ​​of these two monitoring quantities at future moments, thereby constructing a safe temperature range. In the face of a cascade utilization energy storage system built with retired battery clusters, the status and operation mode of each retired battery are different. This deterministic prediction method may dilute the correlation constraint between the maximum temperature value and the minimum temperature value, and it is difficult to obtain a safe temperature range that conforms to different operation modes, resulting in false alarms and inaccurate predictions. Therefore, the present invention proposes to use a probabilistic prediction method to determine the safe temperature range of the cascade utilization energy storage system, by constructing a probability distribution that describes the possible values ​​of the temperature monitoring quantity of the cascade utilization energy storage system, and determining the upper and lower temperature limits with correlation constraints based on the temperature probability distribution, and then constructing a scaling factor based on the prediction error of the temperature prediction mean to adapt to the different operation modes of the cascade utilization energy storage system, so as to more accurately determine the safe temperature range of the cascade utilization energy storage system.

[0111] The input data of the temperature probability distribution prediction model include: historical values ​​of temperature and historical values ​​of temperature-related monitoring variables;

[0112] The input data of the temperature probability distribution prediction model satisfies the following relationship:

[0113]

[0114] Where, is the input data of the temperature probability distribution prediction model, Establish moments for the sliding window, is the sliding window length, from time To time Constructing a sliding window, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities;

[0115] Since the graph neural network in the temperature prediction model has the same structure as the graph neural network in the temperature probability distribution prediction model, the input data of the temperature probability distribution prediction model is the same as the input data of the temperature prediction model;

[0116] The output data of the temperature probability distribution prediction model includes: temperature probability distribution within the prediction window; wherein the temperature probability distribution includes but is not limited to Gaussian distribution;

[0117] When the temperature probability distribution within the prediction window takes Gaussian distribution When , the output data of the temperature probability distribution prediction model satisfies the following relationship:

[0118]

[0119] Where, is the output data of the temperature probability distribution prediction model, Establish moments for the sliding window, is the prediction window length, from time To time Constitute the prediction window, 、 、 、……、 is the mean temperature within the prediction window, 、 、 、……、 is the standard deviation of temperature within the prediction window.

[0120] In an embodiment, the graph neural network used is a Fourier graph neural network, and the temperature probability distribution prediction model is composed of a Fourier graph neural network and the last layer of Bayesian. The temperature-related monitoring quantities in the embodiment include: state of charge, current, voltage, maximum temperature, and minimum temperature. The length m of the sliding window is 6 in the embodiment, and the length n of the prediction window is 1 in the embodiment. In the present invention, the graph neural network in the temperature probability distribution prediction model has the same structure as the graph neural network in the temperature prediction model. Since a fully connected layer is used in the temperature prediction model, a deterministic prediction is achieved, while the temperature probability distribution prediction model achieves probabilistic prediction by introducing an uncertainty quantization mechanism using the last layer of Bayesian. When the model is actually deployed and applied, the temperature prediction model focuses on capturing the mapping relationship between input data and deterministic prediction values, thereby providing a more stable and clear error calculation benchmark for the subsequent temperature prediction error model, avoiding the temperature probability distribution prediction model from generating a relatively broad mean estimate due to uncertainty estimation, thereby reducing the performance of the temperature prediction error model.

[0121] Step 4: Use shallow machine learning methods to establish a temperature prediction error model.

[0122] In the embodiment, establishing the temperature prediction error model based on the support vector regression model is a non-limiting and preferred choice.

[0123] Specifically, according to the temperature within the prediction window, the temperature prediction error model is used to output the absolute value of the prediction error of the temperature of the cascade utilization energy storage system within the prediction window.

[0124] When obtaining the upper and lower temperature limits corresponding to each retired battery cluster, if the same set of uncertainty assessment criteria is used for retired batteries with different states and operating modes, the obtained temperature range will be relatively wide, leading to problems such as false alarms or inaccurate temperature ranges. Therefore, the present invention establishes a temperature prediction error model to learn the temperature prediction error pattern within the prediction time period, obtains a matching prediction error based on the temperature prediction results, and thus corrects the temperature range established based on the probability distribution at each moment to adapt to the different operating modes of the cascade utilization energy storage system;

[0125] When the temperature prediction error model learns the temperature prediction error, the input data of the temperature prediction error model includes: the temperature within the prediction period; the input data of the temperature prediction error model satisfies the following relationship:

[0126]

[0127] Where, is the input data of the temperature prediction error model, Establish moments for the sliding window, is the prediction window length, from time To time Constitute the prediction window, 、 、 、……、 is the temperature within the prediction window;

[0128] The output data of the temperature prediction error model includes: the absolute value of the temperature prediction error within the prediction time period; the output data of the temperature prediction error model satisfies the following relationship:

[0129]

[0130] Where, is the output data of the temperature prediction error model, Establish moments for the sliding window, is the prediction window length, from time To time Constitute the prediction window, 、 、 、……、 is the absolute value of the temperature prediction error within the prediction window.

[0131] Step 5: Use the historical values ​​of temperature and the historical values ​​of temperature-related monitoring quantities to train the temperature prediction model and the temperature prediction error model, and set the parameters of the graph neural network in the trained temperature prediction model as the initial parameters of the graph neural network in the temperature probability distribution prediction model; train the temperature probability distribution prediction model after the initial parameters are set, and according to the temperature probability distribution in the prediction window output by the trained temperature probability distribution prediction model, the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution in the prediction window output by the trained temperature prediction error model.

[0132] Specifically, the temperature prediction model and the temperature prediction error model are trained using the historical values ​​of temperature and the historical values ​​of temperature-related monitoring variables, including:

[0133] The historical values ​​of the temperature in the sliding window and the historical values ​​of the temperature-related monitoring quantity are used as the first training data set, and the historical values ​​of the temperature in the prediction window are used as the first label data;

[0134] The temperature prediction model is trained using the first training data set and the first label data. In an embodiment, the temperature within the prediction window output by the temperature prediction model during the training process is a deterministic estimate of the temperature historical value within the prediction window, and has a corresponding relationship.

[0135] The temperature within the prediction window output by the temperature prediction model constitutes a second training data set, and the absolute value of the difference between the temperature within the prediction window and the corresponding first label data is used as the second label data;

[0136] The temperature prediction error model is trained using the second training data set and the second label, and the trained temperature prediction error model outputs an absolute value of the prediction error of the temperature within the prediction window;

[0137] Through the above training process, the parameters of the temperature prediction model and the parameters of the temperature prediction error model are optimized; at this time, the parameters of the graph neural network in the trained temperature prediction model are extracted and set as the initial parameters of the graph neural network in the temperature probability distribution prediction model; the characteristics of the same graph neural network structure in the temperature prediction model and the temperature probability distribution prediction model are fully utilized, and in the embodiment, the training of the temperature prediction model and the temperature prediction error model can be carried out offline. After the transfer and setting of the neural network parameters, the uncertainty estimation mechanism is introduced through the last layer of Bayes without changing the topological structure of the neural network. Therefore, the training process of the temperature prediction model and the temperature prediction error model is actually also the initial training process of the temperature probability distribution prediction model, and in subsequent training, only the temperature probability distribution prediction model needs to be fine-tuned, which is beneficial to reduce the amount of calculation and improve the model training efficiency.

[0138] Specifically, the temperature probability distribution prediction model after the initial parameter setting is trained using the first training data set and the first label data, and the temperature probability distribution within the prediction window is output by the trained temperature probability distribution prediction model. The absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window is output by the trained temperature prediction error model; in an embodiment, the output temperature probability distribution within the prediction window is an uncertainty estimate of the temperature historical value within the prediction window, and also has a corresponding relationship.

[0139] Step 6: Establish a temperature interval model using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window, and the scaling factor; when the confidence mean of the temperature interval is equal to the set threshold, determine the optimal value of the scaling factor.

[0140] Specifically, the temperature range model satisfies the following relationship:

[0141]

[0142] Where, and are the upper and lower bound correction values ​​of the temperature probability distribution within the prediction window, respectively. is the mean temperature within the prediction window, is the variance of temperature within the prediction window, is the absolute value of the prediction error of the temperature probability distribution within the prediction window, is the scaling factor.

[0143] The temperature upper limit correction value and the temperature lower limit correction value constitute a temperature interval, and the confidence mean of the temperature interval satisfies the following relationship:

[0144]

[0145] Where, is the confidence mean of the temperature interval, is the total number of groups of temperature upper limit correction value and temperature lower limit correction value, is the jth label value, is the jth indicator function, when It takes 1 when it is within the temperature range, and 0 otherwise.

[0146] When the confidence mean is equal to the set confidence level, the scaling factor is determined In the embodiment, the confidence level is set to 0.95.

[0147] Step 7: Obtain the monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in operation, and use the temperature probability distribution prediction model, the temperature prediction error model, and the optimal value of the scaling factor to obtain the safe temperature range of the cascade utilization energy storage system.

[0148] The monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascaded utilization energy storage system in operation are real-time operating data; if the actual monitored temperature is outside the predicted safe temperature range, it indicates that the cascaded utilization energy storage system may be in an abnormal operating state and an alarm needs to be issued.

[0149] Based on the current moment Establish a lookback window of length m, starting from time To time Constitute the lookback window, from time To time The forecast period, is the length of the forecast period;

[0150] The monitoring values ​​of the temperature and temperature-related monitoring quantities within the lookback window are obtained and input into the temperature probability distribution prediction model to output the temperature probability distribution of the prediction period; when the temperature probability distribution of the prediction period obeys the Gaussian distribution, the temperature mean of the prediction period is determined; the temperature mean of the prediction period is input into the temperature prediction error model to obtain the absolute value of the prediction error of the temperature mean of the prediction period; the temperature mean of the prediction period, the absolute value of the prediction error of the temperature mean of the prediction period and the optimal value of the scaling factor are used to determine the temperature upper limit correction value and the temperature lower limit correction value of the prediction period according to the temperature interval model to form a safe temperature interval for the prediction period.

[0151] When predicting the safe temperature range based on the real-time operating data of the cascaded energy storage system, the input data of the temperature prediction error model includes: the average temperature within the prediction period; the input data of the temperature prediction error model satisfies the following relationship:

[0152]

[0153] Where, is the input data of the temperature prediction error model, For the current moment, is the length of the prediction period, from time To time The forecast period, 、 、 、……、 is the mean temperature during the forecast period.

[0154] In the embodiment, the safe temperature range obtained by test data is as follows Figure 4 As shown, the figure consists of the prediction results based on multiple sets of test data. Figure 4 It can be seen that the final safe temperature range obtained has good accuracy in most cases, which can basically cover the maximum and minimum temperature values, effectively avoiding frequent false alarms.

[0155] The present invention also proposes a probabilistic prediction system for the safe temperature range of a cascade utilization energy storage system, comprising:

[0156] An acquisition module is used to obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring quantities;

[0157] The model building module is used to build a temperature prediction model based on the graph neural network and the fully connected layer, and a temperature probability distribution prediction model based on the graph neural network and the last Bayesian layer; and a temperature prediction error model is built using shallow machine learning methods;

[0158] A model training module is used to train the temperature prediction model and the temperature prediction error model using historical temperature values ​​and historical values ​​of temperature-related monitoring variables, and to set the parameters of the graph neural network in the trained temperature prediction model as the initial parameters of the graph neural network in the temperature probability distribution prediction model; the temperature probability distribution prediction model after the initial parameters are set is trained, and the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean in the prediction window based on the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model;

[0159] The scaling factor optimization module is used to establish a temperature interval model using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean within the prediction window, and the scaling factor; when the confidence mean of the temperature interval is equal to the set threshold, the optimal value of the scaling factor is determined;

[0160] The prediction module is used to obtain the monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in the operating state, and use the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor to obtain the safe temperature range of the cascade utilization energy storage system.

[0161] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0162] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punched card or raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0163] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0164] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, the state information of the computer-readable program instructions is used to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), so that the electronic circuit can execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A probabilistic prediction method for the safe temperature range of a cascaded energy storage system, characterized in that: include: Obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring quantities; A temperature prediction model was established based on a graph neural network and a fully connected layer, and a temperature probability distribution prediction model was established based on a graph neural network and the last Bayesian layer. A temperature prediction error model was established using a shallow machine learning method. The temperature prediction model and temperature prediction error model are trained using historical temperature values ​​and historical values ​​of temperature-related monitoring variables. The parameters of the graph neural network in the trained temperature prediction model are set as the initial parameters of the graph neural network in the temperature probability distribution prediction model. The temperature probability distribution prediction model with the initial parameters set is trained. Based on the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model, the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window. The temperature range model is established using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean within the prediction window, and the scaling factor, which satisfies the following relationship: Where, and are the upper and lower bound correction values ​​of the temperature probability distribution within the prediction window, respectively. is the mean temperature within the prediction window, is the variance of temperature within the prediction window, is the absolute value of the prediction error of the temperature probability distribution within the prediction window, is the scaling factor; The temperature upper limit correction value and the temperature lower limit correction value constitute a temperature interval, and the confidence mean of the temperature interval satisfies the following relationship: Where, is the confidence mean of the temperature interval, is the total number of groups of temperature upper limit correction value and temperature lower limit correction value, is the jth label value, is the jth indicator function, which is It takes 1 when it is in the temperature range, otherwise it takes 0; When the confidence mean of the temperature interval is equal to the set threshold, the optimal value of the scaling factor is determined; The monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in operation are obtained, and the safe temperature range of the cascade utilization energy storage system is obtained using the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor.

2. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 1 is characterized in that: Obtain the historical values ​​of the temperature of the cascaded energy storage system under normal operating conditions and the historical values ​​of temperature-related monitoring variables, including: Collect historical values ​​of different temperature-related monitoring quantities, including state of charge, current, voltage, and temperature; Calculate the correlation coefficients between the various associated monitoring quantities, and determine the monitoring quantities that have a correlation relationship based on the correlation coefficients; Divide the historical values ​​of temperature and temperature-related monitoring variables into sequences of different time periods, and detect suspected abnormal data in each time period sequence; If other monitoring quantities that are associated with the monitoring quantity to which the suspected abnormal data belongs also have suspected abnormal data at the same time node, the suspected abnormal data is determined to be abnormal state data; otherwise, with the suspected abnormal data as the center of the sliding window, it is determined whether the absolute value of the difference between the normalized value of the abnormal data and the mean of the normalized data in the sliding window is greater than the set abnormal threshold. If it is greater, the suspected abnormal data is determined to be abnormal state data; after eliminating the abnormal state data and performing interpolation and repair, together with the historical values ​​of the temperature, a multi-segment historical normal data time series is formed.

3. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 2 is characterized in that: Each period of normal historical data time series satisfies the following relationship: Where X is the historical normal data time series, 、 、 、……、 The historical value time series of the temperature is formed, 、 、 、……、 For the A time series of historical values ​​of temperature-related monitoring quantities, , is the total number of monitored quantities, is the length of each time series in the historical normal data time series X; In each period of historical normal data time series, the time nodes in the historical normal data time series are Establish the moment for the sliding window, is the sliding window length, is the prediction window length, from time To time Constructing a sliding window, from time To time Constitute the prediction window.

4. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 1 is characterized in that: The fully connected layer is the last layer of the temperature prediction model; The last Bayesian layer is the last layer of the temperature probability distribution prediction model; Moreover, the graph neural network in the temperature prediction model has the same structure as that in the temperature probability distribution prediction model.

5. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 3 is characterized in that: Based on the historical values ​​of temperature within the sliding window and the historical values ​​of temperature-related monitoring variables, a temperature prediction model is used to output the temperature of the cascade utilization energy storage system within the prediction window. The input data of the temperature prediction model satisfies the following relationship: Where, is the input data of the temperature prediction model, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities; The output data of the temperature prediction model satisfies the following relationship: Where, is the output data of the temperature prediction model, 、 、 、……、 is the temperature within the prediction window.

6. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 3 is characterized in that: Based on the historical values ​​of temperature within the sliding window and the historical values ​​of temperature-related monitoring variables, the temperature probability distribution prediction model is used to output the temperature probability distribution of the cascade utilization energy storage system within the prediction window; The input data of the temperature probability distribution prediction model satisfies the following relationship: Where, is the input data of the temperature probability distribution prediction model, 、 、 、……、 is the historical value of temperature in the sliding window, 、 、 、……、 is the historical value of the temperature-related monitoring quantity in the sliding window, , is the total number of temperature-related monitoring quantities.

7. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 6 is characterized in that: Set the temperature probability distribution within the prediction window to follow Gaussian distribution , the output data of the temperature probability distribution prediction model satisfies the following relationship: Where, is the output data of the temperature probability distribution prediction model, 、 、 、……、 is the mean temperature within the prediction window, 、 、 、……、 is the standard deviation of temperature within the prediction window.

8. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 5 is characterized in that: According to the temperature within the prediction window, the temperature prediction error model is used to output the absolute value of the prediction error of the temperature of the cascade utilization energy storage system within the prediction window; When the temperature prediction error model learns the temperature prediction error, the input data of the temperature prediction error model satisfies the following relationship: Where, is the input data of the temperature prediction error model, 、 、 、……、 is the temperature within the prediction window; The output data of the temperature prediction error model satisfies the following relationship: Where, is the output data of the temperature prediction error model, 、 、 、……、 is the absolute value of the temperature prediction error within the prediction window.

9. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 8, characterized in that: The temperature prediction model and the temperature prediction error model are trained using the historical values ​​of temperature and the historical values ​​of temperature-related monitoring variables, including: The historical values ​​of the temperature in the sliding window and the historical values ​​of the temperature-related monitoring quantity are used as the first training data set, and the historical values ​​of the temperature in the prediction window are used as the first label data; The temperature prediction model is trained using the first training data set and the first label data; during the training process, the temperature within the prediction window output by the temperature prediction model is a deterministic estimate of the temperature history value within the prediction window, and has a corresponding relationship; The temperature within the prediction window output by the temperature prediction model constitutes a second training data set, and the absolute value of the difference between the temperature within the prediction window and the corresponding first label data is used as the second label data; The temperature prediction error model is trained using the second training data set and the second label, and the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature within the prediction window.

10. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 9 is characterized in that: The temperature probability distribution prediction model after the initial parameter setting is trained using the first training data set and the first label data. According to the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model, the absolute value of the prediction error of the temperature mean corresponding to the temperature probability distribution within the prediction window is output by the trained temperature prediction error model.

11. The probabilistic prediction method for the safe temperature range of the cascade utilization energy storage system according to claim 1 is characterized in that: When the confidence mean is equal to the set confidence level, the scaling factor is determined The confidence level is set to 0.

95.

12. The probabilistic prediction method for the safe temperature range of a cascade utilization energy storage system according to claim 1, characterized in that: Based on the current moment Establish a lookback window of length m, starting from time To time Constitute the lookback window, from time To time The forecast period, is the length of the forecast period; Obtaining the monitored values ​​of the temperature and temperature-related monitoring variables within the lookback window and inputting them into the temperature probability distribution prediction model to output the temperature probability distribution for the prediction period; determining the temperature mean for the prediction period based on the temperature probability distribution for the prediction period; inputting the temperature mean for the prediction period into the temperature prediction error model to obtain the absolute value of the prediction error of the temperature mean for the prediction period; The input data of the temperature prediction error model satisfies the following relationship: Where, is the input data of the temperature prediction error model, 、 、 、……、 is the mean temperature during the forecast period; Using the temperature mean of the prediction period, the absolute value of the prediction error of the temperature mean of the prediction period and the optimal value of the scaling factor, according to the temperature interval model, the temperature upper limit correction value and the temperature lower limit correction value of the prediction period are determined to form a safe temperature interval for the prediction period.

13. A probabilistic prediction system for a safe temperature range of a cascaded energy storage system, used to implement the probabilistic prediction method for a safe temperature range of a cascaded energy storage system according to any one of claims 1 to 12, characterized in that: include: An acquisition module is used to obtain historical values ​​of the temperature of the cascade utilization energy storage system under normal operating conditions and historical values ​​of temperature-related monitoring quantities; The model building module is used to build a temperature prediction model based on the graph neural network and the fully connected layer, and a temperature probability distribution prediction model based on the graph neural network and the last Bayesian layer; and a temperature prediction error model is built using shallow machine learning methods; A model training module is used to train the temperature prediction model and the temperature prediction error model using historical temperature values ​​and historical values ​​of temperature-related monitoring variables, and to set the parameters of the graph neural network in the trained temperature prediction model as the initial parameters of the graph neural network in the temperature probability distribution prediction model; the temperature probability distribution prediction model after the initial parameters are set is trained, and the trained temperature prediction error model outputs the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean in the prediction window based on the temperature probability distribution within the prediction window output by the trained temperature probability distribution prediction model; The scaling factor optimization module is used to establish a temperature range model using the temperature probability distribution within the prediction window, the absolute value of the prediction error of the temperature probability distribution corresponding to the temperature mean within the prediction window, and the scaling factor, satisfying the following relationship: Where, and are the upper and lower bound correction values ​​of the temperature probability distribution within the prediction window, respectively. is the mean temperature within the prediction window, is the variance of temperature within the prediction window, is the absolute value of the prediction error of the temperature probability distribution within the prediction window, is the scaling factor; The temperature upper limit correction value and the temperature lower limit correction value constitute a temperature interval, and the confidence mean of the temperature interval satisfies the following relationship: Where, is the confidence mean of the temperature interval, is the total number of groups of temperature upper limit correction value and temperature lower limit correction value, is the jth label value, is the jth indicator function, which is It takes 1 when it is in the temperature range, otherwise it takes 0; When the confidence mean of the temperature interval is equal to the set threshold, the optimal value of the scaling factor is determined; The prediction module is used to obtain the monitoring values ​​of the temperature and temperature-related monitoring quantities of the cascade utilization energy storage system in the operating state, and use the temperature probability distribution prediction model, the temperature prediction error model and the optimal value of the scaling factor to obtain the safe temperature range of the cascade utilization energy storage system.

14. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Images