UPS recession performance prediction method and system based on decision tree

Through the UPS decay performance prediction method based on the decision tree, combined with the convolutional neural network and the gradient-enhanced decision tree algorithm, the performance decay related features of the UPS device are extracted, and the problem of poor performance decay prediction effect in the existing technology is solved, and the implementation of the realization of the realization of UPS device is achieved. Higher prediction accuracy and maintenance efficiency.

CN118916793APending Publication Date: 2024-11-08STATE GRID JIBEI ELECTRIC POWER CO LTD TANGSHAN POWER SUPPLY CO +1

Patent Information

Application Number
CN202410952447.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The performance degradation problem of existing UPS equipment during use leads to a decrease in equipment reliability and stability. The existing prediction methods are poor in prediction and cannot detect and deal with potential problems in a timely manner.

Method used

UPS decay performance prediction method based on decision tree is adopted, and performance decay related features are extracted by collecting device information and historical operation data, and performance decay prediction is constructed by combining convolutional neural network-long and short-term memory models, recurrent neural networks and gradient-enhanced decision tree algorithms. Model, predict the degree of equipment performance decay in real time, and provide maintenance measures based on the predicted results.

Benefits of technology

It improves the accuracy and reliability of UPS equipment performance decay prediction, can promptly detect potential problems, take maintenance measures in advance, extend the service life of the equipment, and improve system reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118916793A_ABST
    Figure CN118916793A_ABST
Patent Text Reader

Abstract

The invention provides a decision tree-based UPS decline performance prediction method and system. The method comprises the steps of collecting equipment information and historical operation data of to-be-predicted UPS equipment; the collected equipment information and historical operation data are preprocessed; extracting related characteristics of performance degradation of the UPS equipment as a data set; constructing a UPS equipment performance degradation prediction model based on a convolutional neural network-long and short-term memory model, a recurrent neural network model and a gradient boosting decision tree algorithm model, and training the UPS equipment performance degradation prediction model by using the data set; the trained UPS equipment performance degradation prediction model can be used for real-time prediction, equipment information and real-time operation data of to-be-predicted UPS equipment are collected at first, then preprocessing is carried out, relevant features of UPS equipment performance degradation are extracted and input into the trained UPS equipment performance degradation prediction model, and the performance degradation degree of the UPS equipment is predicted in real time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of uninterruptible power supply, and relates to intelligent management of UPS charging and discharging, and in particular to a UPS degradation performance prediction method and system based on a decision tree. Background Art

[0002] With the development of science and technology and the advancement of information construction, uninterruptible power supply systems (UPS) have been widely used in all walks of life. However, UPS equipment may experience performance degradation during use, affecting the reliability and stability of the equipment.

[0003] Patent 1: CN201420746357.X proposes a UPS battery remote intelligent charge and discharge management system, including an antenna, a modem connected to the antenna, at least one detection branch, and a capacity test module. The detection branch includes a battery pack and a detection switch connected in series. The battery pack is connected to the moving end of the detection switch, and the two fixed ends of the detection switch are respectively connected to the power supply output end and the detection end. The UPS maintenance method of this patent is post-maintenance, which cannot detect and handle equipment problems in time, bringing hidden dangers to production and safety.

[0004] Patent 2: CN202311570735.3 proposes a UPS maintenance and status monitoring method and system. When the preset maintenance cycle is reached, the controllable switch of the AC input circuit is controlled to be disconnected, and the battery pack voltage is monitored; the battery pack voltage is compared with the preset discharge lower limit value, and the controllable switch of the AC input circuit is controlled to be closed based on the comparison result, and the AC power is switched back to the AC power in time, and the battery pack is charged; the battery pack voltage and / or charging current are monitored, and the battery pack charging is stopped based on the battery pack voltage and / or charging current; the battery maintenance data of the current maintenance cycle is converted into a waveform graph, and the waveform graph is analyzed for abnormal curves based on historical maintenance data to obtain the status monitoring results of the battery pack, so as to realize online monitoring of the health status of the battery; the patent uses relatively little prediction data, only the battery pack voltage and charging current, and the abnormality is found by curve analysis, which leads to the fact that the battery abnormality cannot be predicted when it is not severe, but the abnormality occurs before the next preset maintenance cycle, so the prediction effect is not good.

[0005] Patent 3: CN202211521104.8 proposes a method for predicting the health status of a rail transit UPS battery, conducts charge and discharge cycle experiments on the battery, and obtains performance parameter data under different working conditions; preprocesses the data, eliminates abnormal data during the charge and discharge process, and performs variational mode decomposition VMD on the capacity data; eliminates the noise data generated by capacity regeneration; extracts battery timing features through a bidirectional long short-time neural network BiLSTM; introduces an Attention layer after the network hidden layer to strengthen the weight of the key features of the battery, obtains the battery capacity measurement results through the full connection layer, and calculates the battery health status index SOH based on the capacity prediction results; This patent uses a single neural network to predict different types of data, resulting in poor reliability. Summary of the invention

[0006] In order to solve the problem of poor prediction reliability in the prior art, the present invention provides a UPS degradation performance prediction method and system based on a decision tree.

[0007] The present invention adopts the following technical solution.

[0008] On one hand, the present invention discloses a UPS degradation performance prediction method based on a decision tree, characterized in that the method comprises the following steps:

[0009] Step 1: Collect the equipment information and historical operation data of the uninterruptible power supply UPS device to be predicted;

[0010] Step 2: Preprocess the collected equipment information and historical operation data;

[0011] Step 3: Extract UPS equipment performance degradation related features as a data set;

[0012] Step 4: Build a UPS equipment performance degradation prediction model based on the convolutional neural network-long short-term memory model, the recurrent neural network model, and the gradient boosting decision tree algorithm model, and use the data set to train the UPS equipment performance degradation prediction model;

[0013] Step 5: Collect the equipment information and real-time operation data of the UPS device to be predicted, pre-process the data according to steps 2-3, and extract the relevant features of UPS device performance degradation and input them into the UPS device performance degradation prediction model trained in step 4 to predict the performance degradation degree of the UPS device in real time.

[0014] More preferably,

[0015] In step 1, the device information of the UPS device collected includes rated load, rated input voltage and rated output voltage;

[0016] Historical operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS in a week, the most recent continuous operation time of the UPS, the number of charge and discharge times of the UPS in a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status;

[0017] The battery condition includes the result of the last battery health status assessment and the time since the last battery health status assessment. The battery health status is marked through regular performance tests, marked as 2 healthy, 1 sub-healthy, and 0 damaged.

[0018] More preferably,

[0019] In step 2, One-Hot is used to encode the device type connected to the UPS device; then the outliers in all the data are cleaned and the missing values ​​are filled.

[0020] More preferably,

[0021] In step 3, the extracted UPS device performance degradation related characteristics include UPS operation characteristic values, electrical related parameter characteristic values, load rate, voltage fluctuation range, current fluctuation range, ambient temperature, ambient humidity, the last battery health status assessment result, the time since the last battery health status assessment and the type of equipment connected to the UPS.

[0022] More preferably,

[0023] The calculation formula of the operating characteristic value of the UPS is as follows:

[0024] C=R K r k +H i h i +E r e r +J f j f

[0025] Among them, C is the operating characteristic value of UPS, R K is the cumulative running time of the UPS in a week, r k is the weight of the accumulated running time of UPS in a week; H i is the continuous operation time of UPS this time, h i is the weight of the UPS continuous operation time this time; E r is the last continuous operation time of the UPS, e r is the weight of the last continuous operation time of UPS; J f is the number of charge and discharge times of UPS in a week, jf is the weight of the number of charge and discharge times of the UPS in a week; k 、h i 、e r 、j f Satisfies the following relationship:

[0026] r k +h i +e r +j f =1;

[0027] in, j f The value is 0.3~0.5, h i The value is 1.2e r ~1.5e r .

[0028] More preferably,

[0029] The calculation formula of the characteristic value of the electrical related parameters is as follows:

[0030] E L =V id v ID +V od v OD +(P i -P o ) L +Q n T n q L

[0031] Among them, E L is the characteristic value of electrical parameters; V id is the difference between the input voltage during operation and the rated input voltage, v ID is the input voltage deviation weight; V od is the difference between the output voltage during operation and the rated output voltage, v OD is the output voltage deviation weight; P i is the input power, P o is the output power, p L is the weight of the difference between input and output power; Q n is the number of devices connected to the UPS, T n is the battery temperature, q L is the weight of the number of devices connected to the UPS and the battery temperature; where v ID 、v OD 、p L ,q L Satisfies the following relationship:

[0032] v ID +vOD +p L +q L =1;

[0033] Input voltage deviation weight v ID Calculated using the following formula:

[0034]

[0035] in,

[0036] max() is the maximum value function, V max is the maximum input voltage in the historical operation data, V min is the minimum input voltage in the historical operation data, V ie is the rated input voltage value;

[0037] Output voltage deviation weight v OD =0.5-v ID ;

[0038] The weight p of the difference between input and output power L The value ranges from 0.1 to 0.25;

[0039] The weights q of the number of devices connected to the UPS and the battery temperature L =0.5-p L .

[0040] More preferably,

[0041] The load rate calculation formula is as follows:

[0042]

[0043] Where, L is the load factor, l o is the load, l e is the rated load;

[0044] The voltage fluctuation range is the maximum value of the input voltage fluctuation and the output voltage fluctuation within a fixed time period;

[0045] The current fluctuation range is the maximum value of the input current fluctuation and the output current fluctuation within a fixed time period.

[0046] More preferably,

[0047] In step 3, the data set also includes labels corresponding to performance degradation related features, and the labels include 0, 1, 2 and 3; among which 0 represents normal, 1 represents the early stage of performance degradation, 2 represents the middle stage of performance degradation, and 3 represents the late stage of performance degradation.

[0048] More preferably,

[0049] In step 4, the convolutional neural network-long short-term memory model CNN-LSTM, the recurrent neural network model RNN and the gradient boosting decision tree algorithm model GBDT are combined into a UPS equipment performance degradation prediction model;

[0050] The input of the convolutional neural network-long short-term memory model is the load factor, the types of various equipment connected to the UPS, and the characteristic values ​​of electrical related parameters;

[0051] The input of the recurrent neural network model is the operating characteristic value, voltage fluctuation range and current fluctuation range of the UPS;

[0052] The input of the gradient boosting decision tree algorithm model is the last battery health status assessment result, the time since the last battery health status assessment, the ambient temperature and the ambient humidity.

[0053] More preferably,

[0054] Step 5 includes the following steps:

[0055] 5.1 Use the trained UPS equipment performance degradation prediction model to make real-time predictions, and use the prediction algorithm to calculate the final prediction results:

[0056] 5.2 Determine whether the final prediction result is 0. If not, remind the user to adopt the selected maintenance measures for maintenance according to the final prediction result.

[0057] More preferably,

[0058] In step 5.1, the prediction result of the convolutional neural network-long short-term memory model is recorded as the first prediction result z r1 , the prediction result of the recurrent neural network model is recorded as the second prediction result z r2 , the prediction result of the gradient boosting decision tree algorithm model is recorded as the third prediction result z r3 , the prediction algorithm is used to obtain the final prediction result of the UPS equipment performance degradation prediction model according to the first prediction result, the second prediction result and the third prediction result. The formula of the prediction algorithm is as follows:

[0059]

[0060] Among them, P k is the probability of event k occurring, where the values ​​of k are 0, 1, 2, and 3, corresponding to the four labels in the data set; z k is the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k; e is the base of the natural logarithm;

[0061] The calculation formula of the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k is as follows:

[0062]

[0063] Among them, λ1 is the prediction relative accuracy of the first prediction result, λ2 is the prediction relative accuracy of the second prediction result; λ3 is the prediction relative accuracy of the third prediction result;

[0064] λ i The calculation formula is as follows, i is 1, 2, 3;

[0065]

[0066] Among them, z ri is the i-th prediction result, λ i is the relative accuracy of the prediction of the i-th prediction result, is the mean of the first prediction result, the second prediction result and the third prediction result, m z is the maximum value of the absolute value obtained by subtracting any two of the first prediction result, the second prediction result, and the third prediction result;

[0067] Calculate the probability P of event k occurring through the prediction algorithm k , select the one with the highest probability of occurrence k, and record it as the risk probability P lk , and its corresponding k value is recorded as the final prediction result k L ; In particular, when m z =0, the first prediction result is recorded as the final prediction result, and the risk probability is recorded as 1.

[0068] More preferably,

[0069] In step 5.2, when the final prediction result is 1 or 2, the maintenance measures are selected as follows: strengthen heat dissipation, reduce UPS load, and perform charging and discharging operations to prevent failures; when the power of the UPS when it was last shut down is not within the rated power range, give priority to powering the set key loads; when the battery voltage is too high, disconnect the charging circuit;

[0070] When the final prediction result is 2, maintenance measures also include dust removal, static electricity protection, dehumidification, and controlling the internal temperature of the UPS to no more than 45 degrees, and asking professionals to regularly maintain the UPS;

[0071] When the final prediction result is 3, the maintenance measure selected is to immediately shut down the UPS and replace the UPS device.

[0072] On the other hand, the present application claims protection for a UPS degradation performance prediction system based on a decision tree using the above-mentioned UPS degradation performance prediction method, characterized in that it comprises a UPS device (1), a UPS charging and discharging intelligent management data acquisition module (2), a control module (3), a UPS charging and discharging intelligent management data analysis and processing module (4), a UPS charging and discharging intelligent management feature extraction module (5), a UPS charging and discharging intelligent management data network analysis module (6), a data backup and recovery module (7) and a user interface module (8):

[0073] The UPS device (1) is a device to be detected;

[0074] The UPS charging and discharging intelligent management data acquisition module (2) collects device information and real-time operation data of the UPS device from the UPS device (1), and then transmits the data to the control module (3) and the UPS charging and discharging intelligent management data analysis and processing module (4); wherein the device information of the UPS device includes the rated load, the rated input voltage and the rated output voltage; the real-time operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS within a week, the most recent continuous operation time of the UPS, the number of UPS charging and discharging times within a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status; the battery status includes the last battery health status assessment result and the time since the last battery health status assessment; wherein the battery health status is marked through regular performance testing and expert evaluation, and is marked as 2 healthy, 1 sub-healthy and 0 damaged;

[0075] The UPS charging and discharging intelligent management data analysis and processing module (4) uses One-Hot to encode the device type connected to the UPS device for the device information and real-time operation data of the UPS device; then cleans the abnormal values ​​in all the data and fills in the missing values;

[0076] The UPS charging and discharging intelligent management feature extraction module (5) extracts UPS equipment performance degradation related features from the data preprocessed by the UPS charging and discharging intelligent management data analysis and processing module (4), and then transmits the extracted UPS equipment performance degradation related features to the UPS charging and discharging intelligent management data network analysis module (6);

[0077] The UPS charging and discharging intelligent management data network analysis module (6) inputs the features related to the UPS equipment performance degradation into the trained UPS equipment performance degradation prediction model for real-time prediction, obtains the final prediction result, determines whether the final prediction result is 0, and if not, provides the selected maintenance measures according to the final prediction result; then transmits the obtained final prediction result and the selected maintenance measures to the control module (3); wherein the UPS equipment performance degradation prediction model is composed of a convolutional neural network-long short-term memory model, a recurrent neural network model and a gradient boosting decision tree algorithm model;

[0078] The control module (3) controls the data backup and recovery module (7) to back up or restore the device information, real-time operation data, and the final prediction result of the performance degradation of the UPS device;

[0079] The user interface module (8) reads and displays the device information, real-time operation data and selected maintenance measures of the UPS device in the control module (3); the user can maintain the UPS device according to the maintenance measures provided by the user interface module (8).

[0080] More preferably,

[0081] A UPS charging and discharging intelligent management data network analysis module (6), comprising a UPS equipment performance degradation prediction unit (9) and a UPS charging and discharging intelligent management decision unit (10);

[0082] The UPS device performance degradation prediction unit (9) inputs the UPS device performance degradation related features extracted by the UPS charge and discharge intelligent management feature extraction module (5) into the trained UPS device performance degradation prediction model for real-time prediction to obtain a final prediction result, and transmits the final prediction result to the UPS charge and discharge intelligent management decision unit (10);

[0083] The UPS charging and discharging intelligent management decision unit (10) determines whether the final prediction result is 0, and if not 0, provides selected maintenance measures according to the final prediction result and transmits them to the control module (3).

[0084] On the other hand, the present application discloses an electronic device, including a processor and a storage medium; the characteristics are:

[0085] The storage medium is used to store instructions;

[0086] The processor is configured to operate according to the instructions to execute the UPS degradation performance prediction method based on the decision tree.

[0087] The present application also discloses a computer-readable storage medium having a computer program stored thereon, wherein the computer program is characterized in that when the program is executed by a processor, the decision tree-based UPS degradation performance prediction method is implemented.

[0088] The beneficial effect of the present invention is that, compared with the prior art, the device information and historical operation data of the UPS device are collected in the scheme of the present invention; the collected device information of the UPS device includes the rated load, the rated input voltage and the rated output voltage; the historical operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS within a week, the most recent continuous operation time of the UPS, the number of charge and discharge times of the UPS within a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery condition. The battery condition includes the last battery health status assessment result and the time from the last battery health status assessment; the battery health status is marked by regular performance testing and expert evaluation, and is marked as 2 healthy, 1 sub-healthy, and 0 damaged. After feature extraction of the collected data, the UPS device performance degradation prediction model is trained. The UPS device performance degradation prediction model of the present invention uses CNN-LSTM, RNN and GBDT to perform combined prediction of UPS device performance degradation, so that the final prediction result obtained by using the UPS device performance degradation prediction model for real-time data after training has high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 It is a flow chart of the UPS degradation performance prediction method based on the decision tree of the present invention;

[0090] Figure 2 Schematic diagram of a UPS degradation performance prediction system based on a decision tree according to the present invention. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in 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, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0092] Embodiment 1:

[0093] The embodiment of the present invention provides a UPS degradation performance prediction method based on a decision tree, see the attached Figure 1It includes collecting UPS data, data preprocessing, feature extraction, combining CNN-LSTM, RNN and GBDT to build a UPS equipment performance degradation prediction model and real-time collection of UPS data for prediction.

[0094] Step 1: Collect the equipment information and historical operation data of the uninterruptible power supply UPS device to be predicted.

[0095] In this step, historical UPS operation data that can reflect and predict UPS performance degradation are collected, with a sampling interval of 1 second.

[0096] The historical operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS in a week, the most recent continuous operation time of the UPS, the number of charge and discharge times of the UPS in a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status. The battery status includes the last battery health status assessment result and the time since the last battery health status assessment; the battery health status is marked through regular performance tests, marked as 2 healthy, 1 sub-healthy, and 0 damaged.

[0097] The device information includes rated load, rated input voltage and rated output voltage.

[0098] The above data can directly reflect the working status and health status of the UPS, and thus serve as training samples for the neural network.

[0099] Step 2: Preprocess the device information and historical operation data of the UPS device collected in step 1, that is, use One-Hot to encode the device type connected to the UPS device and then clean the abnormal values ​​and fill the missing values ​​in all the data to make it more suitable for subsequent processing;

[0100] 2.1 Use One-Hot to encode the device type connected to the UPS device collected in step 1;

[0101] 2.2 After encoding the device type connected to the UPS device in step 2.1, clean the outliers in all the data.

[0102] The method for cleaning outliers in any data is as follows:

[0103] when

[0104]

[0105] When

[0106] Among them, Z (h,t) is the data of the hth day at time t; n is the number of data sample points collected in one day, i is a natural number between 1 and n, and Z (h,i) is the data of the hth day and the i-th sampling time;

[0107] when

[0108]

[0109] When

[0110] 2.3 After cleaning all the outliers in the data in step 2.2, fill in the missing values.

[0111] The missing value filling is to fill the missing value with the average value of the corresponding data of the previous day.

[0112] Step 3: Extract UPS equipment performance degradation related features from the preprocessed data, then normalize the extracted features respectively, and then set labels for the normalized features; use the normalized UPS equipment performance degradation related features and corresponding labels to construct a data set.

[0113] 3.1 Extract features related to UPS equipment performance degradation from the preprocessed data.

[0114] The characteristics related to the performance degradation of the UPS device include UPS operation characteristic values, electrical-related parameter characteristic values, load rate, voltage fluctuation range, current fluctuation range, ambient temperature, ambient humidity, the last battery health status assessment result, the time since the last battery health status assessment and the type of equipment connected to the UPS.

[0115] The calculation formula of the operating characteristic value of the UPS is as follows:

[0116] C=R K r k +H i h i +E r e r +J f j f

[0117] Among them, C is the operating characteristic value of UPS, R K is the cumulative running time of the UPS in a week, r k is the weight of the accumulated running time of UPS in a week; H i is the continuous operation time of UPS this time, h i is the weight of the UPS continuous operation time this time; E r is the last continuous operation time of the UPS, er is the weight of the last continuous operation time of UPS; J f is the number of charge and discharge times of UPS in a week, j f is the weight of the number of charge and discharge times of the UPS in a week; k 、h i 、e r 、j f Satisfies the following relationship:

[0118] r k +h i +e r +j f =1;

[0119] in, j f The value is 0.3~0.5, h i The value is 1.2e r ~1.5e r .

[0120] The calculation formula of the characteristic value of the electrical related parameters is as follows:

[0121] E L =V id v ID +V od v OD +(P i -P o ) L +Q n T n q L

[0122] Among them, E L is the characteristic value of electrical parameters; V id is the difference between the input voltage during operation and the rated input voltage, v ID is the input voltage deviation weight; V od is the difference between the output voltage during operation and the rated output voltage, v OD is the output voltage deviation weight; P i is the input power, P o is the output power, p L is the weight of the difference between input and output power; Q n is the number of devices connected to the UPS, T n is the battery temperature, q L is the weight of the number of devices connected to the UPS and the battery temperature; where v ID 、v OD 、p L ,q L Satisfies the following relationship:

[0123] v ID +v OD +p L +q L =1;

[0124] Input voltage deviation weight v ID Calculated using the following formula:

[0125]

[0126] in,

[0127] max() is the maximum value function, V max is the maximum input voltage in the historical operation data, V min is the minimum input voltage in the historical operation data, V ie is the rated input voltage value;

[0128] Output voltage deviation weight v OD =0.5-v ID ;

[0129] The weight p of the difference between input and output power L The value ranges from 0.1 to 0.25;

[0130] The weights q of the number of devices connected to the UPS and the battery temperature L =0.5-p L .

[0131] The load rate calculation formula is as follows:

[0132]

[0133] Where, L is the load factor, l o is the load, l e is the rated load;

[0134] The voltage fluctuation range is the maximum value of the input voltage fluctuation and the output voltage fluctuation within a fixed time period; wherein the input voltage fluctuation is the input voltage value minus the average value of the input voltage within a fixed time interval, and the output voltage fluctuation is the output voltage value minus the average value of the output voltage within a fixed time interval.

[0135] The current fluctuation range is the maximum value of the input current fluctuation and the output current fluctuation within a fixed time period; wherein the input current fluctuation is the input current value minus the average value of the input current within a fixed time interval, and the output current fluctuation is the output current value minus the average value of the output current within a fixed time interval.

[0136] 3.2 Normalize the performance degradation related features of each UPS device extracted in step 3.1 respectively.

[0137] The normalization formula is as follows:

[0138]

[0139] Among them, x′ is the normalized result, x is the original data, μ is the mean of the data, and σ is the standard deviation of the data.

[0140] 3.3 Setting labels for the UPS device performance degradation related features normalized in step 3.2;

[0141] The label is the target value or result corresponding to the feature. The UPS status corresponding to the normalized UPS device performance degradation related features is marked. The UPS status is divided into four types: 0, 1, 2, and 3; among them, 0 means normal, 1 means early performance degradation, 2 means mid-performance degradation, and 3 means late performance degradation. The UPS status will be used as the output value of the neural network for training.

[0142] 3.4 A data set is constructed using the normalized equipment performance degradation related features in step 3.2 and the labels of the corresponding normalized UPS equipment performance degradation related features set in step 3.3.

[0143] Step 4: Build a UPS equipment performance degradation prediction model based on the convolutional neural network-long short-term memory model, the recurrent neural network model, and the gradient boosting decision tree algorithm model, and use the data set to train the UPS equipment performance degradation prediction model;

[0144] 4.1 Construct a UPS equipment performance degradation prediction model;

[0145] Taking into account the complexity and diversity of UPS equipment performance degradation, an ensemble learning method is adopted to construct a UPS equipment performance degradation prediction model by combining multiple algorithms, including convolutional neural network-long short-term memory model CNN-LSTM, recurrent neural network model RNN and gradient boosting decision tree algorithm model GBDT.

[0146] Build a CNN-LSTM model, including the input layer, Conv1D layer, first Pooling layer, Dropout layer, LSTM layer, second Pooling layer, Dense layer, Softmax layer and output layer;

[0147] The input data first passes through the Conv1D layer to extract features from the time series data; the Conv1D layer is connected to the first Pooling layer to compress the features extracted by the Conv1D layer; the first Pooling layer is connected to the Dropout layer to reduce overfitting; the Dropout layer sends the extracted features to the LSTM layer, which effectively captures the temporal dependencies of these features, and is connected to the second Pooling layer to compress the data again; the second Pooling layer is connected to the Dense layer to integrate all feature representations into one value, and finally the Dense layer is connected to the Softmax layer to output the classification results;

[0148] The input of the CNN-LSTM model is the load rate in the training set, the various types of equipment connected to the UPS, and the electrical-related parameter feature values;

[0149] Build an RNN model whose input is the operating characteristic values, voltage fluctuation range, and current fluctuation range of the UPS in the training set;

[0150] The GBDT model is constructed. The input of the model is the last battery health assessment result in the training set, the time since the last battery health assessment, the ambient temperature, and the ambient humidity.

[0151] 4.2 Use the data set obtained in step 3 to train the UPS equipment performance degradation prediction model constructed in step 4.1;

[0152] Use the data set obtained in step 3.4 to simultaneously train the CNN-LSTM, RNN, and GBDT in the UPS equipment performance degradation prediction model, and calculate the prediction loss values ​​of the three models separately through the cross entropy loss function, so that the model pays more attention to the early prediction of performance degradation.

[0153] Step 5: Collect the equipment information and real-time operation data of the UPS device to be predicted, pre-process the data according to steps 2-3, and extract the relevant features of the UPS device performance degradation and input them into the UPS device performance degradation prediction model trained in step 4 to predict the performance degradation degree of the UPS device in real time; finally, determine whether the final prediction result is normal. If not, remind the user to adopt the selected maintenance measures to avoid equipment failure.

[0154] 5.1 Use the trained UPS equipment performance degradation prediction model to make real-time predictions, and use the prediction algorithm to calculate the final prediction results.

[0155] The real-time operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS in a week, the most recent continuous operation time of the UPS, the number of charge and discharge times of the UPS in a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status. The battery status includes the last battery health status assessment result and the time since the last battery health status assessment; the battery health status is marked through regular performance tests, marked as 2 healthy, 1 sub-healthy, and 0 damaged.

[0156] The prediction result of CNN-LSTM is recorded as the first prediction result z r1 , the prediction result of RNN is recorded as the second prediction result z r2 , record the prediction result of GBDT as the third prediction result z r3 .

[0157] The formula of the prediction algorithm is as follows:

[0158]

[0159] Among them, P k is the probability of event k occurring, and the values ​​of k are 0, 1, 2, and 3, which correspond to the four prediction results in the data set: normal, early stage of performance decline, mid-stage of performance decline, and late stage of performance decline; z k is the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k; e is the base of the natural logarithm.

[0160] The calculation formula of the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k is as follows:

[0161]

[0162] Among them, λ1 is the first prediction result z r1 The relative accuracy of the prediction, λ2 is the second prediction result z r2 The relative accuracy of the prediction; λ3 is the third prediction result z r3 The relative accuracy of the predictions;

[0163] λ i The calculation formula is as follows, i is 1, 2, 3;

[0164]

[0165] Among them, z ri is the i-th prediction result, λ i is the i-th prediction result z ri The relative accuracy of the predictions, is the first prediction result z r1 , the second prediction result z r2 and the third prediction result z r3 The mean value, m z is the first prediction result z r1 , the second prediction result z r2 and the third prediction result z r3 The maximum absolute value of any two subtracted from each other.

[0166] Calculate the probability P of event k occurring through the prediction algorithm k , select the one with the highest probability of occurrence k, and record it as the risk probability P lk , and its corresponding k value is recorded as the final prediction result k L ; In particular, when m z =0, the first prediction result is recorded as the final prediction result, and the risk probability is recorded as 1;

[0167] 5.2 Determine whether the final prediction result is 0. If not, remind the user to adopt the selected maintenance measures for maintenance according to the final prediction result.

[0168] In step 5.2, when the final prediction result is 1 or 2, the maintenance measures are selected as follows: strengthen heat dissipation, reduce UPS load, and perform charging and discharging operations to prevent failures; when the power of the UPS when it was last shut down is not within the rated power range, give priority to powering the set key loads; when the battery voltage is too high, disconnect the charging circuit;

[0169] When the final prediction result is 2, maintenance measures also include dust removal, static electricity protection, dehumidification, and controlling the internal temperature of the UPS to no more than 45 degrees, and asking professionals to regularly maintain the UPS;

[0170] When the final prediction result is 3, the maintenance measure selected is to immediately shut down the UPS and replace the UPS device.

[0171] The embodiment of the present invention also provides a system based on the UPS charging and discharging intelligent management method, see the attached Figure 2 , including UPS equipment 1, UPS charging and discharging intelligent management data acquisition module 2, control module 3, UPS charging and discharging intelligent management data analysis and processing module 4, UPS charging and discharging intelligent management feature extraction module 5, UPS charging and discharging intelligent management data network analysis module 6, data backup and recovery module 7 and user interface module 8; UPS charging and discharging intelligent management data network analysis module 6, including UPS equipment performance degradation prediction unit 9 and UPS charging and discharging intelligent management decision unit 10.

[0172] The present invention monitors the data of UPS equipment in real time, extracts features related to performance degradation, and uses the UPS charging and discharging intelligent management data network analysis module 6 to make predictions and decisions, so as to timely discover potential problems of the equipment, propose maintenance measures, and improve its operating efficiency.

[0173] The UPS device 1 is a device to be detected;

[0174] The UPS charging and discharging intelligent management data acquisition module 2 collects device information and real-time operation data of the UPS device from the UPS, and then transmits the data to the control module 3 and the UPS charging and discharging intelligent management data analysis and processing module 4.

[0175] Use sensors, interface circuits and other equipment to collect real-time operation data, and then collect the real-time operation data through a high-precision ADC analog-to-digital converter; the equipment information of the UPS device includes the rated load, rated input voltage and rated output voltage; the real-time operation data includes the UPS's current continuous operation time, the UPS's cumulative operation time within a week, the UPS's most recent continuous operation time, the number of UPS charges and discharges within a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery condition. The battery condition includes the last battery health status assessment result and the time since the last battery health status assessment; the battery health status is marked through regular performance tests and expert evaluations, marked as 2 healthy, 1 sub-healthy, and 0 damaged.

[0176] The UPS charging and discharging intelligent management data analysis and processing module 4 receives the device information and real-time operation data of the UPS device collected by the UPS charging and discharging intelligent management data acquisition module 2, and uses One-Hot to encode the device type connected to the UPS device; then cleans the abnormal values ​​in all the data and fills the missing values;

[0177] The UPS charging and discharging intelligent management feature extraction module 5 extracts features related to UPS device performance degradation from the data preprocessed by the UPS charging and discharging intelligent management data analysis and processing module 4; and then transmits the extracted UPS device performance degradation related features to the UPS charging and discharging intelligent management data network analysis module 6.

[0178] The UPS charging and discharging intelligent management data network analysis module 6 inputs the features related to the UPS equipment performance degradation into the trained UPS equipment performance degradation prediction model for real-time prediction, obtains the final prediction result, determines whether the final prediction result is 0, and if not, provides the selected maintenance measures according to the final prediction result; then transmits the obtained final prediction result and the selected maintenance measures to the control module 3. The UPS equipment performance degradation prediction model is composed of CNN-LSTM, RNN and GBDT.

[0179] The UPS charging and discharging intelligent management data network analysis module 6 includes a UPS equipment performance degradation prediction unit 9 and a UPS charging and discharging intelligent management decision unit 10 .

[0180] The UPS device performance degradation prediction unit 9 inputs the UPS device performance degradation related features extracted by the UPS charge and discharge intelligent management feature extraction module 5 into the trained UPS device performance degradation prediction model for real-time prediction to obtain the final prediction result, and transmits the final prediction result to the UPS charge and discharge intelligent management decision unit 10;

[0181] The UPS charging and discharging intelligent management decision unit 10 determines whether the final prediction result is 0. If not, it provides the selected maintenance measures according to the final prediction result and transmits them to the control module 3.

[0182] As the battery is used, its performance will gradually decline, manifested as decreased capacity, increased internal resistance, etc. In order to extend the battery life and improve the reliability of the system, it is necessary to make decisions based on different degradation performance states and adopt different maintenance measures.

[0183] When the final prediction result of the UPS device performance degradation prediction unit 9 shows that the UPS device may be about to fail, that is, the prediction result of the UPS device performance degradation prediction unit 9 is not 0, the UPS charge and discharge intelligent management decision unit 10 will make a decision and select maintenance measures. When the final prediction result is 1 or 2, the maintenance measures are selected as follows: strengthen heat dissipation, reduce UPS load, and perform charge and discharge operations to prevent failures; when the power of the UPS when it was last shut down is not within the rated power range, give priority to powering the set key loads; when the battery voltage is too high, disconnect the charging circuit;

[0184] When the final prediction result is 2, maintenance measures also include dust removal, static electricity protection, dehumidification, and controlling the internal temperature of the UPS to no more than 45 degrees, and asking professionals to regularly maintain the UPS;

[0185] When the final prediction result is 3, the maintenance measure selected is to immediately shut down the UPS and replace the UPS device.

[0186] The control module 3 controls the data backup and recovery module 7 to back up and restore the device information and real-time operation data of the UPS device, as well as the final prediction results of the UPS device performance degradation; this ensures that the UPS data can be backed up and restored in a timely manner, reducing the risk of data loss. Specifically, in this structure, the data backup and recovery module 7 can regularly back up the device information and real-time operation data of the UPS device, as well as the final prediction results of the UPS device performance degradation, and restore them when necessary, ensuring the integrity of the data and the normal operation of the system while ensuring the security of the UPS device data.

[0187] The user interface module 8 provides a visual interface to read and display the device information, real-time operating data and selected maintenance measures of the UPS device in the control module 3; the user can maintain the UPS device according to the maintenance measures provided by the user interface module (8); specifically, in this structure, the user interface module 8 provides an intuitive interface for the user, and the user can conveniently view the device information, real-time operating data and selected maintenance measures of the UPS device; and it is convenient for the user to perform corresponding maintenance operations based on this information.

[0188] The present disclosure may be a system, a method and / or a 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.

[0189] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may 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 of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted 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 through a fiber optic cable), or an electrical signal transmitted through a wire.

[0190] 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, optical fiber transmissions, wireless transmissions, 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 for storage in the computer-readable storage medium in each computing / processing device.

[0191] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related 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++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely 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., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0192] 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 relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A UPS degradation performance prediction method based on decision tree, characterized in that: The method comprises the following steps: Step 1: Collect the equipment information and historical operation data of the uninterruptible power supply UPS device to be predicted; Step 2: Preprocess the collected equipment information and historical operation data; Step 3: Extract UPS equipment performance degradation related features as a data set; Step 4: Build a UPS equipment performance degradation prediction model based on the convolutional neural network-long short-term memory model, the recurrent neural network model, and the gradient boosting decision tree algorithm model, and use the data set to train the UPS equipment performance degradation prediction model; Step 5: Collect the equipment information and real-time operation data of the UPS device to be predicted, pre-process the data according to steps 2-3, and extract the relevant features of UPS device performance degradation and input them into the UPS device performance degradation prediction model trained in step 4 to predict the performance degradation degree of the UPS device in real time.

2. The UPS degradation performance prediction method based on decision tree according to claim 1 is characterized in that: In step 1, the device information of the UPS device collected includes rated load, rated input voltage and rated output voltage; Historical operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS in a week, the most recent continuous operation time of the UPS, the number of charge and discharge times of the UPS in a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status; The battery status includes the result of the last battery health status assessment and the time since the last battery health status assessment; The battery health status is marked through regular performance tests, with 2 healthy, 1 sub-healthy, and 0 damaged.

3. The UPS degradation performance prediction method based on decision tree according to claim 1 is characterized in that: In step 2, One-Hot is used to encode the device type connected to the UPS device; then the outliers in all the data are cleaned and the missing values ​​are filled.

4. The UPS degradation performance prediction method based on decision tree according to claim 1 is characterized in that: In step 3, the extracted UPS device performance degradation related characteristics include UPS operation characteristic values, electrical related parameter characteristic values, load rate, voltage fluctuation range, current fluctuation range, ambient temperature, ambient humidity, the last battery health status assessment result, the time since the last battery health status assessment and the type of equipment connected to the UPS.

5. The UPS degradation performance prediction method based on decision tree according to claim 4 is characterized in that: The calculation formula of the operating characteristic value of the UPS is as follows: C=R K r k +H i h i +E r e r +J f j f Among them, C is the operating characteristic value of UPS, R K is the cumulative running time of the UPS in a week, r k is the weight of the accumulated running time of UPS in a week; H i is the continuous operation time of UPS this time, h i is the weight of the UPS continuous operation time this time; E r is the last continuous operation time of the UPS, e r is the weight of the last continuous operation time of UPS; J f is the number of charge and discharge times of UPS in a week, j f is the weight of the number of charge and discharge times of the UPS in a week; k 、h i 、e r 、j f Satisfies the following relationship: r k +h i +e r +j f =1; in, j f The value is 0.3~0.5, h i The value is 1.2e r ~1.5e r .

6. The UPS degradation performance prediction method based on decision tree according to claim 4 is characterized in that: The calculation formula of the characteristic value of the electrical related parameters is as follows: E L =V id v ID +V od v OD +(P i -P o )p L +Q n T n q L Among them, E L is the characteristic value of electrical parameters; V id is the difference between the input voltage during operation and the rated input voltage, v ID is the input voltage deviation weight; V od is the difference between the output voltage during operation and the rated output voltage, v OD is the output voltage deviation weight; P i is the input power, P o is the output power, p L is the weight of the difference between input and output power; Q n is the number of devices connected to the UPS, T n is the battery temperature, q L is the weight of the number of devices connected to the UPS and the battery temperature; where v ID 、v OD 、p L ,q L Satisfies the following relationship: v ID +v OD +p L +q L =1; Input voltage deviation weight v ID Calculated using the following formula: in, max() is the maximum value function, V max is the maximum input voltage in the historical operation data, V min is the minimum input voltage in the historical operation data, V ie is the rated input voltage value; Output voltage deviation weight v OD =0.5-v ID ; The weight p of the difference between input and output power L The value ranges from 0.1 to 0.25; The weights q of the number of devices connected to the UPS and the battery temperature L =0.5-p L .

7. The UPS degradation performance prediction method based on decision tree according to claim 4 is characterized in that: The load rate calculation formula is as follows: Where, L is the load factor, l o is the load, l e is the rated load; The voltage fluctuation range is the maximum value of the input voltage fluctuation and the output voltage fluctuation within a fixed time period; The current fluctuation range is the maximum value of the input current fluctuation and the output current fluctuation within a fixed time period.

8. The UPS degradation performance prediction method based on decision tree according to claim 1 is characterized in that: In step 3, the data set also includes labels corresponding to performance degradation related features, and the labels include 0, 1, 2 and 3; among which 0 represents normal, 1 represents the early stage of performance degradation, 2 represents the middle stage of performance degradation, and 3 represents the late stage of performance degradation.

9. The UPS degradation performance prediction method based on decision tree according to claim 1 is characterized in that: In step 4, the convolutional neural network-long short-term memory model CNN-LSTM, the recurrent neural network model RNN and the gradient boosting decision tree algorithm model GBDT are combined into a UPS equipment performance degradation prediction model; The input of the convolutional neural network-long short-term memory model is the load factor, the types of various equipment connected to the UPS, and the characteristic values ​​of electrical related parameters; The input of the recurrent neural network model is the operating characteristic value, voltage fluctuation range and current fluctuation range of the UPS; The input of the gradient boosting decision tree algorithm model is the last battery health status assessment result, the time since the last battery health status assessment, the ambient temperature and the ambient humidity.

10. The UPS degradation performance prediction method based on decision tree according to claim 1, characterized in that: Step 5 includes the following steps: 5.1 Use the trained UPS equipment performance degradation prediction model to make real-time predictions, and use the prediction algorithm to calculate the final prediction results: 5.2 Determine whether the final prediction result is 0. If not, remind the user to adopt the selected maintenance measures for maintenance according to the final prediction result.

11. The UPS degradation performance prediction method based on decision tree according to claim 10, characterized in that: In step 5.1, the prediction result of the convolutional neural network-long short-term memory model is recorded as the first prediction result z r1 , the prediction result of the recurrent neural network model is recorded as the second prediction result z r2 , the prediction result of the gradient boosting decision tree algorithm model is recorded as the third prediction result z r3 , the prediction algorithm is used to obtain the final prediction result of the UPS equipment performance degradation prediction model according to the first prediction result, the second prediction result and the third prediction result. The formula of the prediction algorithm is as follows: Among them, P k is the probability of event k occurring, where the values ​​of k are 0, 1, 2, and 3, corresponding to the four labels in the data set; z k is the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k; e is the base of the natural logarithm; The calculation formula of the degree to which the predicted value of the UPS equipment performance degradation prediction model deviates from k is as follows: Among them, λ1 is the prediction relative accuracy of the first prediction result, λ2 is the prediction relative accuracy of the second prediction result; λ3 is the prediction relative accuracy of the third prediction result; λ i The calculation formula is as follows, i is 1, 2, 3; Among them, z ri is the i-th prediction result, λ i is the relative accuracy of the prediction of the i-th prediction result, is the mean of the first prediction result, the second prediction result and the third prediction result, m z is the maximum value of the absolute value obtained by subtracting any two of the first prediction result, the second prediction result, and the third prediction result; Calculate the probability P of event k occurring through the prediction algorithm k , select the one with the highest probability of occurrence k, and record it as the risk probability P lk , and its corresponding k value is recorded as the final prediction result k L ; In particular, when m z =0, the first prediction result is recorded as the final prediction result, and the risk probability is recorded as 1.

12. The UPS degradation performance prediction method based on decision tree according to claim 10, characterized in that: In step 5.2, when the final prediction result is 1 or 2, the maintenance measures are selected as follows: strengthen heat dissipation, reduce UPS load, and perform charging and discharging operations to prevent failures; when the power of the UPS when it was last shut down is not within the rated power range, give priority to powering the set key loads; when the battery voltage is too high, disconnect the charging circuit; When the final prediction result is 2, maintenance measures also include dust removal, static electricity protection, dehumidification, and controlling the internal temperature of the UPS to no more than 45 degrees, and asking professionals to regularly maintain the UPS; When the final prediction result is 3, the maintenance measure selected is to immediately shut down the UPS and replace the UPS device.

13. A UPS degradation performance prediction system based on a decision tree using the UPS degradation performance prediction method according to any one of claims 1 to 12, characterized in that: The system comprises a UPS device (1), a UPS charging and discharging intelligent management data acquisition module (2), a control module (3), a UPS charging and discharging intelligent management data analysis and processing module (4), a UPS charging and discharging intelligent management feature extraction module (5), a UPS charging and discharging intelligent management data network analysis module (6), a data backup and recovery module (7) and a user interface module (8): The UPS device (1) is a device to be detected; The UPS charging and discharging intelligent management data acquisition module (2) collects device information and real-time operation data of the UPS device from the UPS device (1), and then transmits the data to the control module (3) and the UPS charging and discharging intelligent management data analysis and processing module (4); wherein the device information of the UPS device includes the rated load, the rated input voltage and the rated output voltage; the real-time operation data includes the current continuous operation time of the UPS, the cumulative operation time of the UPS within a week, the most recent continuous operation time of the UPS, the number of UPS charging and discharging times within a week, the load, the type of equipment connected to the UPS, the number of equipment connected to the UPS, the battery temperature, the ambient temperature, the ambient humidity, the input voltage, the output voltage, the input current, the output current, the input power, the output power and the battery status; the battery status includes the last battery health status assessment result and the time since the last battery health status assessment; wherein the battery health status is marked through regular performance testing and expert evaluation, and is marked as 2 healthy, 1 sub-healthy and 0 damaged; The UPS charging and discharging intelligent management data analysis and processing module (4) uses One-Hot to encode the device type connected to the UPS device for the device information and real-time operation data of the UPS device; then cleans the abnormal values ​​in all the data and fills the missing values; The UPS charging and discharging intelligent management feature extraction module (5) extracts UPS equipment performance degradation related features from the data preprocessed by the UPS charging and discharging intelligent management data analysis and processing module (4), and then transmits the extracted UPS equipment performance degradation related features to the UPS charging and discharging intelligent management data network analysis module (6); The UPS charging and discharging intelligent management data network analysis module (6) inputs the features related to the UPS equipment performance degradation into the trained UPS equipment performance degradation prediction model for real-time prediction, obtains the final prediction result, determines whether the final prediction result is 0, and if not, provides the selected maintenance measures according to the final prediction result; then transmits the obtained final prediction result and the selected maintenance measures to the control module (3); wherein the UPS equipment performance degradation prediction model is composed of a convolutional neural network-long short-term memory model, a recurrent neural network model and a gradient boosting decision tree algorithm model; The control module (3) controls the data backup and recovery module (7) to back up or restore the device information, real-time operation data, and the final prediction result of the performance degradation of the UPS device; The user interface module (8) reads and displays the device information, real-time operation data and selected maintenance measures of the UPS device in the control module (3); the user can maintain the UPS device according to the maintenance measures provided by the user interface module (8).

14. The UPS degradation performance prediction system based on decision tree according to claim 13, characterized in that: A UPS charging and discharging intelligent management data network analysis module (6), comprising a UPS equipment performance degradation prediction unit (9) and a UPS charging and discharging intelligent management decision unit (10); The UPS device performance degradation prediction unit (9) inputs the UPS device performance degradation related features extracted by the UPS charge and discharge intelligent management feature extraction module (5) into the trained UPS device performance degradation prediction model for real-time prediction to obtain a final prediction result, and transmits the final prediction result to the UPS charge and discharge intelligent management decision unit (10); The UPS charging and discharging intelligent management decision unit (10) determines whether the final prediction result is 0, and if not 0, provides selected maintenance measures according to the final prediction result and transmits them to the control module (3).

15. An electronic device, 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 UPS degradation performance prediction method based on a decision tree according to claims 1-12.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the UPS degradation performance prediction method based on a decision tree described in claims 1-12 is implemented.

Citation Information

Patent Citations

  • Rail transit UPS storage battery health state prediction method

    CN115730525A

  • UPS maintenance and state monitoring method and system

    CN117559627A

  • Remote intelligent charging and discharging management system for UPS storage battery

    CN204189917U

Cited By

  • Circuit assembly selecting and matching method and system and medium

    CN119397995A