Generator starting state detection method, device and equipment and storage medium

Abstract

The embodiment of the invention provides a generator starting state detection method and device, equipment and a storage medium. According to the method, the starting state of the generator can be detected based on a detection data set of the generator which is collected in the starting process and changes along with starting time. According to the method, each collected detection data group is divided to obtain a plurality of time windows, so that each time window is deepened, and according to the starting time of each detection data group in the time window, through a predetermined target mapping relation between the time and the voltage of the generator in a normal starting state, the voltage of the generator is determined to be higher than the starting time of each detection data group in the time window. The method comprises the following steps: determining a reference voltage value corresponding to each detection data group in a time window according to a preset time window, thereby determining a reference voltage value and an early warning threshold value based on each reference voltage value, and further analyzing an abnormal monitoring result of each detection data group based on the reference voltage value and the early warning threshold value. And fine dynamic changes of the detection data set in the starting process are captured by going deep into each time window, so that the starting state of the generator is detected.

Description

Technical Field

[0001] This application belongs to the field of generator status detection, and particularly relates to a method, device, equipment, and storage medium for detecting the starting status of a generator. Background Art

[0002] In the modern power protection system, generators are usually used as backup power supplies or emergency power supplies and are widely used in key facilities such as data centers, hospitals, and communication base stations. Its starting process includes receiving a start signal, generating torque through a starting motor to drive the crankshaft to rotate, completing the cycle of intake, compression, power generation, and exhaust, and finally outputting a stable voltage when the rotational speed rises to a preset rotational speed. Any abnormality in any link during the starting process of the generator may lead to starting failure. Therefore, in order to ensure rapid power restoration and maintain the normal operation of key equipment in case of an emergency, it is particularly important to monitor the starting status of the generator.

[0003] Currently, sensors are generally installed to monitor parameters such as temperature and rotational speed during the starting process of the generator in real time. Once these parameters exceed the warning threshold, it is considered that the generator has a risk of abnormal starting. However, parameters such as temperature and rotational speed can only reflect the surface status of the generator and cannot deeply reveal the abnormal hidden dangers existing during the starting process. Summary of the Invention

[0004] The embodiments of this application provide a method, device, equipment, and computer storage medium for detecting the starting status of a generator, which can reveal the abnormal hidden dangers existing during the starting process of the generator.

[0005] In a first aspect, the embodiments of this application provide a method for detecting the starting status of a generator, and the method includes:

[0006] Sending a start signal to the generator to be abnormally detected, and collecting a set of detection data on the output voltage of the generator changing with the starting time;

[0007] Dividing each set of detection data to obtain multiple time windows;

[0008] For each time window, according to the starting time of each set of detection data within the time window, through the target mapping relationship between time and voltage of the generator in the normal starting state determined in advance, determining the reference voltage value corresponding to each set of detection data within the time window;

[0009] For each time window, according to the reference voltage values corresponding to each set of detection data within the time window, determining the reference voltage value of the time window and the warning threshold of the time window;

[0010] For each group of detection data, determine the abnormal detection result of the group of detection data according to the output voltage of the group of detection data, the reference voltage value of the time window to which the group of detection data belongs, and the early warning threshold of the time window to which the group of detection data belongs;

[0011] Detect the starting state of the generator according to the abnormal detection results of each group of detection data.

[0012] In an implementable embodiment, the method further includes:

[0013] Obtain the nominal output voltage and ideal starting time of each model of the generator;

[0014] For each model, determine the initial mapping relationship between time and voltage of the generator of this model in the normal starting state according to the nominal output voltage of this model, the ideal starting time of this model, and the preset voltage change parameter;

[0015] For each model, collect a sample data group of the output voltage of the generator of this model changing with the starting time in the normal starting state;

[0016] Determine the predicted output voltage of each sample data group through the initial mapping relationship;

[0017] Determine the voltage fluctuation gradient according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group;

[0018] With the goal of minimizing the voltage fluctuation gradient, through the gradient descent method, iteratively update the initial mapping relationship until the preset stop condition is met, and obtain the target mapping relationship corresponding to this model.

[0019] In an implementable embodiment, determining the voltage fluctuation gradient according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group specifically includes:

[0020] For each sample data group, determine the deviation value of the sample data group according to the output voltage of the sample data group and the predicted output voltage of the sample data group;

[0021] Determine the total deviation value according to the deviation values of each sample data group;

[0022] Take the partial derivative of the total deviation value with respect to the voltage change parameter as the voltage fluctuation gradient.

[0023] In an implementable embodiment, with the goal of minimizing the voltage fluctuation gradient, through the gradient descent method, iteratively update the initial mapping relationship until the preset stop condition is met, and obtain the target mapping relationship corresponding to this model, specifically including:

[0024] According to the voltage fluctuation gradient, perform a first iterative update on the voltage change parameter, and based on the voltage change parameter after the first iterative update, determine the initial mapping relationship after the first iterative update, where the voltage change parameter is negatively correlated with the voltage fluctuation gradient;

[0025] Based on each sample data group, with the goal of minimizing the voltage fluctuation gradient, through the gradient descent method, perform multiple iterative updates on the initial mapping relationship after the first iterative update until a preset stop condition is met, and obtain the target mapping relationship corresponding to this model.

[0026] In an implementable embodiment, based on each sample data group, with the goal of minimizing the voltage fluctuation gradient, through the gradient descent method, perform multiple iterative updates on the initial mapping relationship after the first iterative update until a preset stop condition is met, and obtain the target mapping relationship corresponding to this model. Specifically, it includes:

[0027] For each iterative update in the multiple iterative updates, through the initial mapping relationship after the previous iterative update, determine the predicted output voltage of each sample data group in this iterative update;

[0028] For each iterative update in the multiple iterative updates, according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group in this iterative update, determine the voltage fluctuation gradient of this iterative update;

[0029] For each iterative update in the multiple iterative updates, according to the voltage fluctuation gradient of this iterative update, perform this iterative update on the voltage change parameter after the previous iterative update, and based on the voltage change parameter after this iterative update, determine the initial mapping relationship after this iterative update;

[0030] For each iterative update in the multiple iterative updates, determine whether the voltage fluctuation gradient of this iterative update is less than a preset gradient threshold;

[0031] If so, use the initial mapping relationship after this iterative update as the target mapping relationship corresponding to this model;

[0032] If not, continue with the next iterative update.

[0033] In an implementable embodiment, according to the start times of each detection data group within this time window, through the pre-determined target mapping relationship between the time and voltage of the generator in the normal start state, determine the reference voltage values corresponding to each detection data group within this time window. Specifically, it includes:

[0034] Determine the model of the generator to be abnormally detected as the target model;

[0035] From the target mapping relationships corresponding to each model, call the target mapping relationship corresponding to the target model.

[0036] According to the start times of the detection data groups within this time window, determine the reference voltage values corresponding to each detection data group within this time window through the target mapping relationship corresponding to the target model.

[0037] In an implementable embodiment, according to the reference voltage values corresponding to each detection data group within this time window, determine the reference voltage value of this time window and the warning threshold of this time window, specifically including:

[0038] Take the average of the reference voltage values corresponding to each detection data group within this time window to obtain the average reference voltage;

[0039] According to the reference voltage values corresponding to each detection data group within this time window and the average reference voltage, determine the standard deviation of the reference voltage;

[0040] Take the average reference voltage as the reference voltage value of this time window;

[0041] According to the standard deviation of the reference voltage and a preset threshold, determine the warning threshold of this time window.

[0042] In an implementable embodiment, according to the output voltage of this detection data group, the reference voltage value of the time window to which this detection data group belongs, and the warning threshold of the time window to which this detection data group belongs, determine the abnormal detection result of this detection data group, specifically including:

[0043] Determine the absolute difference between the output voltage of this detection data group and the reference voltage value of the time window to which this detection data group belongs;

[0044] When the absolute difference is greater than the warning threshold of the time window to which this detection data group belongs, the abnormal detection result of this detection data group is abnormal;

[0045] When the absolute difference is not greater than the warning threshold of the time window to which this detection data group belongs, the abnormal detection result of this detection data group is normal.

[0046] In an implementable embodiment, according to the abnormal detection results of each detection data group, detect the starting state of the generator, specifically including:

[0047] When the abnormal detection results of each detection data group are all normal, the generator is in a normal starting state;

[0048] When the abnormal detection results of each detection data group are not all normal, the generator is in an abnormal starting state.

[0049] In an implementable embodiment, the method includes dividing each detection data group to obtain a plurality of time windows, specifically including:

[0050] Responding to a division instruction from the user;

[0051] Dividing each detection data group according to the target quantity set by the division instruction to obtain a plurality of time windows.

[0052] In a second aspect, an embodiment of the present application provides a generator starting state detection device, including:

[0053] A data acquisition module, configured to send a start signal to the generator to be abnormally detected and acquire a detection data group of the output voltage of the generator changing with the start time;

[0054] An accuracy division module, configured to divide each detection data group according to a preset quantity to obtain a plurality of time windows;

[0055] A reference voltage module, configured to, for each time window, determine the reference voltage value corresponding to each detection data group in the time window according to the start time of each detection data group in the time window through a pre-determined target mapping relationship between time and voltage of the generator in a normal starting state;

[0056] An early warning parameter module, configured to, for each time window, determine the reference voltage value and the early warning threshold of the time window according to the reference voltage values corresponding to each detection data group in the time window;

[0057] A first detection module, configured to, for each detection data group, determine the abnormal detection result of the detection data group according to the output voltage of the detection data group, the reference voltage value of the time window to which the detection data group belongs, and the early warning threshold of the time window to which the detection data group belongs;

[0058] A target detection module, configured to detect the starting state of the generator according to the abnormal detection results of each detection data group.

[0059] In a third aspect, an embodiment of the present application provides a generator starting state detection device, including: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement any one of the above generator starting state detection methods.

[0060] Fourthly, an embodiment of the present application provides a computer storage medium. Computer program instructions are stored on the computer storage medium. When the computer program instructions are executed by a processor, the above-mentioned generator starting state detection method is implemented.

[0061] A generator starting state detection method, device, equipment and computer storage medium according to an embodiment of the present application can detect the starting state of the generator based on a set of detection data of the generator changing with the starting time collected during the starting process. By dividing the collected sets of detection data, multiple time windows are obtained, and then each time window is delved into. According to the starting time of each set of detection data within the time window, through a pre-determined target mapping relationship between time and voltage of the generator in the normal starting state, the reference voltage value corresponding to each set of detection data within the time window is determined, so as to analyze the difference between the output voltage of each set of detection data and the reference voltage value, capture the small dynamic changes during the starting process of the detection data set, and achieve the detection of the generator starting state.

[0062] In addition, the target mapping relationship between time and voltage of each model of generator in the normal starting state can be pre-determined, so as to call the corresponding model's target mapping relationship when detecting the starting state of the generator. At the same time, the reference voltage mean value and reference voltage standard deviation corresponding to the reference voltage value are determined, and then the reference voltage mean value is used as the reference voltage value, and an early warning threshold is determined based on the reference voltage standard deviation and a preset threshold, so as to analyze the fluctuation situation of each set of detection data based on the reference voltage value and the early warning threshold. The statistics (reference voltage value and early warning threshold) dynamically determined based on the actually collected sets of detection data improve the accuracy and reliability of the generator starting state detection. Description of the Drawings

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0064] Figure 1 is a flowchart of a generator starting state detection method provided by an embodiment of the present application;

[0065] Figure 2 is a flowchart of iteratively updating and determining the target mapping relationship provided by an embodiment of the present application;

[0066] Figure 3 is a schematic diagram of the principle of a generator starting state detection system provided by an embodiment of the present application;

[0067] Figure 4It is a schematic structural diagram of a generator startup state detection device provided by an embodiment of the present application;

[0068] Figure 5 It is a schematic structural diagram of a generator startup state detection device provided by an embodiment of the present application. Specific embodiments

[0069] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without some of these specific details. The following description of the embodiments is only intended to provide a better understanding of the present application by showing examples of the present application.

[0070] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, elements defined by the statement "comprising..." do not preclude the existence of additional identical elements in the process, method, article or device comprising the said elements.

[0071] In the technical solution of the present application, the specific application scenarios of a generator startup state detection method, device, storage medium and device provided by the embodiment are not strictly limited, and can be flexibly selected according to actual needs.

[0072] It should be noted that the application scenarios described in the present application are only for more clearly explaining the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the present application. Those of ordinary skill in the art will know that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems. A generator startup state detection method provided by an embodiment of the present application can be applied to various application scenarios that require generator state detection.

[0073] In the modern power security system, generators are usually used as backup power sources or emergency power sources and are widely applied in key facilities such as data centers, hospitals, and communication base stations. After receiving a start signal, the starting motor drives the crankshaft of the generator to rotate. Under the action of the rotating crankshaft, the generator starts the working cycle of intake, compression, power generation, and exhaust. As the rotational speed of the generator gradually increases and reaches the preset rotational speed, the generator starts to output voltage, and then the voltage gradually reaches the target value.

[0074] Since generators are usually idle for a long time and are only put into use during power outages for emergency or regular tests, this operating characteristic makes it easy to cause serious safety hazards if the generator fails at a critical moment. For example, in places with extremely high requirements for power continuity such as hospitals and data centers, it may lead to catastrophic consequences such as medical equipment shutdown and data loss.

[0075] Currently, the monitoring system of generators mainly relies on adding sensors such as temperature and rotational speed on the generators to monitor parameters such as temperature and rotational speed, and issues an alarm when these parameters exceed the pre-set warning thresholds. However, parameters such as temperature and rotational speed can only reflect the surface state of the generator and cannot deeply reveal the abnormal hidden dangers during the starting process.

[0076] To solve the problems of the existing technology, the embodiments of the present application provide a method, device, equipment, and computer storage medium for detecting the starting state of a generator.

[0077] First, the method for detecting the starting state of a generator provided by the embodiments of the present application will be introduced below. Among them, regarding the execution entity adopted by the embodiments of the present application, specifically, it can be a terminal device with data processing and analysis capabilities, such as a desktop computer, a laptop computer, etc., or a server, etc. In addition, the execution entity adopted by the embodiments of the present application can also be an execution entity in the form of software, such as a client installed in a terminal device, a software program, etc. For the sake of convenience of description, below, the embodiments of the present application use a server as the execution entity to illustrate a method for detecting the starting state of a generator provided by the present application.

[0078] Figure 1 The flowchart of the method for detecting the starting state of a generator provided by an embodiment of the present application is shown. As Figure 1 shown, the method may include the following steps:

[0079] S100: Send a start signal to the generator to be abnormally detected, and collect a detection data set of the output voltage of the generator changing with the start time.

[0080] In one or more embodiments of the present application, in order to detect the start-up state of the generator to be abnormally detected in subsequent steps, in this step, the server needs to send a start signal to the generator to be abnormally detected, so as to collect in real time a set of detection data on the output voltage of the generator changing with the start time after the generator starts.

[0081] Specifically, the server can send a start signal to the generator to be abnormally detected and collect a set of detection data on the output voltage of the generator changing with the start time.

[0082] It should be noted that after the server sends a start signal to the generator to be abnormally detected, it should promptly issue a data collection instruction to quickly and accurately collect a set of detection data on the output voltage changing with the start time, that is, each set of detection data consists of a start time and the output voltage output by the generator at that start time. The output voltage of the generator reflects the power generation capacity of the generator, and the change law of the output voltage directly reflects the operating state of the internal components of the generator (such as the rotational speed stability of the rotor, the working condition of the excitation system, etc.). Therefore, in this step, the server needs to collect and record a set of detection data on the output voltage changing with the start time from the time when the start signal is issued to the completion of the generator start for subsequent analysis of the start-up state of the generator.

[0083] Of course, in the present application, there is no limitation on the time interval and duration of data collection, which can be set according to actual needs. Of course, the duration can be set based on the ideal start time of the generator. For example, the duration can be twice the ideal start time to ensure the collection of a set of detection data for the entire start process. For example, after the start signal is sent, the server can collect a set of detection data composed of "start time, output voltage" every 0.1 second and continue to collect for 1.3 seconds. For example, the first set of detection data is "0S, 0V"... the 13th set of detection data is "1.3S, 100V". In addition, after the server collects a set of detection data on the output voltage of the generator changing with the start time, it can perform preprocessing operations (such as removing noise, etc.) on each set of detection data to improve the quality of the set of detection data.

[0084] S101: Divide each set of detection data to obtain multiple time windows.

[0085] In one or more embodiments of the present application, the start of the generator is a dynamically changing process, and its output voltage exhibits different change characteristics in different start-up periods. If the entire start process is analyzed, the changes in different start-up periods are ignored, thus masking potential hidden dangers. Therefore, in order to achieve a more detailed and accurate detection of the start-up state of the generator in subsequent steps, in this step, the server needs to divide each set of detection data to obtain multiple time windows.

[0086] Specifically, the server can divide each group of detection data to obtain multiple time windows.

[0087] It should be noted that several consecutive sub-intervals with a certain time length obtained after dividing the generator startup process are the time windows. Each time window contains one or more groups of detection data, and these groups of detection data record the change of the output voltage of the generator with the startup time within this time window. In this application, there is no limitation on the specific method of dividing to obtain multiple time windows. For example, it can be divided according to a preset quantity, divided according to a preset time length, etc., which can be set according to actual needs. If the detection data group obtained during the division process is not sufficient to form a complete time length, it can be used as a separate time window. Continuing with the above example, dividing the 13 groups of detection data collected above according to a time length of 0.5 seconds, three time windows can be obtained, that is, the 1st to 5th groups of detection data are one time window, and the time window is 0.0S - 0.4S; the 6th to 10th groups of detection data are one time window, and the time window is 0.5S - 0.9S; the 11th to 13th groups of detection data are one time window, and the time window is 1.0S - 1.3S. Of course, the more the number of time windows, the shorter the time period covered by each time window, and the higher the analysis accuracy. However, the workload of subsequent in-depth statistical calculations for each time window becomes more complex. Therefore, in this application, the number of divided time windows can be set according to actual needs. In one or more embodiments of this application, the server can respond to the user's division instruction and divide each group of detection data according to the target quantity set by this division instruction to obtain multiple time windows.

[0088] S102: For each time window, according to the startup time of each group of detection data within this time window, through the pre-determined target mapping relationship between time and voltage of the generator in the normal startup state, determine the reference voltage value corresponding to each group of detection data within this time window.

[0089] In one or more embodiments of this application, in order to determine the reference voltage value of each time window and the warning threshold of each time window in the subsequent steps. In this step, the server needs to determine the reference voltage value corresponding to each group of detection data within each time window.

[0090] Specifically, the server can, for each time window, according to the startup time of each group of detection data within this time window, through the pre-determined target mapping relationship between time and voltage of the generator in the normal startup state, determine the reference voltage value corresponding to each group of detection data within this time window.

[0091] It should be noted that in this application, there is no limitation on the specific method of determining the target mapping relationship between time and voltage of the generator in the normal startup state, which can be set according to actual needs. For example, for each model of generator, sample data groups of the output voltage of the generator of this model with the startup time in the normal startup state are collected multiple times. Through the various sample data groups collected multiple times and through statistical analysis, the target mapping relationship of this model can be obtained. In one or more embodiments of this application, the server can determine the model of the generator to be detected for anomalies as the target model, and call the target mapping relationship corresponding to this target model from the target mapping relationships respectively corresponding to each pre-determined model, so as to determine the reference voltage values respectively corresponding to each detection data group within this time window according to the startup time of each detection data group within this time window through the target mapping relationship corresponding to this target model.

[0092] S103: For each time window, determine the reference voltage value of this time window and the warning threshold of this time window according to the reference voltage values respectively corresponding to each detection data group within this time window.

[0093] In one or more embodiments of this application, in order to determine the anomaly detection results of each detection data group in subsequent steps. In this step, the server can, for each time window, determine the reference voltage value of this time window and the warning threshold of this time window according to the reference voltage values respectively corresponding to each detection data group within this time window.

[0094] Specifically, the server can, for each time window, determine the reference voltage value of this time window and the warning threshold of this time window according to the reference voltage values respectively corresponding to each detection data group within this time window.

[0095] It should be noted that the reference voltage value refers to the reference voltage value for each set of detection data within each time window during the generator startup process. It is obtained through statistical methods and is used to characterize the voltage reference level within that time window, reflecting the expected value of the generator output voltage within that time window under normal operating conditions. The warning threshold refers to the boundary value obtained through statistical methods for each set of detection data within each time window during the generator startup process, used to determine whether the output voltage is abnormal, reflecting the fluctuation range of the generator output voltage within that time window under normal operating conditions. In this application, there is no limitation on the specific method for determining the reference voltage value and the warning threshold for this time window, which can be set according to actual needs. For example, the reference voltage value can be the mean of each reference voltage value (reflecting the average level of the data), the weighted average value (referring to the importance of different data points), the median (reflecting the typical level of the data), etc. The warning threshold can be the standard deviation of each reference voltage value (reflecting the degree of dispersion of the data), the variance (reflecting the fluctuation range of the data), etc. That is, different statistical quantities reflect the characteristics and distribution laws of the data from different perspectives. In one or more embodiments of this application, the server can calculate the average of the reference voltage values corresponding to each set of detection data within each time window to obtain the reference voltage value. Based on the reference voltage values corresponding to each set of detection data within that time window and the reference voltage mean, the reference voltage standard deviation is determined. The reference voltage mean is used as the reference voltage value for that time window, and based on the reference voltage standard deviation and a preset threshold, the warning threshold for that time window is determined. The process of determining the reference voltage value and the warning threshold is as follows:

[0096]

[0097] Among them, there are T sets of detection data within each time window. "i - T + 1 to i" is the range of the startup time acquisition sequence numbers within the time window. μ(T) is the reference voltage mean, σ(T) is the reference voltage standard deviation, W is the warning threshold, and θ is the threshold, which can be set according to actual needs. For example, different types of generators can be set with different threshold values.

[0098] S104: For each set of detection data, determine the abnormal detection result of this set of detection data based on the output voltage of this set of detection data, the reference voltage value of the time window to which this set of detection data belongs, and the warning threshold of the time window to which this set of detection data belongs.

[0099] In one or more embodiments of this application, in order to detect the startup state of the generator in subsequent steps, the server can, for each set of detection data, determine the abnormal detection result of this set of detection data based on the output voltage of this set of detection data, the reference voltage value of the time window to which this set of detection data belongs, and the warning threshold of the time window to which this set of detection data belongs.

[0100] Specifically, for each detection data group, the server can determine the anomaly detection result of the detection data group based on the output voltage of the detection data group, the reference voltage value of the time window to which the detection data group belongs, and the warning threshold of the time window to which the detection data group belongs.

[0101] It should be noted that in this application, there is no limitation on the specific method for determining the anomaly detection result based on the output voltage, the reference voltage value, and the warning threshold. In one or more embodiments of this application, the server can determine the absolute difference between the output voltage of the detection data group and the reference voltage value of the time window to which the detection data group belongs. When the absolute difference is greater than the warning threshold of the time window to which the detection data group belongs, the anomaly detection result of the detection data group is abnormal. When the absolute difference is not greater than the warning threshold of the time window to which the detection data group belongs, the anomaly detection result of the detection data group is normal.

[0102] When the reference voltage value is the average value of the reference voltages and W is the product of the threshold and the standard deviation of the reference voltages, the process of determining the anomaly detection result of the detection data group is as follows:

[0103] |v i -μ(T)|>W (4)

[0104] It should be noted that in addition to using normal or abnormal to represent the anomaly detection result, "0" and "1" can also be used to represent the anomaly detection result. For example, "0" can be used to represent normal and "1" can be used to represent abnormal. That is, in this application, there is no limitation on the expression form of the anomaly detection result, which can be set according to actual needs.

[0105] S105: Detect the starting state of the generator according to the anomaly detection results of the respective detection data groups.

[0106] In one or more embodiments of this application, the server can detect the starting state of the generator based on the anomaly detection results of the respective detection data groups determined in step S104.

[0107] Specifically, the server can detect the starting state of the generator according to the anomaly detection results of the respective detection data groups.

[0108] It should be noted that in this application, there is no limitation on the specific method for detecting the starting state of the generator, which can be set according to actual needs. For example, a target detection ratio can be set. When the ratio between the number of normal abnormal detection results and the number of abnormal abnormal detection results is greater than the preset target detection ratio, it is considered that the generator is in a normal starting state. On the contrary, when the ratio between the normal abnormal detection results and the abnormal abnormal detection results is not greater than the preset target detection ratio, it is considered that the generator is in an abnormal starting state. In one or more embodiments of this application, the server can determine that the generator is in a normal starting state when the abnormal detection results of each detection data group are all normal, and determine that the generator is in an abnormal starting state when the abnormal detection results of each detection data group are not all normal. In addition, when the generator is in an abnormal starting state, potential problems of the generator can be explored in depth to ensure the stable operation of the generator.

[0109] In the above method, by dividing each detection data group collected during the generator startup process, multiple time windows are obtained, and then each time window is delved into. According to the startup time of each detection data group within this time window, through the pre-determined target mapping relationship between time and voltage of the generator in the normal startup state, the reference voltage value corresponding to each detection data group within this time window is determined. Thus, by statistically analyzing the difference between the output voltage of the detection data group and the reference voltage value, the subtle dynamic changes of the detection data group during the startup process are captured, realizing the detection of the generator startup state. That is to say, when the generator is in an abnormal startup state, it can be analyzed from the detection data group of the time window, so as to discover potential problems as early as possible and conduct maintenance detection in advance based on the predicted state detection results to ensure the stable operation of the generator.

[0110] In step S102, the target mapping relationship between time and voltage of the generator in the normal startup state can be determined through statistical analysis. In addition, this application provides an embodiment of iteratively updating to obtain the target mapping relationship. That is, in one or more embodiments of this application, the server can determine the target mapping relationship between time and voltage of each model of the generator in the normal startup state, specifically as follows:

[0111] First, the server can obtain the nominal output voltage and ideal startup time of each model of the generator. The nominal output voltage refers to the output voltage value designed for the generator under normal working conditions (such as rated speed, rated load, etc.). For example, for a generator with a nominal output voltage of 220V, its output voltage should be close to 220V during normal operation. The ideal startup time refers to the shortest time required for the generator to reach the nominal output voltage from startup, and this parameter reflects the startup performance of the generator.

[0112] Secondly, for each model, the server can determine the initial mapping relationship between time and voltage of the generator of this model in the normal startup state according to the nominal output voltage of this model, the ideal startup time of this model, and the preset voltage change parameters. Collect the sample data group of the output voltage of the generator of this model changing with the startup time in the normal startup state, and determine the predicted output voltage of each sample data group through this initial mapping relationship.

[0113] Finally, the server can determine the voltage fluctuation gradient based on the deviation value between the output voltage value of each sample data group and the predicted output voltage of each sample data group. Furthermore, with the goal of minimizing this voltage fluctuation gradient, through the gradient descent method, iteratively update the initial mapping relationship of this model until the preset stop condition is met, and obtain the target mapping relationship corresponding to this model.

[0114] It should be noted that in this application, the specific method for determining the voltage fluctuation gradient is not limited and can be set according to actual needs, such as the difference method, etc. In one or more embodiments of this application, the server can determine the deviation value of each sample data group according to the output voltage of this sample data group and the predicted output voltage of this sample data group. Determine the total deviation value according to the deviation values of each sample data group, and then take the partial derivative of the total deviation value with respect to the voltage change parameter as the voltage fluctuation gradient. The voltage fluctuation gradient reflects the influence degree and direction of the voltage change parameter on the total deviation value. Iteratively updating the initial mapping relationship with the goal of minimizing the voltage fluctuation gradient is to make the total deviation value close to zero by adjusting the voltage change parameter, so as to obtain a target mapping relationship with high accuracy and good reliability.

[0115] The process of the first iterative update and subsequent iterative updates of the above initial mapping relationship is as follows:

[0116] In one or more embodiments of this application, first, the server can perform the first iterative update on the voltage change parameter according to the voltage fluctuation gradient, and determine the initial mapping relationship after the first iterative update based on the voltage change parameter after the first iterative update, where the voltage change parameter is negatively correlated with the voltage fluctuation gradient.

[0117] Secondly, the server can, based on each sample data group, with the goal of minimizing the voltage fluctuation gradient, perform multiple iterative updates on the initial mapping relationship after the first iterative update through the gradient descent method until the preset stop condition is met, and obtain the target mapping relationship corresponding to this model. Among them, in this application, the specific content of this preset stop condition is not limited, such as stopping the iteration when the number of iterative updates reaches the preset number of times.

[0118] In one or more embodiments of the present application, the process of performing multiple iterative updates on the initial mapping relationship after the first iterative update using the gradient descent method is as follows:

[0119] First, for each iterative update among the multiple iterative updates, the server can determine the predicted output voltage of each sample data group in this iterative update through the initial mapping relationship after the previous iterative update. According to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group in this iterative update, the voltage fluctuation gradient of this iterative update is determined.

[0120] Second, for each iterative update among the multiple iterative updates, the server can perform this iterative update on the voltage change parameter after the previous iterative update according to the voltage fluctuation gradient of this iterative update, and determine the initial mapping relationship after this iterative update based on the voltage change parameter after this iterative update.

[0121] Finally, for each iterative update among the multiple iterative updates, the server can determine whether the voltage fluctuation gradient of this iterative update is less than the preset gradient threshold; if so, the initial mapping relationship after this iterative update is used as the target mapping relationship corresponding to this model; if not, the next iterative update continues.

[0122] The calculation process of obtaining the target mapping relationship through the above iterative update is as follows:

[0123]

[0124] Among them, n is the nominal output voltage, m is the ideal start time, k = 1 is the preset voltage change parameter, t i is the start time, and the calculation formula (5) is the initial mapping relationship when k = 1. v i is the output voltage of the collected sample data group, β is the deviation value between the output voltage of the sample data group and the predicted output voltage of the sample data group. G(k) is the total deviation value, G ′ is the voltage fluctuation gradient. In the calculation formula (5), k is the voltage change parameter after the previous iterative update, that is, v ’ is the initial mapping relationship after the previous iterative update, then in the calculation formula (9), k ′ is the voltage change parameter after this iterative update, α is the learning rate / update step size, which can be set according to actual needs, such as set to 0.001. In the calculation formula (9), the voltage change parameter is updated along the reverse direction of the voltage fluctuation gradient, that is, the voltage change parameter and the voltage fluctuation gradient are negatively correlated. In the above, k ′ is brought back to the calculation formula (5) to repeat multiple iterations until the preset stop condition is reached, and the initial mapping relationship after this iterative update is used as the target mapping relationship corresponding to this model.

[0125] As shown in Figure 2 FIG. 4, it is a schematic flowchart of an iterative update for determining a target mapping relationship provided by an embodiment of the present application.

[0126] In addition, the present application provides a generator starting state detection system. As shown in Figure 3 FIG. 5, the system includes a system control unit, a data acquisition unit, and an analysis and prediction unit. For details, reference can be made to the content of the above-mentioned generator starting state detection method, which will not be elaborated here.

[0127] The system control unit is responsible for coordinating the collaborative operation of the data storage, the analysis and prediction unit, and the data acquisition unit, controlling the data flow direction, ensuring that when a generator starting signal is received, a data acquisition instruction is issued in a timely manner, and the collected output voltage together with the time stamp is transmitted to the storage quickly and accurately. When analysis is required, the detection data sets are orderly delivered to the analysis device to ensure the smoothness and efficiency of the system operation.

[0128] The data acquisition unit is responsible for acquiring the output voltage. The data acquisition unit is connected to the generator control system or the diesel generator and mains power control system. When receiving the engine starting signal, it starts to acquire the output voltage of the generator output end changing with time, and records all the detection data sets during the process from the system issuing the starting signal to the start completion.

[0129] The analysis and prediction unit is used to retrieve the detection data sets of the generator starting time and output voltage in the data storage, and use time window statistical analysis to predict starting hidden dangers.

[0130] Figure 4 FIG. 6 is a schematic structural diagram of a generator starting state detection device provided by an embodiment of the present application. As shown in Figure 4 FIG. 7, the device may include a data acquisition module 400, a precision division module 401, a reference voltage module 402, an early warning parameter module 403, a first detection module 404, and a target detection module 405.

[0131] The data acquisition module 400 is configured to send a starting signal to the generator to be abnormally detected, and acquire the detection data sets of the output voltage of the generator changing with the starting time;

[0132] The precision division module 401 is configured to divide each detection data set according to a preset quantity to obtain a plurality of time windows;

[0133] The reference voltage module 402 is configured to, for each time window, determine the reference voltage value corresponding to each detection data set in the time window according to the starting time of each detection data set in the time window and the target mapping relationship between time and voltage of the generator in a normal starting state determined in advance;

[0134] An early warning parameter module 403, configured to determine a reference voltage value and an early warning threshold value of each time window according to the reference voltage values respectively corresponding to each detection data group within the time window for each time window.

[0135] A first detection module 404, configured to determine an abnormality detection result of each detection data group according to the output voltage of the detection data group, the reference voltage value of the time window to which the detection data group belongs, and the early warning threshold value of the time window to which the detection data group belongs.

[0136] A target detection module 405, configured to detect the starting state of the generator according to the abnormality detection results of each detection data group.

[0137] In an implementable embodiment, the device includes an iterative update module 406, specifically configured to obtain the nominal output voltage and the ideal starting time of each model of the generator; for each model, determine an initial mapping relationship between time and voltage of the generator of this model in a normal starting state according to the nominal output voltage of this model, the ideal starting time of this model, and a preset voltage change parameter; for each model, collect sample data groups of the output voltage of the generator of this model changing with the starting time in a normal starting state; determine the predicted output voltage of each sample data group through the initial mapping relationship; determine a voltage fluctuation gradient according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group; with the minimization of the voltage fluctuation gradient as the goal, iteratively update the initial mapping relationship through the gradient descent method until a preset stop condition is satisfied, and obtain the target mapping relationship corresponding to this model.

[0138] In an implementable embodiment, the iterative update module 406 may also be configured to determine the deviation value of each sample data group according to the output voltage of the sample data group and the predicted output voltage of the sample data group; determine the total deviation value according to the deviation values of each sample data group; and use the partial derivative of the total deviation value with respect to the voltage change parameter as the voltage fluctuation gradient.

[0139] In an implementable embodiment, the iterative update module 406 is further configured to perform a first iterative update on the voltage change parameter according to the voltage fluctuation gradient, and determine an initial mapping relationship after the first iterative update based on the voltage change parameter after the first iterative update, where the voltage change parameter is negatively correlated with the voltage fluctuation gradient; based on each sample data group, with the minimization of the voltage fluctuation gradient as the target, perform multiple iterative updates on the initial mapping relationship after the first iterative update through the gradient descent method until a preset stop condition is met, and obtain the target mapping relationship corresponding to this model.

[0140] In an implementable embodiment, for each iterative update in the multiple iterative updates, the iterative update module 406 is further configured to determine the predicted output voltage of each sample data group in this iterative update through the initial mapping relationship after the previous iterative update; for each iterative update in the multiple iterative updates, determine the voltage fluctuation gradient of this iterative update according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group in this iterative update; for each iterative update in the multiple iterative updates, perform this iterative update on the voltage change parameter after the previous iterative update according to the voltage fluctuation gradient of this iterative update, and determine the initial mapping relationship after this iterative update based on the voltage change parameter after this iterative update; for each iterative update in the multiple iterative updates, determine whether the voltage fluctuation gradient of this iterative update is less than a preset gradient threshold; if so, use the initial mapping relationship after this iterative update as the target mapping relationship corresponding to this model; if not, continue with the next iterative update.

[0141] In an implementable embodiment, the reference voltage module 402 is specifically configured to determine the model of the generator to be abnormally detected as the target model; call the target mapping relationship corresponding to the target model from the target mapping relationships corresponding to each model. According to the start times of each detection data group within this time window, determine the reference voltage value corresponding to each detection data group within this time window through the target mapping relationship corresponding to the target model.

[0142] In an implementable embodiment, the warning parameter module 403 is specifically configured to calculate the average of the reference voltage values corresponding to each detection data group within this time window to obtain the reference voltage average value; determine the reference voltage standard deviation according to the reference voltage values corresponding to each detection data group within this time window and the reference voltage average value; use the reference voltage average value as the reference voltage value of this time window; determine the warning threshold of this time window according to the reference voltage standard deviation and a preset threshold.

[0143] In an implementable embodiment, the first detection module 404 is specifically configured to determine the absolute difference between the output voltage of the detection data group and the reference voltage value of the time window to which the detection data group belongs; when the absolute difference is greater than the warning threshold of the time window to which the detection data group belongs, the anomaly detection result of the detection data group is abnormal; when the absolute difference is not greater than the warning threshold of the time window to which the detection data group belongs, the anomaly detection result of the detection data group is normal.

[0144] In an implementable embodiment, the target detection module 405 is specifically configured to determine that the generator is in a normal start state when the anomaly detection results of all detection data groups are normal; and determine that the generator is in an abnormal start state when the anomaly detection results of all detection data groups are not all normal.

[0145] In an implementable embodiment, the precision division module 401 is specifically configured to respond to a division instruction from a user; and divide each detection data group according to the target quantity set by the division instruction to obtain a plurality of time windows.

[0146] Figure 5 FIG. shows a schematic hardware structure diagram of a generator start state detection device provided by an embodiment of the present application.

[0147] The generator start state detection device may include a processor 501 and a memory 502 storing computer program instructions.

[0148] Specifically, the processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0149] The memory 502 may include a mass storage for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In one example, the memory 502 may include removable or non-removable (or fixed) media, or the memory 502 is a non-volatile solid state memory. The memory 502 may be internal or external to the integrated gateway disaster recovery device.

[0150] In one example, the memory 502 may be a Read Only Memory (ROM). In one example, the ROM may be a mask-programmed ROM, a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), an Electrically Alterable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0151] The memory 502 may include a read only memory (ROM), a random access memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical, or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described in reference to the method according to one aspect of the present disclosure.

[0152] The processor 501 reads and executes the computer program instructions stored in the memory 502 to implement Figure 1 a generator startup state detection method in the illustrated embodiment.

[0153] In one example, a generator startup state detection device may further include a communication interface 503 and a bus 504. Among them, as Figure 5 shown, the processor 501, the memory 502, and the communication interface 503 are connected through the bus 504 to complete communication with each other.

[0154] The communication interface 503 is mainly used to implement communication between each module, device, unit, and / or device in the embodiments of the present application.

[0155] The bus 504 includes hardware, software, or both, and couples the components of the online data flow metering device to each other. By way of example and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 304 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0156] The generator starting state detection device can detect the starting state of the generator based on the detection data set collected during the starting process, so as to implement the combination with Figure 1 a generator starting state detection method described.

[0157] In addition, in combination with a generator starting state detection method in the above embodiments, the embodiments of the present application can provide a computer storage medium to implement. Computer program instructions are stored on the computer storage medium; when the computer program instructions are executed by a processor, the methods for detecting any generator starting state in the above embodiments are implemented.

[0158] The embodiments of the present application also provide a computer program product, including a computer program, and when the computer program is executed by a processor, the methods for detecting any generator starting state in the above embodiments are implemented.

[0159] It should be clear that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, the detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0160] The functional blocks shown in the above-described structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments used to perform the required tasks. The programs or code segments can be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave over a transmission medium or a communication link. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact disc read-only memory (CD-ROM), optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segments can be downloaded via a computer network such as the Internet, an intranet, and so on.

[0161] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, can be different from the order in the embodiments, or several steps can be executed simultaneously.

[0162] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block in the flowcharts and / or block diagrams, and the combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the functions / actions specified in one or more blocks of the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It should also be understood that each block in the block diagrams and / or flowcharts, and the combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0163] As described above, this is only the specific implementation manner of the present application. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present application.

Claims

1. A method for detecting the starting state of a generator, characterized in that, Including: Sending a start signal to the generator to be abnormally detected, and collecting a detection data set of the output voltage of the generator varying with the start time; Dividing each detection data set to obtain multiple time windows; For each time window, according to the start times of the detection data sets within the time window, and through the pre-determined target mapping relationship between time and voltage of the generator in the normal start state, determining the reference voltage values respectively corresponding to the detection data sets within the time window; For each time window, determining the reference voltage value of the time window and the warning threshold of the time window according to the reference voltage values respectively corresponding to the detection data sets within the time window; For each detection data set, determining the abnormal detection result of the detection data set according to the output voltage of the detection data set, the reference voltage value of the time window to which the detection data set belongs, and the warning threshold of the time window to which the detection data set belongs; Detecting the start state of the generator according to the abnormal detection results of the detection data sets.

2. The method according to claim 1, wherein The method further includes: Obtaining the nominal output voltage and the ideal start time of each model of the generator; For each model, determining the initial mapping relationship between time and voltage of the generator of this model in the normal start state according to the nominal output voltage of this model, the ideal start time of this model, and the preset voltage change parameter; For each model, collecting a sample data set of the output voltage of the generator of this model varying with the start time in the normal start state; Determining the predicted output voltage of each sample data set through the initial mapping relationship; Determining the voltage fluctuation gradient according to the deviation value between the output voltage of each sample data set and the predicted output voltage of each sample data set; Taking the minimization of the voltage fluctuation gradient as the goal, and through the gradient descent method, iteratively updating the initial mapping relationship until the preset stop condition is met, to obtain the target mapping relationship corresponding to this model.

3. The method according to claim 2, wherein Determining the voltage fluctuation gradient according to the deviation value between the output voltage of each sample data set and the predicted output voltage of each sample data set, specifically including: For each sample data set, determining the deviation value of the sample data set according to the output voltage of the sample data set and the predicted output voltage of the sample data set; Determining the total deviation value according to the deviation values of the sample data sets; Taking the partial derivative of the total deviation value with respect to the voltage change parameter as the voltage fluctuation gradient.

4. The method according to claim 2, characterized in that, Taking the minimization of the voltage fluctuation gradient as the goal, and through the gradient descent method, iteratively updating the initial mapping relationship until the preset stop condition is met, to obtain the target mapping relationship corresponding to this model, specifically including: According to the voltage fluctuation gradient, performing the first iterative update on the voltage change parameter, and based on the voltage change parameter after the first iterative update, determining the initial mapping relationship after the first iterative update, and the voltage change parameter and the voltage fluctuation gradient are negatively correlated; Based on each sample data group, aiming at minimizing the voltage fluctuation gradient, through the gradient descent method, the initial mapping relationship after the first iterative update is iteratively updated multiple times until the preset stop condition is satisfied, and the target mapping relationship corresponding to this model is obtained.

5. The method according to claim 4, characterized in that, Based on each sample data group, aiming at minimizing the voltage fluctuation gradient, through the gradient descent method, the initial mapping relationship after the first iterative update is iteratively updated multiple times until the preset stop condition is satisfied, and the target mapping relationship corresponding to this model is obtained. Specifically, it includes: For each iterative update in the multiple iterative updates, through the initial mapping relationship after the previous iterative update, determine the predicted output voltage of each sample data group in this iterative update; For each iterative update in the multiple iterative updates, according to the deviation value between the output voltage of each sample data group and the predicted output voltage of each sample data group in this iterative update, determine the voltage fluctuation gradient of this iterative update; For each iterative update in the multiple iterative updates, according to the voltage fluctuation gradient of this iterative update, perform this iterative update on the voltage change parameters after the previous iterative update, and based on the voltage change parameters after this iterative update, determine the initial mapping relationship after this iterative update; For each iterative update in the multiple iterative updates, judge whether the voltage fluctuation gradient of this iterative update is less than the preset gradient threshold; If so, use the initial mapping relationship after this iterative update as the target mapping relationship corresponding to this model; If not, continue with the next iterative update.

6. The method according to claim 2, wherein According to the start times of each detection data group within this time window, through the pre-determined target mapping relationship between the time and voltage of the generator in the normal start state, determine the reference voltage values corresponding to each detection data group within this time window. Specifically, it includes: Determine the model of the generator to be abnormally detected as the target model; From the target mapping relationships corresponding to each model, call the target mapping relationship corresponding to the target model; According to the start times of each detection data group within this time window, through the target mapping relationship corresponding to the target model, determine the reference voltage values corresponding to each detection data group within this time window.

7. The method according to claim 1, characterized in that, According to the reference voltage values corresponding to each detection data group within this time window, determine the reference voltage value of this time window and the warning threshold of this time window. Specifically, it includes: Take the average of the reference voltage values corresponding to each detection data group within this time window to obtain the average reference voltage; According to the reference voltage values corresponding to each detection data group within this time window and the average reference voltage, determine the standard deviation of the reference voltage; Take the average reference voltage as the reference voltage value of this time window; According to the standard deviation of the reference voltage and the preset threshold, determine the warning threshold of this time window.

8. The method according to claim 1, wherein According to the output voltage of this detection data group, the reference voltage value of the time window to which this detection data group belongs, and the warning threshold of the time window to which this detection data group belongs, determine the abnormal detection result of this detection data group. Specifically, it includes: Determine the absolute difference between the output voltage of the detected data set and the reference voltage value of the time window to which the detected data set belongs; In the case where the absolute difference is greater than the warning threshold of the time window to which the detected data set belongs, the anomaly detection result of the detected data set is abnormal; In the case where the absolute difference is not greater than the warning threshold of the time window to which the detected data set belongs, the anomaly detection result of the detected data set is normal.

9. The method according to claim 1, wherein Detect the starting state of the generator according to the anomaly detection results of each detected data set, specifically including: In the case where the anomaly detection results of each detected data set are all normal, the generator is in a normal starting state; In the case where the anomaly detection results of each detected data set are not all normal, the generator is in an abnormal starting state.

10. The method according to claim 1, wherein Divide each detected data set to obtain multiple time windows, specifically including: Respond to the user's division instruction; Divide each detected data set according to the target number set by the division instruction to obtain multiple time windows.

11. A generator starting state detection device, characterized in that, The device includes: A data acquisition module, configured to send a start signal to the generator to be abnormally detected and acquire a detected data set of the output voltage of the generator changing with the start time; An accuracy division module, configured to divide each detected data set according to a preset number to obtain multiple time windows; A reference voltage module, configured to, for each time window, determine the reference voltage value corresponding to each detected data set in the time window according to the start time of each detected data set in the time window through a pre-determined target mapping relationship between time and voltage of the generator in a normal starting state; A warning parameter module, configured to, for each time window, determine the reference voltage value and the warning threshold of the time window according to the reference voltage values corresponding to each detected data set in the time window; A first detection module, configured to, for each detected data set, determine the anomaly detection result of the detected data set according to the output voltage of the detected data set, the reference voltage value of the time window to which the detected data set belongs, and the warning threshold of the time window to which the detected data set belongs; A target detection module, configured to detect the starting state of the generator according to the anomaly detection results of each detected data set.

12. A generator startup state detection device, characterized in that, The device includes: a processor and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the generator starting state detection method according to any one of claims 1 to 7.

13. A computer-readable storage medium, characterized in that, Computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are executed by the processor, the generator starting state detection method according to any one of claims 1 to 7 is implemented.

Images