Monitoring threshold range determination method and device, equipment, medium and product

Abstract

The invention discloses a monitoring threshold range determination method and device, equipment, a medium and a product. The method comprises the steps that a first historical data set is obtained, the first historical data set comprises at least one piece of historical data of a sensor in a system, and the historical data are data collected by the sensor when the system is in a normal operation state; performing a monitoring threshold range determination operation on the historical data of each sensor to obtain a monitoring threshold range of each sensor; the monitoring threshold range determination operation comprises the following steps: inputting each piece of historical data into the monitoring model to obtain an evaluation value corresponding to each piece of historical data; the monitoring threshold range of the sensor is determined according to the historical data and the evaluation value corresponding to the historical data, through the technical scheme, the monitoring threshold range of the sensor can be automatically generated, and then the labor cost and the time cost for setting the monitoring threshold range of the sensor are reduced.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular, to a method, apparatus, device, medium, and product for determining a monitoring threshold range. Background Art

[0002] Performing condition monitoring on large and medium-sized complex equipment can improve the reliability and safety of the equipment and system, reduce the risk coefficient, and reduce the maintenance cost.

[0003] Currently, condition monitoring generally relies on the fixed threshold range method. This method sets a fixed value as the threshold and monitors the steady-state data through the fixed threshold range. Once the monitored data exceeds the fixed threshold, an alarm will be triggered. For example, the upper limit value of a certain temperature measurement point is 55°C, and an alarm will be generated when the data transmitted from this temperature measurement point is greater than 55.

[0004] This method is simple and reliable and easy to implement, but there are also problems such as the need to manually set the upper and lower limits of the threshold, which will consume a large amount of human and time costs when there are a large number of monitoring points. Summary of the Invention

[0005] Embodiments of the present invention provide a method, apparatus, device, medium, and product for determining a monitoring threshold range, which can automatically generate a monitoring threshold range, thereby reducing the human and time costs of setting the monitoring threshold range of sensors.

[0006] According to one aspect of the present invention, there is provided a method for determining a monitoring threshold range, including:

[0007] Obtain a first historical data set, where the first historical data set includes at least one historical data of sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operating state;

[0008] Perform a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor;

[0009] The monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; and determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

[0010] According to another aspect of the present invention, there is provided a monitoring threshold range determination apparatus, where the monitoring threshold range determination apparatus includes:

[0011] A first historical data set acquisition module, configured to obtain a first historical data set, where the first historical data set includes at least one historical data of sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operating state;

[0012] A monitoring threshold range determination module for each sensor, which is used to perform a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor; the monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; and determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

[0013] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the monitoring threshold range determination method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the monitoring threshold range determination method according to any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, there is provided a computer program product, and the computer program implements the monitoring threshold range determination method as described in any one of the embodiments of the present invention when executed by a processor.

[0019] In the embodiment of the present invention, by first obtaining a first historical data set, and then performing a monitoring threshold range determination operation on the historical data of each sensor in the first historical data set to obtain the monitoring threshold range of each sensor; the monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; and determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data, it is possible to automatically generate the monitoring threshold range of the sensor, thereby reducing the labor cost and time cost of setting the monitoring threshold range of the sensor.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Description of the Drawings

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present invention, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other relevant drawings can also be obtained based on these drawings.

[0022] Figure 1 is a flowchart of a method for determining a monitoring threshold range in an embodiment of the present invention;

[0023] Figure 2 is a schematic structural diagram of a device for determining a monitoring threshold range in an embodiment of the present invention;

[0024] Figure 3 is a schematic structural diagram of an electronic device in an embodiment of the present invention. Detailed Embodiments

[0025] In order to enable those skilled in the art of the present technology to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device comprising a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0027] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0028] Embodiment 1

[0029] Figure 1The flowchart of a method for determining a monitoring threshold range provided by an embodiment of the present invention. This embodiment is applicable to the situation of determining the monitoring threshold range of sensors. This method can be executed by a device for determining the monitoring threshold range in the embodiment of the present invention, and the device can be implemented in a software and / or hardware manner, such as Figure 1 As shown, the method specifically includes the following steps:

[0030] S110, obtain a first historical data set.

[0031] In this embodiment, the first historical data set includes: at least one historical data of sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operating state. It should be noted that the sensors in the embodiments of the present invention are all set monitoring points.

[0032] In this embodiment, the system may include one device or multiple devices, and a large number of sensors are equipped on each device. For example, temperature sensors, pressure sensors, flow sensors, vibration sensors, etc.

[0033] It should be noted that since the data collected when the system is in a normal operating state generally changes continuously. For example, the change rates of temperature, flow, and pressure are generally within a certain range, and the data exceeding this range is the data collected when the system is in an abnormal operating state.

[0034] In this embodiment, the change rate of each historical data in the first historical data set is less than or equal to a change rate threshold.

[0035] Optionally, obtaining the first historical data set includes:

[0036] Obtain an initial historical data set, where the initial historical data set includes: at least one historical data collected by sensors in the system and the acquisition timestamp corresponding to each historical data.

[0037] In this embodiment, if there are N sensors in the system and m historical data are collected for each sensor, the initial historical data set is an m×N matrix.

[0038] Determine the change rate of each historical data according to each historical data and the acquisition timestamp corresponding to each historical data.

[0039] In this embodiment, the method for determining the change rate of each historical data according to each historical data and the acquisition timestamp corresponding to each historical data can be: determine adjacent historical data according to the acquisition timestamp corresponding to each historical data, and determine the ratio of the difference between adjacent historical data to the difference between the corresponding sampling timestamps as the change rate of the historical data.

[0040] The historical data whose change rate is greater than the change rate threshold is eliminated from the initial historical data set to obtain a target historical data set.

[0041] In this embodiment, the historical data with a change rate greater than the change rate threshold is determined as data collected when the system is in an abnormal operation state. The data collected when the system is in an abnormal operation state is removed from the initial historical data set, and the remaining data is data collected when the system is in a normal operation state.

[0042] The target historical data set is split to obtain a first historical data set and a second historical data set.

[0043] In this embodiment, the target historical data set is split to obtain the first historical data set and the second historical data set in a manner of: splitting the target historical data set according to a preset ratio to obtain the first historical data set and the second historical data set. It should be noted that the preset ratio can be 1:1 or 2:1, and the embodiment of the present invention is not limited to this.

[0044] In a specific example, if there are N sensors in the system, the historical data with a time length of m obtained by the i-th sensor is expressed as:

[0045] x i =[x1,x2...x m ];

[0046] In this embodiment, x i is the historical data of the i-th sensor, and the number of historical data collected by each sensor is m.

[0047] The first historical data set includes historical data of N sensors. The first historical data set D can be expressed as:

[0048] D=[x1,x2...x i ...x N ];

[0049] That is, D is an m×N matrix.

[0050] It should be noted that the historical data in the initial historical dataset can be historical data for one month, two months, one year, or two years. It is necessary to ensure that the amount of historical data is large enough, not less than 5000 pieces. If there is a lack of sufficient data, a reasonable monitoring model cannot be established. Since the data collected by the sensors during normal operation of the system is required for modeling, the historical data in the initial historical dataset needs to be divided into normal operation data and abnormal operation data. The former is used for modeling and verification, while the latter is only used for verification. The division of normal operation data and abnormal operation data can be carried out by a method based on the change rate of each historical data to divide the normal operation data and abnormal operation data in the initial historical dataset. The change rate of historical data is obtained by calculating the difference between the historical data collected at time t + 1 and the historical data collected at time t. Since the data collected by the sensors during normal operation of the system generally changes continuously. For example, the change rates of temperature, flow rate, and pressure are generally within a certain range. Data exceeding this range is the data collected by the sensors when the system is in an abnormal operation state. By screening based on the change rate, the historical data with too large change rates in the initial historical dataset can be excluded. After excluding the historical data with a change rate greater than the change rate threshold in the initial historical dataset, a target historical dataset is obtained. Then, based on a preset ratio, the target historical dataset is split to obtain a first historical dataset and a second historical dataset.

[0051] S120. Perform an operation to determine the monitoring threshold range for the historical data of each sensor to obtain the monitoring threshold range for each sensor.

[0052] In this embodiment, the operation to determine the monitoring threshold range includes: inputting each historical data into the monitoring model to obtain an evaluation value corresponding to each historical data; determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

[0053] In this embodiment, setting the monitoring threshold range for each sensor is for the monitoring of the system.

[0054] In this embodiment, the monitoring model can be: a monitoring model selected based on the kernel regression state evaluation method.

[0055] It should be noted that kernel regression is a non - parametric estimation method used for regression analysis and prediction of data. Bias measures the average difference between the model prediction result and the true value, reflecting the accuracy of the model. In kernel regression, the bias is related to the choice of kernel function, the size of the bandwidth, and the distribution of the data. Selecting a monitoring model based on the kernel regression state evaluation method requires comprehensive consideration of multiple factors and finding the most suitable model according to the actual business requirements to achieve accurate evaluation of the data collected by the sensors.

[0056] In this embodiment, a monitoring model is established based on the second historical data set.

[0057] In this embodiment, the method for determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data may be: determining the difference between the evaluation value corresponding to each historical data and the corresponding historical data as the evaluation residual of each historical data; determining the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data. The method for determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data may also be: determining the difference between the evaluation value corresponding to each historical data and the corresponding historical data as the evaluation residual of each historical data; determining the monitoring threshold range of the sensor according to the mean and variance of the evaluation residuals of each historical data.

[0058] In this embodiment, a monitoring model is established according to the data collected by the sensor when the system is operating normally. For example, the kernel regression state evaluation method and the second historical data set can be used to establish the monitoring model. The monitoring model can output corresponding evaluation values for the input historical data, and then obtain the difference between the historical data and the evaluation value (i.e., the evaluation residual).

[0059] It should be noted that when the system is operating normally, the evaluation residual will fluctuate around 0, and the variance will not change too much; when the system is operating abnormally, the evaluation residual will deviate from 0, and the variance will suddenly increase. The monitoring model can be used to distinguish between the normal operation and abnormal operation of the system.

[0060] Optionally, determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data includes:

[0061] Determining the difference between each historical data and the corresponding evaluation value as the evaluation residual of each historical data.

[0062] Determining the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data.

[0063] In this embodiment, the method for determining the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data may be: taking the difference between the mean of the evaluation residuals of each historical data and the standard deviation of a preset multiple as the lower bound threshold; taking the sum of the mean of the evaluation residuals of each historical data and the standard deviation of a preset multiple as the upper bound threshold; taking the upper bound threshold and the lower bound threshold as the monitoring threshold range of the sensor.

[0064] Optionally, determining the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data includes:

[0065] Take the difference between the mean of the evaluation residuals of the historical data and the standard deviation multiplied by a preset multiple as the lower bound threshold.

[0066] In this embodiment, the preset multiple can be a preset value, and the set value needs to satisfy the following formula:

[0067]

[0068] In this embodiment, X is the evaluation residual, μ is the mean (mathematical expectation) of the evaluation residuals, σ is the standard deviation of the evaluation residuals, and k is a constant greater than 1. For any evaluation residual X with finite variance, at least of the probability that its value deviates from the mean by no more than k standard deviations, that is, at most of the probability that its value deviates from the mean by more than k standard deviations.

[0069] In this embodiment, the lower bound threshold is μ - kσ. μ is the mean of the evaluation residuals, k is the preset multiple, and σ is the standard deviation of the evaluation residuals.

[0070] Take the sum of the mean of the evaluation residuals of the historical data and the standard deviation multiplied by a preset multiple as the upper bound threshold.

[0071] In this embodiment, the upper bound threshold is μ + kσ.

[0072] Take the upper bound threshold and the lower bound threshold as the monitoring threshold range of the sensor.

[0073] In this embodiment, the monitoring threshold range of the sensor is [μ - kσ, μ + kσ].

[0074] In a specific example, a monitoring model is established. Through this monitoring model, the evaluation value of the input historical data can be obtained. By taking the difference between the evaluation value and the historical data, the evaluation residual can be obtained. The evaluation residual is expressed as:

[0075] e = x actual - x evaluate ;

[0076] In this embodiment, x actual is the historical data, and x evaluate is the evaluation value corresponding to x actual .

[0077] Corresponding to each moment, each sensor will generate an evaluation residual e. For the same sensor, when it is operating stably, the obtained evaluation residuals will be normally distributed; if the operation changes, the evaluation residuals will change during the change process and no longer conform to the normal distribution; for different sensors, the obtained evaluation residuals are independently distributed from each other.

[0078] Initialize the monitoring threshold range based on the Chebyshev inequality. The Chebyshev inequality can describe the degree to which a random variable deviates from its mean when the distribution of the random variable X is unknown. The general form of the Chebyshev formula is as follows:

[0079]

[0080] In this embodiment, X is a random variable, μ is the sample mean (mathematical expectation) of the random variable, σ is the standard deviation of the random variable, and k is a constant greater than 1. For any random variable X with finite variance, at least of the probability that its value will deviate from the mean by no more than k standard deviations, that is, at most of the probability that its value will deviate from the mean by more than k standard deviations.

[0081] Here, k is preset, which reflects a limit value set by the engineer for the normal data fluctuation. For example, when the system is running normally, the mean μ of the evaluation residual is approximately 0, and the standard deviation σ = 0.2. If k = 5 is set manually, then we can obtain:

[0082]

[0083] That is

[0084] The probability that the evaluation residual is between [-1, 1] is at least 96%, and the probability in (-∞, -1] and [1, +∞) is at most 4%. In this embodiment, the remaining normal operation data not used for modeling (that is, the first historical data set) is used as the input of the monitoring model to obtain the corresponding evaluation residual. Calculate the mean and standard deviation of the evaluation residual, and set k to 10, then it can be ensured that the probability that the evaluation residual is within 10 standard deviations is at least 0.99.

[0085] After determining the value of K, the monitoring threshold range can be obtained. The upper threshold of the monitoring threshold range is μ + kσ, and the lower threshold of the monitoring threshold range is μ - kσ. The probability that the parameter falls within the monitoring threshold range [μ - kσ, μ + kσ] is at least Then this monitoring threshold range can be called the Chebyshev monitoring threshold range. Using this monitoring threshold range as the initialized monitoring threshold range not only has interpretability but also applies to parameters with unknown distributions. If the parameter to be measured is within the monitoring threshold range, it can be judged that the parameter is normal; if the parameter to be measured is outside the interval, it can be judged that the parameter is abnormal; if the parameter to be measured is outside the monitoring threshold range for a long time, then an alarm needs to be issued to inform the engineer that the equipment is abnormal.

[0086] Optionally, after performing the monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor, it further includes:

[0087] Generate a verification data set based on the other historical data in the initial historical data set except the historical data in the first historical data set.

[0088] In this embodiment, the verification data set includes: the historical data collected by the sensor when the system is operating normally and the historical data collected by the sensor when the system is operating abnormally.

[0089] Verify the monitoring threshold range of each sensor based on the sequential probability ratio test method and the verification data set.

[0090] In this embodiment, the way to verify the monitoring threshold range of each sensor based on the sequential probability ratio test method and the verification data set can be: make assumptions based on the sequential probability ratio test method: H0: The observed value Y follows a normal distribution with a mean of M1 = 0 and a standard deviation of δ = σ 2 ; H1: The observed value Y follows a normal distribution with a mean of M2 = kσ and a standard deviation of δ = σ 2 . Define the likelihood function and calculate the likelihood ratio (the ratio of the likelihood functions) based on the defined likelihood function; define the decision criterion, and the decision criterion includes: if A < L n < B, do not make a judgment, select other observed values and continue; if L n > B, then judge to accept H1; if A > L n , then judge to accept H0. α is the false alarm rate, which is the probability of accepting H1 when H0 is true, and β is the missed alarm rate, which is the probability of accepting H0 when H1 is true.

[0091] In a specific example, based on the sequential probability ratio test (Sequential Probability Ratio Test, SPRT) method and the verification data set, a rationality test is performed on the monitoring threshold range of each sensor. First, assumptions need to be made:

[0092] H0: The observed value Y follows a normal distribution with a mean of M1 = 0 and a standard deviation of δ = σ 2

[0093] H1: The observed value Y follows a normal distribution with a mean of M2 = kσ and a standard deviation of δ = σ 2

[0094] In H0 and H1, the observed value Y is the evaluation residual e, δ is the standard deviation of the evaluation residual, M2 represents the mean of the evaluation residuals obtained by evaluating abnormal data, and M2 = kσ is selected. M1 represents the mean of the evaluation residuals obtained by evaluating normal data. Generally speaking, M1 is close to 0. If it deviates far from 0, it may be that the amount of data is not enough, resulting in a certain contingency. It is necessary to consider reorganizing the normal operation data and abnormal operation data, retraining the monitoring model, and obtaining the evaluation residuals again.

[0095] Continuing the assumption, if H0 or H1 is true, the probabilities (1 - α) and (1 - β) are used to decide whether it is H0 or H1 respectively. α is the false alarm rate, which is the probability of accepting H1 when H0 is true, and β is the miss rate, which is the probability of accepting H0 when H1 is true.

[0096] In the SPRT method, the intermediate process value (i.e., the basis for judgment) of the test result is calculated through the likelihood ratio Ln, which can be obtained by calculating a series of observed values of Y:

[0097]

[0098] Pr(y1,y2,..y n |H1) is the probability of the observed value when H1 is true, and Pr(y1,y2,..y n |H0) is the probability of the observed value when H0 is true. Taking the logarithm of the likelihood ratio on both sides of the equation, the result under the assumption of independent observed values can be expressed as:

[0099]

[0100] As long as the result value is in this situation, no judgment is made, and other observed values are selected to continue:

[0101] A < L n < B;

[0102] Where:

[0103]

[0104] If L n > B, it is judged to accept H1; if A > L n, then judge to accept H0. It is possible to set α = 0.04, β = 0.04, A = 0.04 / 0.96 = 0.041, and B = 0.96 / 0.04 = 24. Use the data of normal operation and abnormal operation to calculate. For example, there are 3000 pieces of normal operation data and 3000 pieces of abnormal operation data. Input these data into the monitoring model to obtain the corresponding evaluation residuals, and then calculate the evaluation residuals. If 99% of the 3000 pieces of normal data accept H0, and 99% of the 3000 pieces of abnormal data accept H1, then it indicates that the threshold setting is reasonable; if it is other situations, then it is considered that the monitoring threshold range setting is unreasonable, and it is necessary to re - consider dividing the data, building the model, generating residuals, as well as setting the monitoring threshold range and finally testing the monitoring threshold range. Until the requirements are met.

[0105] Optionally, it further includes:

[0106] Obtain the third historical data set.

[0107] In this embodiment, the third historical data set includes: at least one piece of post - repair data of sensors in the system, and the post - repair data is the data collected by the sensors when the system is in a normal operation state after a preset time from the completion of the planned repair.

[0108] In this embodiment, the planned repair includes: expected repair, overhaul and other situations.

[0109] In this embodiment, after using the monitoring threshold range of each sensor for a period of time, if a planned repair is encountered, since the planned repair requires replacing parts and equipment, it is necessary to re - adjust the monitoring threshold range.

[0110] In this embodiment, the post - repair data in the third historical data set is the data collected by the sensors when the system is stably operating after the planned repair is completed.

[0111] Perform a monitoring threshold range update operation on the post - repair data of each sensor in the third historical data set to obtain the updated monitoring threshold range of each sensor.

[0112] In this embodiment, the monitoring threshold range update operation includes: inputting each piece of post - repair data into the monitoring model to obtain the evaluation value corresponding to each piece of post - repair data.

[0113] In this embodiment, the establishment of the monitoring model is the same as the above - mentioned way of establishing the monitoring model, and will not be elaborated here.

[0114] Determine the difference between each piece of post - repair data and the evaluation value corresponding to each piece of post - repair data as the evaluation residual of each piece of post - repair data.

[0115] Obtain the mean and standard deviation of the evaluation residuals of each piece of post - repair data.

[0116] Based on the standard deviation of each piece of historical data, and the mean and standard deviation of the evaluation residuals of each piece of post - repair data, determine the monitoring threshold range after the sensor is updated.

[0117] In this embodiment, the method of determining the monitoring threshold range after the sensor is updated based on the standard deviation of each piece of historical data, and the mean and standard deviation of the evaluation residuals of each piece of post - repair data can be: Take the standard deviation of each piece of historical data and the mean of the evaluation residuals of each piece of post - repair data as the target standard deviation; Based on the target standard deviation and the mean of the evaluation residuals of each piece of post - repair data, determine the monitoring threshold range after the sensor is updated.

[0118] Optionally, determining the monitoring threshold range after the sensor is updated based on the standard deviation of each piece of historical data, and the mean and standard deviation of the evaluation residuals of each piece of post - repair data includes:

[0119] Take the standard deviation of each piece of historical data and the mean of the evaluation residuals of each piece of post - repair data as the target standard deviation.

[0120] In this embodiment, the standard deviation of each piece of historical data is the standard deviation of each piece of historical data used when calculating the monitoring threshold range last time.

[0121] Based on the target standard deviation and the mean of the evaluation residuals of each piece of post - repair data, determine the monitoring threshold range after the sensor is updated.

[0122] In this embodiment, the method of determining the monitoring threshold range after the sensor is updated based on the target standard deviation and the mean of the evaluation residuals of each piece of post - repair data can be: Take the difference between the mean of the evaluation residuals of each piece of post - repair data and a preset multiple of the target standard deviation as the updated lower - bound threshold; Take the sum of the mean of the evaluation residuals of each piece of post - repair data and a preset multiple of the target standard deviation as the updated upper - bound threshold; Take the upper - bound threshold and the lower - bound threshold as the monitoring threshold range after the sensor is updated.

[0123] In this embodiment, the determination of the preset multiple is the same as the above - mentioned determination method of the preset multiple, and will not be elaborated here.

[0124] In this embodiment, first collect a small batch of post - repair data during stable operation (post - repair data refers to the data collected by the sensor during the normal operation stage of the system after expected repair and overhaul). For the post - repair data, it is necessary to calculate the mean and standard deviation of the evaluation residuals of the post - repair data, take the average of the standard deviation of the evaluation residuals of the post - repair data and the standard deviation of the evaluation residuals of the historical data as the target standard deviation, and based on the target standard deviation and the mean of the evaluation residuals of each piece of post - repair data, determine the monitoring threshold range after the sensor is updated.

[0125] It should be noted that since there is not enough data for interval testing under the new working conditions, no testing will be carried out. After running online for one month or half a year, sufficient normal operation data and abnormal operation data are collected, and then the updated monitoring threshold range of the sensor is tested and adjusted.

[0126] In a specific example, taking the water pump equipment of a certain factory as an example, the water pump contains more than forty sensors, and the sensor types include: temperature, flow, pressure, current, voltage, liquid level, etc. Monitor this equipment and retrieve the historical data of these sensors in the past year from the database. In the early stage, the historical data is divided into normal operation data and abnormal operation data, and a monitoring model is established. Based on the evaluation values output by the monitoring model and the historical data, the monitoring threshold range of the sensor is determined. Conduct a rationality test on the monitoring threshold range of the sensor, and the monitoring threshold range of the sensor that passes the test can be deployed to assist the monitoring model in monitoring. If the equipment undergoes operations such as repair, major repair, and minor repair, and the old monitoring threshold range needs to be updated, then the monitoring threshold range update operation is performed on all the data after repair to obtain the updated monitoring threshold range of each sensor.

[0127] In the prior art, it is necessary to manually set the monitoring threshold range. When there are a large number of sensors, it will consume a large amount of human and time costs. In this embodiment, the monitoring threshold range is initialized based on the Chebyshev inequality, which improves the efficiency of setting the monitoring threshold range and saves labor costs. The system needs to be maintained and repaired every once in a while, and after each operation, the system needs to reset the monitoring threshold range. This embodiment proposes a method for batch updating the monitoring threshold range, which does not require individual manual adjustment. Moreover, in the prior art, after setting the monitoring threshold range, historical normal data and abnormal data are not used to conduct a rationality test on the monitoring threshold range, and the rationality of the set monitoring threshold range cannot be guaranteed. This embodiment proposes to use the method of hypothesis testing to conduct a rationality test on the monitoring threshold range.

[0128] The technical solution of this embodiment is to obtain a first historical data set, where the first historical data set includes: at least one historical data of the sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operation state; perform a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor; the monitoring threshold range determination operation includes: inputting each historical data into the monitoring model to obtain the evaluation value corresponding to each historical data; according to each historical data and the evaluation value corresponding to each historical data, determine the monitoring threshold range of the sensor, which can automatically generate the monitoring threshold range of the sensor, thereby reducing the human and time costs of setting the monitoring threshold range of the sensor.

[0129] Embodiment Two

[0130] Figure 2 The following is a schematic structural diagram of a monitoring threshold range determination device provided by an embodiment of the present invention. This embodiment is applicable to the situation of determining the monitoring threshold range. The device can be implemented in a software and / or hardware manner and can be integrated into any device that provides the function of determining the monitoring threshold range, such as Figure 2 As shown, the monitoring threshold range determination device specifically includes: a first historical data set acquisition module 210 and a monitoring threshold range determination module 220 for each sensor.

[0131] Among them, the first historical data set acquisition module is used to acquire a first historical data set, where the first historical data set includes: at least one historical data of sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operation state;

[0132] The monitoring threshold range determination module for each sensor is used to perform a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor; the monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; and determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

[0133] The above product can execute the method provided by any embodiment of the present invention and has the corresponding functional modules and beneficial effects of the executed method.

[0134] Embodiment III

[0135] Figure 3 The following shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0136] Such as Figure 3As shown, the electronic device 10 includes at least one processor 11 and a memory communicatively connected to the at least one processor 11, such as read-only memory (ROM) 12, random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, ROM 12, and RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0137] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0138] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the monitoring threshold range determination method.

[0139] In some embodiments, the monitoring threshold range determination method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the monitoring threshold range determination method described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the monitoring threshold range determination method by any other appropriate means (e.g., by means of firmware).

[0140] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0141] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0142] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0144] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0145] The computing system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0146] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0147] An embodiment of the present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the monitoring threshold range determination method according to any embodiment of the present invention.

[0148] In the process of implementing the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or, alternatively, can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0149] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining a monitoring threshold range, characterized in that including: Obtain a first historical data set, where the first historical data set includes at least one historical data of sensors in the system, and the historical data is the data collected by the sensors when the system is in a normal operation state; Perform a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor; The monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

2. The method according to claim 1, characterized in that, Determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data includes: Determine the evaluation residual of each historical data as the difference between each historical data and the corresponding evaluation value; Determine the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data.

3. The method according to claim 2, wherein Determining the monitoring threshold range of the sensor according to the mean and standard deviation of the evaluation residuals of each historical data includes: Take the difference between the mean of the evaluation residuals of each historical data and the standard deviation multiplied by a preset multiple as the lower bound threshold; Take the sum of the mean of the evaluation residuals of each historical data and the standard deviation multiplied by a preset multiple as the upper bound threshold; Take the upper bound threshold and the lower bound threshold as the monitoring threshold range of the sensor.

4. The method according to claim 1, characterized in that Obtaining a first historical data set includes: Obtain an initial historical data set, where the initial historical data set includes at least one historical data collected by sensors in the system and the collection timestamp corresponding to each historical data; Determine the change rate of each historical data according to each historical data and the collection timestamp corresponding to each historical data; Exclude the historical data in the initial historical data set whose change rate is greater than the change rate threshold to obtain a target historical data set; Split the target historical data set to obtain a first historical data set and a second historical data set.

5. The method according to claim 4, characterized in that, After performing the monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor, it further includes: Generate a verification data set based on the other historical data in the initial historical data set except the historical data in the first historical data set; Verify the monitoring threshold range of each sensor based on the sequential probability ratio test method and the verification data set.

6. The method according to claim 1, wherein It further includes: Obtain a third historical data set, where the third historical data set includes at least one post-maintenance data of sensors in the system, and the post-maintenance data is the data collected by the sensors when the system is in a normal operation state after a preset time after the planned maintenance is completed; Perform a monitoring threshold range update operation on the post-maintenance data of each sensor in the third historical data set to obtain the updated monitoring threshold range of each sensor; The monitoring threshold range update operation includes: inputting each post-maintenance data into a monitoring model to obtain an evaluation value corresponding to each post-maintenance data; Determine the evaluation residual of each post-maintenance data as the difference between each post-maintenance data and the evaluation value corresponding to each post-maintenance data; Obtain the mean and standard deviation of the evaluation residuals of each post-maintenance data; Determine the monitoring threshold range after sensor update according to the standard deviation of each historical data and the mean and standard deviation of the evaluation residuals of each post-maintenance data.

7. The method according to claim 6, characterized in that, Determine the monitoring threshold range after sensor update according to the standard deviation of each historical data and the mean and standard deviation of the evaluation residuals of each post-maintenance data, including: Take the standard deviation of each historical data and the mean of the evaluation residuals of each post-maintenance data as the target standard deviation; Determine the monitoring threshold range after sensor update according to the target standard deviation and the mean of the evaluation residuals of each post-maintenance data.

8. A monitoring threshold range determination device, characterized in that Including: A first historical data set acquisition module for acquiring a first historical data set, where the first historical data set includes at least one historical data of a sensor in the system, and the historical data is data collected by the sensor when the system is in a normal operating state; A monitoring threshold range determination module for each sensor, which performs a monitoring threshold range determination operation on the historical data of each sensor to obtain the monitoring threshold range of each sensor; the monitoring threshold range determination operation includes: inputting each historical data into a monitoring model to obtain an evaluation value corresponding to each historical data; determining the monitoring threshold range of the sensor according to each historical data and the evaluation value corresponding to each historical data.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to execute the monitoring threshold range determination method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a processor to execute the monitoring threshold range determination method according to any one of claims 1-7 when executed.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the monitoring threshold range determination method according to any one of claims 1-7.

Images